Probing the substrate specificity of Trypanosoma brucei GlcNAc-PI de-N-acetylase with synthetic substrate analogues

Article abstract of DOI:10.1039/c3ob42164c

A series of synthetic analogues of 1-d-(2-amino-2-deoxy-α-d- glucopyranosyl)-myo-inositol 1-(1,2-di-O-hexadecanoyl-sn-glycerol 3-phosphate), consisting of 7 variants of either the d-myo-inositol, d-GlcpN or the phospholipid components, were prepared and tested as substrates and inhibitors of GlcNAc-PI de-N-acetylase, a genetically validated drug target enzyme responsible for the second step in the glycosylphosphatidylinositol (GPI) biosynthetic pathway of Trypanosoma brucei. The d-myo-inositol in the physiological substrate was successfully replaced by cyclohexanediol and is still a substrate for T. brucei GlcNAc-PI de-N-acetylase. However, this compound became sensitive to the stereochemistry of the glycoside linkage (the β-anomer was neither substrate or inhibitor) and the structure of the lipid moiety (the hexadecyl derivatives were inhibitors). Chemistry was successfully developed to replace the phosphate with a sulphonamide, but the compound was neither a substrate or an inhibitor, confirming the importance of the phosphate for molecular recognition. We also replaced the glucosamine by an acyclic analogue, but this also was inactive, both as a substrate and inhibitor. These findings add significantly to our understanding of substrate and inhibitor binding to the GlcNAc-PI de-N-acetylase enzyme and will have a bearing on the design of future inhibitors.

Full text of DOI:10.1039/c3ob42164c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Probing the substrate specificity of Trypanosoma
brucei GlcNAc-PI de-N-acetylase with synthetic
substrate analogues†
Cite this: Org. Biomol. Chem., 2014,
12, 1919
Amy S. Capes,‡ Arthur Crossman,‡ Michael D. Urbaniak, Sophie H. Gilbert,
Michael A. J. Ferguson* and Ian H. Gilbert*
A series of synthetic analogues of 1-D-(2-amino-2-deoxy-α-D-glucopyranosyl)-myo-inositol 1-(1,2-di-
O-hexadecanoyl-sn-glycerol 3-phosphate), consisting of 7 variants of either the D-myo-inositol, D-GlcpN
or the phospholipid components, were prepared and tested as substrates and inhibitors of GlcNAc-PI de-
N-acetylase, a genetically validated drug target enzyme responsible for the second step in the glycosyl-
phosphatidylinositol (GPI) biosynthetic pathway of Trypanosoma brucei. The D-myo-inositol in the
physiological substrate was successfully replaced by cyclohexanediol and is still a substrate for T. brucei
GlcNAc-PI de-N-acetylase. However, this compound became sensitive to the stereochemistry of the gly-
coside linkage (the β-anomer was neither substrate or inhibitor) and the structure of the lipid moiety (the
hexadecyl derivatives were inhibitors). Chemistry was successfully developed to replace the phosphate
with a sulphonamide, but the compound was neither a substrate or an inhibitor, confirming the impor-
tance of the phosphate for molecular recognition. We also replaced the glucosamine by an acyclic ana-
logue, but this also was inactive, both as a substrate and inhibitor. These findings add significantly to our
understanding of substrate and inhibitor binding to the GlcNAc-PI de-N-acetylase enzyme and will have a
bearing on the design of future inhibitors.
Received 31st October 2013,
Accepted 4th February 2014
DOI: 10.1039/c3ob42164c
www.rsc.org/obc
inositol 1-(1,2-di-O-hexadecanoyl-sn-glycerol 3-phosphate) (1,
α-D-GlcpNAc-PI) to 1-D-(2-amino-2-deoxy-α-D-glucopyranosyl)-
Introduction
The enzymes of the glycosylphosphatidylinositol (GPI) bio- myo-inositol 1-(1,2-di-O-hexadecanoyl-sn-glycerol 3-phosphate)
synthetic pathway are located in the endoplasmic reticulum, (2, α-D-GlcpN-PI), Fig. 1.
contain between one and thirteen predicted trans-membrane
This enzyme catalyses the second step in the T. brucei GPI
domains and are mostly present as components of multi- biosynthetic pathway, which is a prerequisite for all sub-
subunit complexes.1 No high-resolution structural data are sequent steps in the pathway.9 In earlier studies, we showed
available on any of these enzymes and our research group has that T. brucei GlcNAc-PI de-N-acetylase is a zinc-dependent
been probing the specificities of several of the enzymes in the metalloenzyme10 and demonstrated, by construction of a con-
GPI pathway of the protozoan parasite Trypanosoma brucei, the dition-null mutant cell line, that it is essential for the blood-
causative agent of African sleeping sickness in humans and stream form of the parasite and, therefore, a genetically
the related disease Nagana in cattle, using synthetic substrate validated drug target.11 Previous studies with other substrate
analogues in vitro.2–8 One of the key enzymes of interest is an analogues showed that the phosphate, 2′-NHAc and 3′-OH
amidase, the GlcNAc-PI de-N-acetylase (EC 3.5.1.89) that de- groups of the natural substrate α-D-GlcpNAc-PI (1) are critical
N-acetylates 1-D-(2-acetamido-2-deoxy-α-D-glucopyranosyl)-myo- for recognition by the T. brucei GlcNAc-PI de-N-acetylase.2–4 In
contrast, the diacylglycerol moiety is not strictly required and
may be efficiently replaced with an octadecyl chain,4 as shown
in analogues 3 and 4. In the case of the T. brucei enzyme, we
Division of Biological Chemistry and Drug Discovery, College of Life Sciences,
had hypothesised that one or more of the inositol 2, 3, 4 and
5-OH groups is/are not required.2–4 We came to this hypothesis
from the ability of the enzyme to recognise and process both
α-D-GlcpNAc-[L]-PI (5) and β-D-GlcpNAc-PI (6). Molecular
dynamics simulations, showed that α and β anomers can
adopt conformations in which the phosphate, the 2′-amide
University of Dundee, Dow Street, Dundee, DD1 5EH, UK.
E-mail: m.a.j.ferguson@dundee.ac.uk, i.h.gilbert@dundee.ac.uk;
Tel: +44 (0) 1382 386240
†Electronic supplementary information (ESI) available: Additional experimental
procedures and characterisation data for the β-anomers 8 and 10 plus 1H and
13C NMR spectra of all the compounds. See DOI: 10.1039/c3ob42164c
‡These authors contributed equally to the work.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1919

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Some previously prepared GPI analogues.
and the 3′-OH overlay. Given the 2′-amide is where the reaction substrate recognition, the glucosamine residue might be sim-
occurs, and evidence suggests that the 3′-OH and phosphate plified to a simple acyclic structure, as in compound 13.
are important for recognition of the substrate with the
The N-acetylated derivatives of the above analogues
enzyme, these conformations are likely to be the active required for biological studies with the de-N-acetylase were
enzyme-bound conformations. In these conformations the ino- prepared from the corresponding amines by standard pro-
sitol 2-, 3-, 4- and 5-hydroxyls are in different orientations for cedures.13 All the analogues were examined for their recog-
the two α and β anomers, implying the hydroxyls are not criti- nition and processing by the T. brucei GlcNAc-PI de-N-
cal for interaction with the enzyme. Further evidence was acetylase.
obtained from a compound in which the inositol 2-hydroxyl
was alkylated. This was also a substrate for the T. brucei
enzyme. One of the goals of the study described in this paper
was to investigate the hypothesis that the inositol 2, 3, 4 and
5-OH groups are not required.
Results and discussion
Synthesis of analogues 7–13
Bearing in mind these key structural features, we have syn- The synthesis of the required α and β-glucosaminyl (1′ → 1)
thesised a variety of analogues to further probe the require- cyclohexanediol building blocks 16 and 17, respectively, began
by reacting the known14 trichloroacetimidate 14 and the com-
ments for substrate recognition by the T. brucei GlcNAc-PI de-
N-acetylase and, specifically, to test:
1. The hypothesis that the inositol 2, 3, 4 and 5-OH groups with a catalytic amount of trimethylsilyl trifluoromethanesul-
are not required for enzyme recognition, a series of pseudodi-
saccharides (7–10, Fig. 2) containing a cyclohexanediol moiety this stage and these anomers were vital in providing the gluco-
mercially available 1R,2R-trans-cyclohexanediol 15, Scheme 1,
fonate (TMSOTf). The separation of anomers was achievable at
in place of the inositol aglycone were prepared.
samine-phosphodiester target analogues discussed herein. For
the sake of brevity, we have chosen to describe in the main text
2. Whether the phosphate group can be replaced by more
cell-permeable sulphonamide isosteres,12 compounds 11 and the formation of the α-anomers, while details for the corres-
12 (Fig. 2) were prepared.
ponding β-anomers appears in the ESI.† Therefore, the pseu-
dodisaccharide 16 was coupled to the hydrogen phosphonate
3. Whether, given the essentiality of the 2′-NHR and 3′-OH
groups but non-essentiality of the 4′- and 6′-OH groups for 18,15 and the ensuing mixture of diastereoisomeric
Fig. 2 Target molecules.
1920 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Synthesis of 7.
phosphonic diesters, was oxidised with iodine in pyridine– reaction that gives an imine which is then further hydrogen-
water16 to give the corresponding phosphodiester 19. The ated to an aminobutanol.17 Consequently, after purchasing a
diester was transformed into the triol 20 after conventional fresh bottle of anhydrous stabilised THF, the second hydro-
O-deacylation of the latter compound with 0.05 M methanolic genolysis attempt at 20 → 7 proceeded without incident.
NaOMe in CH2Cl2–MeOH.
The preparation of the dipalmitoyl glycerol pseudodisac-
Our first attempt at hydrogenolysis of the triethylammo- charide 9 was accomplished from the triacetate 16, Scheme 2.
nium (TEA) salt 20 over Pd(OH)2/C gave, surprisingly, the alkyl However, the acetate protecting groups in 16 are unsuitable
alcohol secondary amine 21. Initially, the azide of 20 was because if they were left in place and removed by base at the
reduced to the primary amine, however, if there are peroxides final step of the synthesis, then those requisite esters of
present in the THF then the formation of THF hydroperoxide the lipid fragment would likewise be saponified. Therefore,
is, apparently, possible which could then lead to an amine– the acetates of 16 needed to be swapped to a more appropriate
THF coupling via a free-radical-based mechanism.17 This inter- protecting group but first, the temporary tert-butyldimethylsilyl
mediate is susceptible to a Pd-mediated THF ring opening (TBDMS) protection of the 2-OH, 16 → 22, was performed and
Scheme 2 Synthesis of 9.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1921

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Synthesis of sulphonamides 11 and 12.
then followed by conventional O-deacylation, as previously oxide 36 with p-methoxybenzyl alcohol (PMBOH), using
described for 19 → 20, furnished the triol 23.
Cu(BF4)2·nH2O as a catalyst21 to give the known racemic PMB
The benzyl group was chosen as the 3′, 4′ and 6′-OH protect- monoprotected contaminated with
cyclohexanediol22
ing group because, from our past experiences synthesising GPI unreacted PMBOH. Chromatographic separation of this PMB
analogues, the benzyl group has been a very reliable protecting cyclohexyl derivative from the excess of PMBOH was not
group via ease of installation and removal. Thus, the triol 23 achievable, in our hands, and so the entire PMB reaction
was benzylated with benzyl bromide in the presence of NaH, residue was acetylated with acetic anhydride in the presence of
as the base, to afford compound 24. We next turned our atten- pyridine and a catalytic amount of 4-(dimethylamino)pyridine
tion towards the reduction of the azide in 24 and because of DMAP to furnish the acetate 37, which was easily separated
the issues with Pd(OH)2 catalysed reduction in the presence of from the acetate of p-methoxybenzyl alcohol by silica gel
peroxidic THF discussed earlier, we chose to reduce the azide column chromatography.23 After deacetylation, the resulting
via the Staudinger reaction18 to give the amine 25 which was alcohol22 residue was alkylated using sodium hydride and allyl
subsequently tert-butyl carbamate (Boc) protected to furnish bromide to give the allyl derivative 38. The epoxide 39 was pre-
26. Desilylation of 26 using HF–pyridine conditions afforded pared upon reacting 38 with 3-chloroperbenzoic acid (mCPBA),
the alcohol 27. The known hydrogen phosphonate 2819 was and the subsequent hydrolysis of epoxide 39 with DMSO, H2O
coupled, as already described, to the 2-OH of 27 which, after and a catalytic amount of KOH,24 worked smoothly to furnish
oxidation, was isolated and characterised as the TEA salt 29. diol 40. The primary alcohol of 40 was protected using tert-
Hydrogenolysis of 29 over Pd(OH)2/C gave the Boc protected butyl(chloro)diphenylsilane (TBDPSCl) and DMAP to give com-
derivative 30 and subsequent cleavage of the Boc group pro- pound 41. The azido group of 43 was satisfactorily installed
duced the deprotected target analogue 9.
(51% yield over two steps) via the mesylate 42 obtained by
Sulphonamides are potential isosteres for the phosphate reacting the secondary alcohol 41 with methanesulfonyl chlor-
group; they have the same tetrahedral shape and polar oxygen ide in the presence of pyridine, followed by treatment of 42
atoms. The synthesis of the sulphonamides 11 and 12, with sodium azide under forcing conditions. The crude mesy-
Scheme 3, was accomplished by reacting commercially late 42 was used directly in the displacement reaction but a
available 1R,2R-1-amino-2-benzyloxycyclohexane 31 with small portion of 42 was purified for a full characterisation of
1-octadecanesulfonyl chloride in the presence of triethylamine this intermediate. The p-methoxybenzyl protecting group of 43
and CH2Cl2 to give the benzyl derivative 32. The benzyl group was removed with mild acid to give alcohol 44 and then it was
was removed by hydrogenolysis over Pd(OH)2/C to furnish the phosphorylated, as previously described, to give the phospho-
alcohol 33. Coupling of 33 with the known trichloroacet- ric diester 45. Thereafter, the removal of the silyl protecting
imidate 3420 resulted in an inseparable mixture of the α,β-ano- group of 45 with 1.0 M tetrabutylammonium fluoride (TBAF)
mers 35 in the ratio of ∼1 : 1, as determined by 1H NMR in THF proceeded smoothly to give 46 which, after hydrogeno-
spectroscopy. Finally, hydrogenolysis of the aforementioned lysis over Pd(OH)2/C, provided the TEA salt 13.
anomers 35 over Pd(OH)2/C and subsequent silica gel column
chromatography (9 : 1 CH2Cl2–MeOH) gave first the β-anomer
11 and then the α-anomer 12.
We were also interested in seeing if we could replace the
glucose ring with an acyclic moiety. From our knowledge of
Biological results
Substrate analogues
the SAR, retaining the 2-amino and 3-hydroxy groups are
important for activity. The synthesis of the amino-phosphate
13, Scheme 4, began by epoxide ring-opening of cyclohexene
The ability of the T. brucei GlcNAc-PI de-N-acetylase to recog-
nise and process the synthetic pseudodisaccharides 7–13 was
tested in T. brucei cell-free system using an LC-MS/MS
1922 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 4 Synthesis of racemic 13.
assay.8,10 The N-acetylated analogues of these compounds were to improved accessibility of the compound, conferred by the
prepared as previously described to give compounds 47–53 single alkyl chain, to the membranes that contain the de-
(Fig. 3).13
N-acetylase enzyme in the cell-free system. Of the synthetic
Using LC-MS/MS, multiple reaction monitoring of charac- pseudodisaccharides (47–50) tested, only the α-anomer of the
teristic transitions for the N-acetylated and corresponding dipalmitoylated compound 49 showed any appreciable turn-
amine form of each compound was used to directly measure over at 22% the rate of α-D-GlcpNAc-PI (1).
the rate of conversion of the N-acetylated compound to the
Inhibitors
free amine (Table 1). As no suitable transition was identified
for the amine form of 13, the enzymatic turnover was accessed
by reacting any free amine formed with d6-Ac2O, and measur-
ing the formation of the d3-N-acetylated form by LC-MS/MS.
Over the range of enzyme concentration that gave a linear
turnover for α-D-GlcpNAc-PI (1) the substrate analogue
α-D-GlcpNAc-IPC18 (3) was de-N-acetylated at 450% the rate of
α-D-GlcpNAc-PI (1). The increased turnover is most likely due
Since the majority of the compounds were not processed by
the T. brucei GlcNAc-PI de-N-acetylase, we tested their ability to
inhibit the turnover of the α-D-GlcpNAc-IPC18 (3) substrate by
the T. brucei GlcNAc-PI de-N-acetylase in the LC-MS/MS assay.
Most compounds showed no inhibitory activity at 100 μM.
However, compounds 47 and 48 showed significant inhibition,
with IC50 values of 11 4 μM and 37 20 μM, respectively,
Fig. 3 N-Acetylated analogues.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1923

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Recognition of synthetic analogues by T. brucei GlcNAc-PI de-N-acetylase
Compound m/z Transition for NH2 m/z Transition for NHAc Fragment assignment Turnover/pmol/106 cells equiv. Relative turnovera
1
3
1012 > 241
673 > 223
608 > 100
608 > 100
906 > 255
906 > 255
633 > 332
633 > 332
972 > 241
715 > 223
650 > 100
650 > 100
948 > 255
948 > 255
592 > 332
592 > 332
563 > 447
C6H10O8P
C6H8O7P
6.1 0.9
27.0 6.0
ND
ND
1.3 0.3
ND
ND
ND
ND
100%
450%
—
—
22%
—
—
—
—
47
48
49
50
51
52
53b
C6H12
C6H12
O
O
O2CC15H31
O2CC15H31
NHSO2C18H37
NHSO2C18H37
C24H47O5P
a Turnover relative to α-D-GlcpNAc-PI (1). b No suitable MRM, ND – turnover not detected. The multiple reaction monitoring (MRM) transition is
shown as [parent ion m/z] > [daughter ion m/z].
Fig. 4 The hydroxamic acid derivative.
Fig. 5 The trypanosome GPI biosynthesis in the cell-free system. A. The
T. brucei cell-free system was incubated without exogenous substrate,
with 1, α-D-GlcpNAc-PI (10 μM), or with 49 (100 μM) in the presence of
GDP-[3H]Mannose to stimulate the production of radiolabelled manno-
sylated GPI intermediates. B. Inhibition of the turnover of 3,
α-D-GlcpNAc-IPC18 (10 μM), in the presence of 48 (100 μM). Glycolipid
indicating that they can be recognised but not processed by
the T. brucei GlcNAc-PI de-N-acetylase. Interestingly, the
potency of 47 and 48 is comparable to that observed with the,
hydroxamic acid pseudodisaccharide analogue 54 (Fig. 4),
where the N-acetyl group is replaced with the zinc-chelating
hydroxamic acid (IC50 = 19 0.5 μM)8 and suggests that the
zinc-chelating group may not be driving the potency of the
latter compound.
products were extracted, separated by high-performance thin-layer
chromatography, and visualised by fluorography. DPM – dolichol-phos-
phate-mannose, M1 – Man1 species, M2 – Man2 species, M3 – Man3
species, A’ – EtNPMan3 species, where the identity and migration of the
species depends on the glycolipid substrate employed.
supports this hypothesis with an important caveat; when these
hydroxyl groups are removed, substrate recognition and turn-
over is dependent on both the stereochemistry of the glyco-
sidic linkage and the lipid composition. With the
diacylglycerol lipid containing compounds 49 and 50, only the
natural α-anomer 49 is both recognised and processed,
whereas the β-anomer 50 is neither a substrate nor an inhibi-
tor. However, neither the α nor β-anomer of the octadecyl lipid
containing compounds 47 and 48 is processed, although both
appear to be recognised and act as inhibitors. These sets of
diastereoisomers differ only in the identity of their lipid com-
ponent, with the more flexible diacylglycerol moiety allowing
the glycan to be recognised and processed. Thus, it appears
that the requirement for the presence of inositol 2, 3, 4, and
5-OH groups for recognition by the T. brucei GlcNAc-PI de-N-
acetylase is nuanced and may depend on the conformational
flexibility of the substrate analogue.
The ability of 49 and 48 to act as a substrate and inhibitor,
respectively, of the T. brucei GPI pathway was confirmed using
the trypanosome cell-free system with [3H]-mannose labelling
(Fig. 5). Priming the cell-free system with 49 produced three
bands corresponding to the addition of 1–3 mannose residues
(Fig. 5A), and, consistent with this assignment, the bands were
sensitive to jackbean α-mannosidase. As these mannosylated
compounds lack the inositol 2-OH group they cannot undergo
inositol acylation, a prerequisite for the transfer of ethanol-
amine, and thus are not processed past the Man3-species.25
Priming the cell-free system with 3 (α-D-GlcpNAc-IPC18 at
10 μM) was efficiently prevented by incubation with 48 at
100 μM (Fig. 5B), confirming that inhibition of the GlcNAc-PI
de-N-acetylase is sufficient to prevent the formation of down-
stream GPI precursors.
Previous studies (see the introduction for more detail) have
shown that both α-D-GlcpNAc-PI (1) and β-D-GlcpNAc-PI (6) are
recognised and processed by the T. brucei GlcNAc-PI de-N-
acetylase, leading to the hypothesis that one or more of the
inositol 2, 3, 4, and 5-OH groups is/are not required. Our data
The inability of the sulphonamide-containing compounds,
51 and 52, to act as substrates or inhibitors confirms the
importance of the phosphate group in substrate recognition. It
1924 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
may be that the presence of the negatively charged phosphate (Sigma-Aldrich) were dissolved in 1 : 1 Et2O–CH2Cl2 (10 mL).
is essential for binding at the enzyme active site. To this solution was added activated 4 Å molecular sieves (1 g)
The inactivity of compound 53 is difficult to interpret. It and TMSOTf (5.4 μL, 0.03 mmol) at rt under argon. The reac-
may be that removal of the inositol 2, 3, 4, and 5-OH groups is tion mixture was stirred at rt overnight, whereafter it was neu-
not compatible with having a modified glucosamine moiety, or tralised with TEA, percolated through a short column of silica
that the entire glucosamine ring is required. Having said this, gel (further elution with EtOAc) and the subsequent eluent
the glucosamine “replacement” in compound 53 is likely to was concentrated under reduced pressure. RBC [elution first
have a considerable degree of conformational flexibility, which with PE (40–60°) and then with 3 : 2 PE (40–60°)–EtOAc] of the
could allow it to take up multiple orientations within the residue gave first the α-linked pseudodisaccharide 16 (255 mg,
active site.
39%) as a waxy solid; [α]D25 +83.7° (c 2.37, CHCl3); δH (500 MHz,
CDCl3) 5.42 (dd, 1H, J2′,3′ = J3′,4′ = 10.3 Hz, H-3′), 5.09 (d, 1H,
J1′,2′ = 3.7 Hz, H-1′), 4.95 (dd, 1H, J3′,4′ = J4′,5′ = 10.3 Hz, H-4′),
4.18 (dd, 1H, J5′,6′a = 4.8, J6′a,6′b = 12.2 Hz, H-6′a), 4.08 (m, 1H,
H-5′), 4.01 (dd, 1H, J5′,6′b = 2.2, J6′a,6′b = 12.2 Hz, H-6′b), 3.58
(dd, 1H, H-2′), 3.44 (m, 1H, H-2), 3.23 (m, 1H, H-1), 3.05 (s,
1H, 2-OH), 2.50–1.90 (m, 11H, 3 × CH3, H-3a and 6a), 1.64 (m,
2H, H-4a and 5a), 1.35–1.15 (m, 4H, H-3b, 4b, 5b and 6b); δC
(125 MHz, CDCl3) 170.6–169.8 (3 × COCH3), 99.3 (C-1′), 87.5
(C-1), 74.1 (C-2), 71.7 (C-3′), 68.6 (C-4′), 67.8 (C-5′), 62.1 (C-2′),
61.9 (C-6′), 32.0, 31.7, 24.4, 23.8, 20.7 (COCH3) 20.6 (COCH3);
HRMS (ESI) calcd for C18H27N3NaO9 [M + Na]+ 452.1640,
found 452.1622 and then the β-anomer 17 (202 mg, 31%) as a
white solid; mp 120–122 °C; [α]2D5 +21.5° (c 1.15, CHCl3); δH
Conclusions
In summary, we have prepared a series of compounds to probe
the substrate specificity and inhibition of enzymes involved at
an early stage of GPI biosynthesis. The enzyme of interest to
us, GlcNAc-PI de-N-acetylase, proved to be fastidious in its
processing of variants of α-D-GlcpN-PI. We conclude that the
glucosamine and the phospholipid moieties are essential for
binding and that, while the D-myo-inositol residue is the pre-
ferred aglycone for recognition by the enzyme, the dispensing
of it entirely with a cyclohexanediol group is tolerated but
should be used with caution. Further work on this enzyme
should focus on using the emerging structure activity relation-
ship data to develop less synthetically complex, cell permeable
analogues, which will be valuable chemical tools and may
serve as leads for a drug discovery programme.
(500 MHz, CDCl3) 4.91 (m, 2H, H-3′ and 4′), 4.43 (d, 1H, J1′,2′
=
8.1 Hz, H-1′), 4.12 (m, 2H, H6′a and 6′b), 3.67 (m, 1H, H-5′),
3.45 (m, 1H, H-2′), 3.40–3.30 (m, 2H, H-1 and 2), 2.50–1.93 (m,
11H, 3 × CH3 and 2H-cyclitol), 1.70–1.60 (m, 2H, cyclitol) 1.35
(m, 1H, cyclitol), 1.17 (m, 3H, cyclitol); δC (125 MHz, CDCl3)
170.6–169.6 (3 × COCH3), 102.1 (C-1′), 88.2, 73.2, 72.1, 71.8
(C-5′), 68.3, 63.7 (C-2′), 61.8 (C-6′), 32.2, 30.9, 24.2, 23.6, 20.7
(COCH3), 20.6 (COCH3); HRMS (ESI) calcd for C18H28N3O9
[M + H]+ 430.1820, found 430.1831.
Experimental
Synthesis general methods
1H, 13C, 31P NMR spectra were recorded on a Bruker AVANCE
500 MHz spectrometer using tetramethylsilane or the residual
solvent as the internal standard. High resolution electrospray
ionisation mass spectra [HRMS (ESI)] were recorded with a
Bruker microTof spectrometer. Melting points were deter-
mined on a Reichert hot-plate apparatus and are uncorrected.
Optical rotations were measured with a Perkin-Elmer 343
polarimeter. Thin layer chromatography (TLC) was performed
on Kieselgel 60 F254 (Merck) plates with various solvent
systems as developers, followed by detection under UV light or
by charring with sulfuric acid–water–ethanol (15 : 85 : 5).
Column chromatography was performed on Kieselgel 60
(0.040–0.063 mm) (Merck). Radial-band chromatography (RBC)
was performed using a Chromatotron (model 7924 T, TC
Research UK) with silica gel F254 TLC standard grade as the
adsorbent. Iatrobeads (6RS-8060) were purchased from SES
Analysesysteme. All reactions were carried out under argon in
commercially available dry solvents, unless otherwise stated.
Triethylammonium 1R,2R-1-O-(2-azido-3,4,6-tri-O-acetyl-2-deoxy-
α-D-glucopyranosyl)-cyclohexanediol 2-(n-octadecylphosphate) 19
Each of the compounds 16 (117 mg, 0.27 mmol) and 1815
(237 mg, 0.54 mmol) were dried overnight over P2O5 in a
vacuum desiccator, whereafter anhyd pyridine was evaporated
therefrom. They were then dissolved in dry pyridine (10 mL),
pivaloyl chloride (216 μL, 1.76 mmol) was added and the
resulting solution was stirred under argon at rt for 1 h. A
freshly prepared solution of iodine (274 mg, 1.08 mmol) in
9 : 1 pyridine–water was then added and stirring of the reaction
mixture was continued for 45 min. After the addition of
CH2Cl2 (20 mL), the organic solution was washed successively
with 5% aq. NaHSO3 (25 mL), water (25 mL), 1 M TEAB buffer
solution (3 × 15 mL), dried (MgSO4) and concentrated under
reduced pressure. RBC of the residue (elution first with CH2Cl2
and then with 9 : 1 CH2Cl2–MeOH) afforded the TEA phos-
phate derivative 19 (140 mg, 60%); [α]2D5 +68.6° (c 1.07, CHCl3);
δH (500 MHz, CDCl3) 5.40 (dd, 1H, J3′,4′ = 9.2 Hz, H-3′), 5.27 (d,
1H, J1′,2′ = 3.7 Hz, H-1′), 4.94 (t, 1H, J4′,5′ = 9.6 Hz, H-4′),
4.24–4.13 (m, 2H, H6′a and H-1 or 2), 4.07–3.98 (m, 2H, H-5′
and 6′b), 3.84–3.76 (m, 3H, OCH2 and H-1 or 2), 3.17 (dd, 1H,
1R,2R-1-O-(2-Azido-3,4,6-tri-O-acetyl-2-deoxy-
α-D-glucopyranosyl)-cyclohexanediol 16 and the β-anomer 17
After drying overnight over P2O5 in a vacuum desiccator, the J2′,3′ = 10.6 Hz, H-2′), 2.83 (q, 6H, J = 6.8 Hz, 3 × CH2CH3),
glycosyl donor14 14 (731 mg, 1.54 mmol) and the acceptor 15 2.05–1.95 (m, 10H, 3 × COCH3 and 1H-cyclitol), 1.87 (m, 1H,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1925

View Article Online
Paper
Organic & Biomolecular Chemistry
cyclitol), 1.62–1.45 (m, 6H, OCH2CH2 and 4H-cyclitol), (C-1 or C-2), 79.3 (C-1 or C-2), 74.0, 71.8, 71.7, 67.1, 62.3, 61.6
1.33–1.14 (41H, [CH2]15, 3 × CH2CH3 and 2H-cyclitol), 0.81 (t, (C-2′), 54.8, 50.6, 48.7 (NCH2), 34.2, 33.6, 31.9, 30.9–30.6, 27.0,
3H, J = 6.8 Hz, CH2CH3); δC (125 MHz, CDCl3) 169.5–168.7 (3 × 25.3, 25.2, 15.0 (CH2CH3); δP (202 MHz, 1 : 1 CDCl3–MeOH-d4)
COCH3), 96.9 (C-1′), 76.4 (C-1 or C-2), 73.9 (C-1 or C-2), 69.4 0.66 (with heteronuclear decoupling); HRMS (ESI) calcd for
(C-3′), 67.9 (C-4′), 66.7 (C-5′), 65.2 (OCH2), 61.1 (C-6′), 60.0 C34H67NO10P [M − H]− 680.4508, found 680.4554.
(C-2′), 44.4 [N(CH2CH3)3], 30.9, 29.7, 28.7–28.1, 24.8, 21.7,
21.0, 20.3, 19.6, 13.1 (CH2CH3), 7.5 [N(CH2CH3)3]; δP
(202 MHz, CDCl3) −0.05 (with heteronuclear decoupling);
HRMS (ESI) calcd for C36H63N3O12P [M − NEt3 − H]− 760.4155,
found 760.4154.
1R,2R-1-O-(2-Amino-2-deoxy-α-D-glucopyranosyl)-
cyclohexanediol 2-(n-octadecylphosphate) 7
A solution of the TEA salt 20 (52 mg, 0.07 mmol) in 1 : 1 stabil-
ised THF–MeOH (5 mL) containing 10–20% Pd(OH)2 on
carbon (5 mg) was stirred under a hydrogen atmosphere at rt
for 1 h. Work-up as described for the derivative 21 gave, after
column chromatography (5 : 1 CH2Cl2–MeOH), the amino com-
pound 7 (33 mg, 77%); [α]D25 +57.5° (c 3.3, 1 : 1 CHCl3–MeOH);
δH (500 MHz, 1 : 1 CDCl3–MeOH-d4), 5.42 (d, 1H, J1′,2′ = 3.9 Hz,
H-1′), 4.05 (m, 1H, H-1 or 2), 3.90–3.65 (6H, OCH2, H-3′, 5′, 6′a,
b), 3.58 (m, 1H, H-1 or 2) 3.32 (m, 3H, H-4′ and MeOH-d4),
3.05 (dd, 1H, J2′,3′ = 10.5 Hz, H-2′), 2.10 (m, 2H, 2H-cyclitol),
1.74–1.57 (4H, OCH2CH2 and 2H-cyclitol), 1.44–1.21 (34H,
[CH2]15 and 4H-cyclitol), 0.90 (t, 3H, J = 6.8 Hz, CH2CH3); δC
(125 MHz, 1 : 1 CDCl3–MeOH-d4) 97.9 (C-1′), 81.8 (C-1 or C-2),
79.9 (C-1 or C-2), 74.5, 71.8, 71.7, 66.7 (OCH2), 62.4 (C-6′), 55.8
(C-2′), 34.0, 33.4, 33.1, 31.9, 31.8, 30.8, 30.5, 26.9, 25.2, 25.1,
23.8, 14.5 (CH2CH3); δP (202 MHz, 1 : 1 CDCl3–MeOH-d4) 0.54
(with heteronuclear decoupling); HRMS (ESI) calcd for
C30H61NO9P [M + H]+ 610.4078, found 610.4050.
Triethylammonium 1R,2R-1-O-(2-azido-2-deoxy-
α-D-glucopyranosyl)-cyclohexanediol 2-(n-octadecylphosphate) 20
To a solution of compound 19 (78 mg, 0.09 mmol) in 1 : 1
CH2Cl2–MeOH (10 mL) was added 5.4 M NaOMe in MeOH
(0.10 mL). The mixture was kept for 3 h at rt and was then neu-
tralised with Amberlite IR-120 (H+) ion-exchange resin, filtered
and the filtrate concentrated under reduced pressure. Column
chromatography (elution first with 3 : 1 CH2Cl2–MeOH and
then with 2 : 1 → 1 : 1) of the residue furnished the TEA salt 20
(45 mg, 67%) as a waxy solid; [α]2D5 +51.6° (c 4.5, 1 : 1 THF–
MeOH); δH (500 MHz, 1 : 1 CDCl3–MeOH-d4) 5.18 (d, 1H, J1′,2′
=
3.5 Hz, H-1′), 4.25 (m, 1H, H-1 or 2), 3.96–3.65 (7H, OCH2,
H-3′, 5′, 6′a,b and H-1 or 2) 3.40 (t, 1H, J3′,4′ = J4′,5′ = 9.6 Hz,
H-4′), 3.13 (q, 6H, J = 7.3 Hz, 3 × CH2CH3), 3.06 (dd, 1H, J2′,3′
=
10.5 Hz, H-2′), 2.03 (m, 1H, cyclitol), 1.90 (m, 1H, cyclitol),
1.64 (m, 6H, OCH2CH2 and 4H-cyclitol), 1.43–1.22 (41H,
[CH2]15, 3 × CH2CH3 and 2H-cyclitol), 0.88 (t, 3H, J = 6.8 Hz,
CH2CH3); δC (125 MHz, 1 : 1 CDCl3–MeOH-d4) 98.7 (C-1′), 76.3
(C-1 or C-2), 74.2 (C-1 or C-2), 73.8 72.4, 72.3, 66.9 (OCH2), 64.5
(C-2′), 62.7, 47.4 [N(CH2CH3)3], 32.0, 31.0–30.6, 29.8, 29.5,
27.1, 23.9, 22.5, 22.1, 15.0 (CH2CH3), 9.6 [N(CH2CH3)3]; δP
1R,2R-1-O-(2-Azido-3,4,6-tri-O-acetyl-2-deoxy-α-D-glucopyranosyl)-
2-O-(tert-butyldimethylsilyl)-cyclohexanediol 22
(202 MHz, 1 : 1 CDCl3–MeOH-d4) −0.47 (with heteronuclear The alcohol 16 (284 mg, 0.66 mmol) was dried overnight in a
decoupling); HRMS (ESI) calcd for C30H57N3O9P [M − NEt3 − desiccator over P2O5 under high vacuum and then dissolved in
H]− 634.3838, found 634.3850.
anhyd. CH2Cl2 (10 mL). To this solution, at room temperature,
was added 2,6-lutidine (154 μL, 1.32 mmol) and tert-butyldi-
methylsilyl trifluoromethane sulfonate (228 μL, 0.99 mmol).
After 30 min, CH2Cl2 (25 mL) and brine (25 mL) were added
1R,2R-1-O-[2-(4-Hydroxybutyl)amino-2-deoxy-
α-D-glucopyranosyl]-cyclohexanediol 2-(n-octadecylphosphate) 21
A solution of the azido compound 20 (45 mg, 0.06 mmol) in and the organic layer separated. The aqueous layer was re-
1 : 1 THF–MeOH (5 mL) containing 10–20% Pd(OH)2 on extracted with CH2Cl2 (25 mL) and the combined organic
carbon (15 mg) was stirred under a hydrogen atmosphere at rt layers were washed with brine (2 × 25 mL), dried (MgSO4) and
for 30 min before it was percolated through a short column of concentrated under reduced pressure. RBC [elution first with
Chelex 100 on a bed of Celite (further elution with 1 : 1 THF– PE (40–60°) and then with 1 : 1 PE(40–60°)–Et2O] of the residue
MeOH). The eluent was concentrated under reduced pressure gave the azide 22 (284 mg, 79%) as an oil; [α]2D5 +90.2° (c 1.52,
and the ensuing residue was purified by column chromato- CHCl3); δH (500 MHz, CDCl3) 5.40 (t, 1H, J2′,3′ = J3′,4′ = 9.7 Hz,
graphy (elution gradient 6 : 1 → 4 : 1 CH2Cl2–MeOH) to give the H-3′), 5.35 (d, 1H, J1′,2′ = 3.3 Hz, H-1′), 5.20 (t, 1H, J3′,4′ = J4′,5′
=
hydroxybutylamino compound 21 (15 mg, 38%); [α]2D5 +34.0° (c 9.7 Hz, H-4′), 4.26 (dd, 1H, J5′,6′a = 4.7, J6′a,6′b = 12.1 Hz, H-6′a),
1.5, 1 : 1 CHCl3–MeOH); δH (500 MHz, 1 : 1 CDCl3–MeOH-d4) 4.10 (m, 2H, H-5′ and 6′b), 3.75 (m, 1H, H-1 or 2), 3.57 (m, 1H,
5.43 (d, 1H, J1′,2′ = 3.6 Hz, H-1′), 4.05 (m, 1H, H-1 or 2), 3.93 H-1 or 2), 3.17 (dd, 1H, H-2′), 2.10–2.03 (3 × s, 9H, 3 × COCH3),
(dd, 1H, J3′,4′ = 10.4 Hz, H-3′), 3.90–3.50 (8H, OCH2 butyl, 1.89 (m, 2H-cyclitol), 1.70–1.25 (6H-cyclitol), 0.89 (s, 9H, 3 ×
POCH2, H-5′, 6′a,b and H-1 or 2), 3.38 (m, 1H, H-4′), 3.17 (t, CH3), 0.12–0.08 (2 × s, 6H, 2 × CH3); δC (125 MHz, CDCl3)
2H, J = 7.6 Hz, NCH2), 3.00 (dd, 1H, J2′,3′ = 10.4 Hz, H-2′), 2.18 170.6–169.7 (3 × COCH3), 97.9 (C-1′), 79.7 (C-1 or 2), 72.6 (C-1
(m, 1H, cyclitol), 2.08 (m, 1H, cyclitol), 1.86 (m, 2H, CH2 or 2), 70.4 (C-3′), 68.7 (C-4′), 67.6 (C-5′), 62.1 (C-6′), 61.0 (C-2′),
butyl), 1.76–1.57 (6H, POCH2CH2, CH2 butyl and 2H-cyclitol), 32.5, 29.8, 25.9, 22.1, 20.7 (COCH3), 20.6 (COCH3), 18.0, −4.1,
1.45–1.20 (34H, [CH2]15 and 4H-cyclitol), 0.89 (t, 3H, J = 6.8 Hz, −4.9; HRMS (ESI) calcd for C24H42N3O9Si [M + H]+ 544.2685,
CH2CH3); δC (125 MHz, 1 : 1 CDCl3–MeOH-d4) 97.3 (C-1′), 83.2 found 544.2698.
1926 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
1R,2R-1-O-(2-Azido-2-deoxy-α-D-glucopyranosyl)-2-O-(tert-
butyldimethylsilyl)-cyclohexanediol 23
starting material. The reaction was then cooled to rt, poured
into water (25 mL) and extracted with CH2Cl2 (3 × 25 mL). The
combined organics were washed successively with water
(25 mL), brine (25 mL), dried (MgSO4) and concentrated under
reduced pressure. RBC [elution gradient hexane → 7 : 1 → 3 : 1
→ 1 : 1 → 1 : 3 hexane–EtOAc] of the residue afforded the
amino derivative 25 (53 mg, 62%); [α]2D5 +65.6° (c 1.09, CHCl3);
δH (500 MHz, CDCl3) 7.35–7.09 (15H, 3 × Ph), 4.98–4.44 (7H,
To a solution of the triacetate 22 (185 mg, 0.34 mmol) in 1 : 1
CH2Cl2–MeOH (92 mL) was added 5.4 M NaOMe in MeOH
(230 μL). The mixture was kept for 30 min at rt and was then
neutralised with Amberlite IR-120 (H+) ion-exchange resin, fil-
tered and the filtrate concentrated under reduced pressure.
The residue, so obtained, was percolated through a short
silica-gel column (further elution with EtOAc) and the eluent
was concentrated under reduced pressure to afford the triol 23
(136 mg, 96%) as a white solid, mp 122–123 °C (from
10 : 1 hexane–Et2O); [α]D25 +84.1° (c 1.63, CHCl3); δH (500 MHz,
H-1′ and 3 × CH2Ar), 3.87 (m, 1H, H-5′), 3.75 (dd, 1H, J5′,6′a
=
3.6, J6′a,6′b = 10.6 Hz, H-6′a), 3.61 (m, 3H, H-4′, 6′b and 1 or 2),
3.55 (t, 1H, J3′,4′ = 9.2 Hz, H-3′), 3.45 (m, 1H, H-1 or 2), 2.77 (dd,
1H, J1′,2′ = 3.7, J2′,3′ = 9.7 Hz, H-2′), 1.98 (m, 1H, cyclitol), 1.70
(m, 1H, cyclitol), 1.63–1.16 (6H, cyclitol), 0.84 (s, 9H, 3 × CH3),
0.00 (2 × s, 6H, 2 × CH3); δC (125 MHz, CDCl3) 138.7–138.0 (Ph),
128.5–127.7 (Ph), 100.2 (C-1′), 84.1 (C-3′), 79.9 (C-1 or 2), 78.9,
75.6, 74.9, 73.5, 71.4 (C-5′), 71.0, 68.7 (C-6′), 56.4 (C-2′), 31.9,
29.4, 25.9, 22.4, 21.8, 18.0, −4.3, −4.4; HRMS (ESI) calcd for
C39H56NO6Si [M + H]+ 662.3871, found 662.3874.
CDCl3) 5.22 (d, 1H, J1′,2′ = 3.4 Hz, H-1′), 4.02 (t, 1H, J3′,4′
=
9.4 Hz, H-3′), 3.90 (dd, 1H, J5′,6′a = 2.3, J6′a,6′b = 11.6 Hz, H-6′a),
3.81 (dd, 1H, J5′,6′b = 2.2, J6′a,6′b = 11.6 Hz, H-6′b), 3.73 (m, 2H,
H-5′ and H-1 or 2), 3.65 (t, 1H, J4′,5′ = 9.4 Hz, H-4′), 3.59 (m, 1H,
H-1 or 2), 3.15 (dd, 1H, J2′,3′ = 10.4 Hz H-2′), 1.85 (m, 2H, cycli-
tol), 1.60 (m, 2H, cyclitol) 1.50–1.20 (4H, cyclitol), 0.89 (s, 9H,
3 × CH3), 0.12–0.08 (2 × s, 6H, 2 × CH3); δC (125 MHz, CDCl3)
98.2 (C-1′), 78.9 (C-1 or 2), 71.5, 71.4, 70.5 (C-4′), 62.9 (C-2′),
61.6 (C-6′), 31.9, 29.4, 25.9, 22.5, 21.7, 18.0, −4.3, −4.9; HRMS
(ESI) calcd for C18H36N3O6Si [M + H]+ 418.2368, found
418.2365.
1R,2R-1-O-[2-N-(tert-Butoxycarbonyl)amino-3,4,6-tri-O-benzyl-
2-deoxy-α-D-glucopyranosyl]-2-O-(tert-butyldimethylsilyl)-
cyclohexanediol 26
The amine 25 (147 mg, 0.22 mmol) was dissolved in EtOAc
(10 mL) at rt. Di-tert-butyldicarbonate (58 mg, 0.26 mmol) was
then added and the mixture was stirred overnight at rt. After-
wards, the reaction mixture was diluted with EtOAc (25 mL)
and then washed successively with water (25 mL), brine
(25 mL), dried (Na2SO4) and concentrated under reduced
pressure. RBC [elution gradient hexane → 7 : 1 → 5 : 1 hexane–
EtOAc] of the residue afforded the Boc protected derivative 26
(132 mg, 79%); [α]2D5 +39.0° (c 1.06, CHCl3); δH (500 MHz,
CDCl3) 7.30–7.05 (15H, 3 × Ph), 4.95 (d, 1H, J1′,2′ = 3.3 Hz,
H-1′), 4.76–4.38 (7H, NH and 3 × CH2Ar), 3.90 (m, 1H, H-2′)
3.83 (m, 1H, H-5′), 3.69 (dd, 1H, J5′,6′a = 4.1, J6′a,6′b = 10.7 Hz,
H-6′a), 3.61 (m, 3H, H-3′, 4′, and 6′b), 3.48 (m, 1H, H-1 or 2),
3.37 (m, 1H, H-1 or 2), 2.01 (m, 1H, cyclitol), 1.74 (m, 1H, cycli-
tol), 1.52 (m, 2H, cyclitol), 1.35 (s, 9H, 3 × BocCH3), 1.30–1.10
(4H, cyclitol), 0.83 (s, 9H, 3 × CH3), 0.00 (2 × s, 6H, 2 × CH3); δC
(125 MHz, CDCl3) 155.3 (CvO), 138.6–138.2 (Ph), 128.4–127.5
(Ph), 99.1 (C-1′), 81.5, 81.3 (C-1 or 2), 79.5, 78.5, 75.3, 75.1,
73.4, 73.2 (C-1 or 2), 71.4 (C-5′), 68.8 (C-6′), 54.6 (C-2′), 33.4,
30.8, 28.5, 26.1, 23.3, 22.9, 18.1, −3.9, −4.3; HRMS (ESI) calcd
for C44H64NO8Si [M + H]+ 762.4396, found 762.4393.
1R,2R-1-O-(2-Azido-3,4,6-tri-O-benzyl-2-deoxy-α-D-glucopyranosyl)-
2-O-(tert-butyldimethylsilyl)-cyclohexanediol 24
To a stirred and cooled (0 °C) solution of the triol 23 (70 mg,
0.17 mmol) in DMF (10 mL) under argon was added NaH
(19 mg, 0.78 mmol) and the solution was stirred for 15 min
before benzyl bromide (93 μL, 0.78 mmol) was added drop-
wise. The reaction mixture was stirred at rt overnight and then
poured slowly and carefully into ice-cold water (50 mL). After
dilution with EtOAc (50 mL), the EtOAc solution was washed
with brine (25 mL), dried (Na2SO4) and concentrated under
reduced pressure. RBC [elution gradient PE (40–60°) → 10 : 1
→ 7 : 1 → 4 : 1 PE (40–60°)–Et2O] of the residue yielded the
fully protected compound 24 (86 mg, 74%); [α]2D5 +57.9° (c 1.78,
CHCl3); δH (500 MHz, CDCl3) 7.33–7.03 (15H, 3 × Ph), 5.17 (d,
1H, J1′,2′ = 3.6 Hz, H-1′), 4.84–4.37 (6H, 3 × CH2Ar), 3.93 (t, 1H,
J3′,4′ = 9.0 Hz, H-3′), 3.83 (m, 1H, H-5′), 3.72–3.60 (m, 3H, H-4′,
6′a and 1 or 2), 3.57 (dd, 1H, J5′,6′b = 2.0, J6′a,6′b = 10.7 Hz, H-6′
b), 3.48 (m, 1H, H-1 or 2), 3.26 (dd, 1H, J2′,3′ = 10.3 Hz, H-2′),
1.80 (m, 2H, cyclitol), 1.32 (m, 2H, cyclitol), 1.40–1.10 (4H,
cyclitol), 0.82 (s, 9H, 3 × CH3), 0.02–0.00 (2 × s, 6H, 2 × CH3); 1R,2R-1-O-[2-N-(tert-Butoxycarbonyl)amino-3,4,6-tri-O-benzyl-
δC (125 MHz, CDCl3) 137.0–136.8 (Ph), 127.4–126.7 (Ph), 97.0 2-deoxy-α-D-glucopyranosyl]-cyclohexanediol 27
(C-1′), 79.2 (C-3′), 77.9 (C-1 or 2), 77.4, 74.3, 74.1, 72.5, 71.3,
To a stirred solution of the silyl derivative 26 (114 mg,
69.8 (C-5′), 67.3 (C-6′), 62.6 (C-2′), 31.3, 28.7, 24.5, 21.7, 21.0,
0.15 mmol) in THF (10 mL) at 0 °C was added ∼70% HF-pyri-
18.4, −5.2, −5.8; HRMS (ESI) calcd for C39H54N3O6Si [M + H]+
688.3776, found 688.3780.
dine (90 μL). The solution was stirred overnight at rt whereafter
a further aliquot of ∼70% HF-pyridine (90 μL) was added and
the solution was left to stir overnight; this process was contin-
ued on day 3. On day 4, TLC revealed the complete disappear-
ance of the starting material, whereafter satd NaHCO3 (1 mL)
1R,2R-1-O-(2-Amino-3,4,6-tri-O-benzyl-2-deoxy-α-D-glucopyranosyl)-
2-O-(tert-butyldimethylsilyl)-cyclohexanediol 25
To a stirred solution of 24 (86 mg, 0.12 mmol) in 10 : 1 THF– was added dropwise to quench the reaction and the resulting
water (5 mL) at 60 °C was added Ph3P (98 mg, 0.38 mmol). solution was poured into brine (25 mL) and extracted with
After 3 h, TLC showed the complete disappearance of the EtOAc (3 × 25 mL). The EtOAc extracts were combined and
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1927

View Article Online
Paper
Organic & Biomolecular Chemistry
washed with brine (25 mL), dried (MgSO4) and concentrated hydrogen for 3 h before it was percolated through a short
under reduced pressure. RBC [elution gradient hexane → column of Chelex 100 on a bed of Celite (further elution with
2 : 1 hexane–EtOAc] of the residue furnished the alcohol 27 1 : 1 THF–n-propanol). The eluent was concentrated under
(97 mg, 49%); [α]2D5 +27.2° (c 1.08, CHCl3); δH (500 MHz, reduced pressure and the ensuing residue was purified by
CDCl3) 7.36–7.10 (15H, 3 × Ph), 5.12 (d, 1H, J1′,2′ = 3.6 Hz, column chromatography (elution gradient 7 : 1 → 4 : 1 CH2Cl2–
H-1′), 4.86–4.44 (7H, NH and 3 × CH2Ar), 3.95 (dd, 1H, J3′,4′
=
MeOH) to give the Boc protected derivative 30 (30 mg, 73%);
J4′,5′ = 9.7 Hz, H-4′), 3.86 (m, 1H, H-2′), 3.75–3.61 (4H, H-3′, 5′ [α]2D5 +39.1° (c 3.00, 1 : 1 CH2Cl2–MeOH); δH (500 MHz, 1 : 1
and 6′a,b), 3.47 (m, 1H, H-1 or 2), 3.32 (m, 1H, H-1 or 2), CDCl3–MeOH-d4) 5.25 (m, 1H, H-2 glycerol), 4.90 (s, 1H, H-1′),
2.10–1.91 (2H, cyclitol), 1.65 (m, 2H, cyclitol), 1.43 (s, 9H, 4.45 (m, 1H, 1- or 3-CHa glycerol), 4.20 (dd, 1H, J = 6.7, J = 11.7
3 × CH3), 1.33–1.16 (4H, cyclitol); δC (125 MHz, CDCl3) Hz, 1- or 3-CHb glycerol), 4.00 (m, 2H, 1- or 3-CH2 glycerol), 3.75
155.6 (CvO), 138.4–138.0 (Ph), 128.4–127.7 (Ph), 99.5 (m, 1H, H-3′ and 4′), 3.65 (m, 1H, H-1 or 2), 3.55 (dd, 1H, J1′,2′
=
(C-1′), 85.0 (C-1 or 2), 80.5, 79.7, 78.5, 75.0, 74.0 (C-1 or 2), 3.0, J2′,3′ = 10.5 Hz, H-2′), 3.47 (m, 1H, H-1 or 2), 3.40 (m, 3H,
73.4, 71.2 (C-4′), 68.7 (C-6′), 54.9 (C-2′), 32.9, 31.7, 28.4, 24.3, H-5′ and 6′a,b), 2.40–2.00 (m, 6H, 2 × COCH2 and 2H cyclitol),
23.9; HRMS (ESI) calcd for C38H50NO8 [M + H]+ 648.3531, 1.74–1.57 (m, 6H, 2 × COCH2CH2 and 2H cyclitol), 1.47 (s, 9H,
found 648.3527.
3 × CH3), 1.40–1.20 (52H, 2 × [CH2]12 and 4H cyclitol), 0.89 (t, 6H,
J = 7.1 Hz, 2 × CH2CH3); δC (125 MHz, 1 : 1 CDCl3–MeOH-d4) 174.6
(CvO), 174.1 (CvO), 158.2 (CvO), 98.9 (C-1′), 82.0, 78.0, 73.3,
73.0, 72.0, 71.1, 64.0, 63.1, 62.0, 57.1, 56.4, 34.7, 34.5, 33.0–29.5,
28.5, 25.4, 24.3, 23.5, 23.1, 14.3; δP (202 MHz, 1 : 1 CDCl3–MeOH-
d4) −0.28 (with heteronuclear decoupling); HRMS (ESI) calcd for
Triethylammonium 1R,2R-1-O-[2-N-(tert-butoxycarbonyl)-
amino-3,4,6-tri-O-benzyl-2-deoxy-α-D-glucopyranosyl]-
cyclohexanediol 2-(1,2-di-O-hexadecanoyl-sn-glycerol
3-phosphate) 29
This compound was obtained from the alcohol 27 (55.7 mg, C52H97NO15P [M − H]− 1006.6601, found 1006.6635.
0.086 mmol) and 1,2-di-O-hexadecanoyl-sn-glycerol 3-hydrogen-
1R,2R-1-O-(2-Amino-2-deoxy-α-D-glucopyranosyl)-
cyclohexanediol 2-(1,2-di-O-hexadecanoyl-sn-glycerol
3-phosphate) 9
phosphonate TEA salt 2819 (126 mg, 0.17 mmol) in the pres-
ence of pivaloyl chloride (69 μL, 0.56 mmol) essentially as
described for the 2-(n-octadecyl phosphate) 19. After the oxi-
dation with iodine (87 mg, 0.34 mmol) in 9 : 1 pyridine–water To a solution of the tert-butoxycarbonyl protected compound
followed by the same aqueous workup as described for 19, 30 (30 mg, 0.030 mmol) in 1 : 1 CH2Cl2–MeOH (1 mL) was
RBC (elution first with CH2Cl2 and then with 20 : 1 → 15 : 1 added 9 : 1 trifluoroacetic acid (TFA)–water (5 mL). After stir-
CH2Cl2–MeOH) afforded the TEA phosphate derivative 29 ring 3 h at rt, toluene (5 mL) was added and the solvents were
(57 mg, 52%) as an opaque oil; [α]2D5 +27.7° (c 1.08, CHCl3); δH removed under reduced pressure. Toluene (2 × 5 mL) was evap-
(500 MHz, CDCl3) 12.5 (brs, 1H, NH TEA salt), 7.36–7.10 (15H, orated off twice from the residue (to remove traces of TFA and
3 × Ph), 5.24 (m, 1H, H-2 glycerol), 4.96 (s, 1H, H-1′), 4.84–4.44 water) to give the pseudodisaccharide phosphate derivative 9
(6H, 3 × CH2Ar), 4.39 (m, 1H, 1- or 3-CHa glycerol), 4.17 (dd, (25 mg, 93%) which did not require any further purification;
1H, J = 6.6, J = 12.0 Hz, 1- or 3-CHb glycerol), 4.13–3.97 (m, 4H, [α]2D5 +28.0° (c 2.50, 1 : 1 CHCl3–MeOH); δH (500 MHz, 1 : 1
H-2′, 1 or 2 cyclitol and 1- or 3-CH2 glycerol), 3.95 (m, 1H, CDCl3–MeOH-d4) 5.28 (d, 1H, J1′,2′ = 3.7 Hz, H-1′), 5.14 (m, 1H,
H-4′), 3.85 (t, J2′,3′ = J3′,4′ = 9.9 Hz, H-3′), 3.74 (dd, 1H, J5′,6′a
=
H-2 glycerol), 4.34 (dd, 1H, J = 3.2, J = 12.0 Hz, 1- or 3-CHa gly-
4.1, J6′a,6′b = 10.7 Hz, H-6′a), 3.66 (m, 2H, H-5′ and 6′b), 3.50 cerol), 4.10 (dd, 1H, J = 6.6, J = 12.0 Hz, 1- or 3-CHb glycerol),
(m, 1H, H-1 or 2), 2.97 (q, 6H, J = 7.1 Hz, 3 × CH2CH3), 2.27 3.96 (m, 1H, H-1 or 2), 3.88 (m, 2H, 1- or 3-CH2 glycerol), 3.71
(m, 4H, 2 × COCH2), 2.21–1.95 (2H, cyclitol), 1.72–1.50 (m, 6H, (m, 3H, H-3′ and 6′a,b), 3.62 (dd, 1H, J4′,5′ = 9.4, J5′,6′ = 3.9 Hz,
2 × COCH2CH2 and 2H cyclitol), 1.44 (s, 9H, 3 × CH3), H-5′), 3.45 (m, 1H, H-1 or 2), 3.33 (t, 1H, J3′,4′ = 9.4 Hz, H-4′),
1.33–1.18 (61H, 3 × CH2CH3, 2 × [CH2]12 and 4H cyclitol), 0.88 2.96 (dd, 1H, J2′,3′ = 10.5 Hz, H-2′), 2.24 (m, 4H, 2 × COCH2),
(t, 6H, J = 7.3 Hz, 2 × CH2CH3); δC (125 MHz, CDCl3) 172.4 2.0 (m, 2H, cyclitol), 1.68–1.47 (m, 6H, 2 × COCH2CH2 and 2H
(CvO), 171.9 (CvO), 155.4 (CvO), 138.9–137.4 (Ph), cyclitol), 1.35–1.14 (52H, 2 × [CH2]12 and 4H cyclitol), 0.79 (t,
127.3–126.3 (Ph), 98.9 (C-1′), 80.0, 76.8, 73.7, 72.7, 72.3, 70.3, 6H, J = 7.1 Hz, 2 × CH2CH3); δC (125 MHz, 1 : 1 CDCl3–MeOH-
69.8, 69.5, 68.1, 67.6, 62.4, 61.8, 61.4, 53.5, 44.2 [N(CH2CH3)3], d4) 175.3 (CvO), 174.8 (CvO), 98.0 (C-1′), 82.5 (C-1 or 2), 80.2
33.3, 33.1, 30.9, 30.2, 28.7–28.1, 27.5, 23.4, 21.7, 13.1, 7.40 (C-1 or 2), 74.2 (C-5′), 71.6, 71.5, 71.4, 64.8 (CH2 glycerol), 63.8
[N(CH2CH3)3]; δP (202 MHz, CDCl3) 0.04 (with heteronuclear (CH2 glycerol), 62.3 (C-6′), 55.7 (C-2′), 35.5, 35.4, 34.0, 33.5,
decoupling); HRMS (ESI) calcd for C73H115NO15P [M − NEt3 − 33.2, 31.0–30.4, 26.2, 25.3, 23.9, 15.1; δP (202 MHz, 1 : 1 CDCl3–
H]− 1276.8010, found 1276.8015.
MeOH-d4) 0.10 (with heteronuclear decoupling); HRMS (ESI)
calcd for C52H97NO15P [M − H]− 906.6077, found 906.6094.
1R,2R-1-O-[2-N-(tert-Butoxycarbonyl)amino-2-deoxy-α-D-
glucopyranosyl]-cyclohexanediol 2-(1,2-di-O-hexadecanoyl-sn-
glycerol 3-phosphate) 30
N-[(1R,2R)-2-(Benzyloxy)cyclohexyl]octadecane-
1-sulphonamide 32
A
solution of the benzylated compound 29 (57 mg, To a solution of CH2Cl2 (10 mL) and triethylamine (2.1 mL)
0.041 mmol) in 1 : 1 THF–n-propanol (10 mL) containing under argon was added (1R, 2R)-1-amino-2-benzyloxycyclo-
10–20% Pd(OH)2 on carbon (15 mg) was stirred under 3 atm of hexane 31 (1.0 g, 4.87 mmol) and 1-octadecanesulfonyl
1928 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
chloride (2.1 g, 5.95 mmol), purchased from Sigma-Aldrich 2H), 3.77–3.57 (m, 7H), 3.47 (m, 1H), 3.42–3.27 (m, 4H), 3.17
and Alfa Aesar, respectively. The reaction mixture was stirred at (m, 1H, H1 α/β or 2 α/β), 3.10–2.96 (m, 4H, 2 × SO2CH2), 2.37
rt overnight, whereafter it was diluted with CH2Cl2 (40 mL), (m, 2H, cyclitol), 2.13 (m, 2H, cyclitol), 1.90–1.18 (α and β
washed successively with water (25 mL), brine (25 mL), dried SO2CH2CH2, [CH2]15 and H’s cyclitol), 0.88 (t, 6H, J = 7.1 Hz, 2
(MgSO4) and concentrated under reduced pressure. RBC × CH2CH3 α and β); δC (125 MHz, CDCl3) 137.9, 137.8, 137.6,
(elution first with hexane and then with a gradient of 5 : 1 → 128.5–127.7 (Ph), 101.2 (C1′β), 99.6 (C1′α), 83.7, 82.7, 82.3,
3 : 1 hexane–EtOAc) gave the sulphonamide 32 (1.9 g, 76%) as 81.4, 78.1, 77.5, 75.8, 75.6, 75.3, 75.1, 74.9, 73.6, 73.5, 71.4,
white needles; mp 63–64 °C; [α]2D5 −35.6° (c 1.00, CHCl3); δH 68.6, 68.2, 66.2, 64.7, 57.9 (C1 α/β or C2 α/β), 57.5, 53.6, 51.9,
(500 MHz, CDCl3) 7.40–7.20 (m, 5H, Ph), 4.68 (d, 1H, J = 1.4 34.0, 32.3, 32.0, 31.9, 31.4, 29.7, 29.5, 29.4, 29.3, 28.6, 28.5,
Hz, OCH2), 4.46 (d, 1H, J = 5.0 Hz, NH), 3.23–3.11 (m, 2H, H1 24.2, 23.8, 23.5, 23.4, 22.7, 14.2 (CH2CH3 α and β); HRMS (ESI)
and 2), 3.03–2.90 (m, 2H, SO2CH2), 2.23 (m, 2H, H3a and 6a), calcd for C51H76N4O7SNa [M + Na]+ 911.5372, found 911.5282.
1.80–1.60 (m, 4H, SO2CH2CH2, H4a and 5a), 1.35–1.11 (m,
N-(1R,2R)-2-O-(2-Amino-2-deoxy-β-D-glucopyranosyl)-
cyclohexyloctadecane-1-sulphonamide 11 and the α-anomer 12
34H, 15 × [CH2]15, H3b, 4b, 5b and 6b), 0.88 (t, 3H, J = 6.8 Hz,
CH2CH3); δC (125 MHz, CDCl3) 138.1, 128.5, 127.8, 127.7, 80.1
(C1 or 2), 70.6 (OCH2), 57.6 (C1 or 2), 53.2 (SO2CH2), 33.2 (C3 A solution of the anomeric mixture 35 (108 mg, 0.12 mmol) in
or 6), 31.9, 30.1 (C3 or 6), 29.7, 29.5, 29.4, 29.1, 28.3, 24.2, 23.8, 5 : 1 THF–AcOH (6 mL) containing 10–20% Pd(OH)2 on carbon
23.5, 22.7, 14.1 (CH2CH3); HRMS (ESI) calcd for C31H56NO3S (25 mg) was stirred under a slight over pressure of hydrogen at
[M + H]+ 522.3975, found 522.3981.
room temperature for 24 h before it was filtered through a bed
of Celite. The catalyst was further washed with 1 : 1 THF–
MeOH (2 × 10 mL) and the washings were combined and con-
N-[(1R,2R)-2-Hydroxycyclohexyl]octadecane-1-sulphonamide 33
A
solution of the benzyloxysulphonamide 32 (200 mg, centrated under reduced pressure. Column chromatography
0.38 mmol) in 5 : 1 THF–AcOH (6 mL) containing 10–20% Pd- (9 : 1 CH2Cl2–MeOH) gave first the β anomer 11 (9.8 mg, 14%)
(OH)2 on carbon (50 mg) was stirred under a slight over as a waxy solid; [α]2D5 +2.5 (c 0.98, MeOH); δH (500 MHz, MeOH-
pressure of hydrogen at room temperature for 2 h before it was d4) 4.44 (d, 1H, J1′,2′ = 8.1 Hz, H1′), 3.94 (dd, 1H, J5′,6′a = 2.1,
percolated through a short column of Celite on a bed of silica J6a′,6′b = 11.7 Hz, H6′a), 3.65–3.50 (m, 2H, H1 or 2 and 6′b),
gel (further elution with EtOAc). The eluent was concentrated 3.34 (m, 2H, H3′ and 5′), 3.22 (q, 1H, J3′,4′ = J4′,5′ = 9.2 Hz, H4′),
under reduced pressure to give the deprotected alcohol 33 3.09 (m, 3H, H1 or 2 and SO2CH2), 2.63 (t, 1H, J2′,3′ = 9.0 Hz,
(140 mg, 85%) as a white solid which was used without any H2′), 2.13 (m, 2H, cyclitol), 1.90–1.20 (38H, SO2CH2CH2,
further purification; mp 101–102 °C; [α]D25 −7.3 (c 3.40, CHCl3); [CH2]15 and 6H cyclitol), 0.89 (t, 3H, J = 7.0 Hz, CH2CH3); δC
δH (500 MHz, CDCl3) 4.54 (d, 1H, J = 7.4 Hz, NH), 3.32 (m, 1H, (125 MHz, MeOH-d4) 100.7 (C1′), 80.5 (C1 or 2), 78.5 (C3′ or
H2), 3.15–3.02 (m, 3H, SO2CH2 and H1), 2.54 (brs, 1H, OH), 5′), 72.2 (C4′), 62.9 (C6′), 58.1 (C1 or 2 or 2′), 58.0 (C1 or 2 or
2.12–2.03 (m, 2H, H3a and 6a), 1.90–1.63 (m, 4H, SO2CH2CH2, 2′), 54.1 (SO2CH2), 35.4, 32.2, 30.8, 30.6, 30.5, 25.4, 25.0, 24.7,
H4a and 5a), 1.50–1.10 (m, 34H, [CH2]15, H3b, 4b, 5b and 6b), 14.5 (CH2CH3); HRMS (ESI) calcd for C30H61N2O7S [M + H]+
0.88 (t, 3H, J = 6.8 Hz, CH2CH3); δC (125 MHz, CDCl3) 73.8 593.4194, found 593.4178. Continued elution gave the α anomer
(C2), 59.8 (C1), 53.5 (SO2CH2), 34.1 (C3 or 6), 31.9, 29.7, 29.6, 12 (14.5 mg, 20%) as an oil; [α]2D5 +66.5 (c 1.45, MeOH); δH
29.5, 29.3, 29.1, 28.3, 24.8 (C4 or 5), 24.0 (C4 or 5), 23.7, 14.1 (500 MHz, MeOH-d4) 5.28 (d, 1H, J1′,2′ = 3.7 Hz, H1′), 3.81 (m,
(CH2CH3); HRMS (ESI) calcd for C24H50NO3S [M + H]+ 1H, H6′a), 3.71 (m, 3H, H3′, 5′ and 6′b), 3.42 (m, 1H, H1 or 2),
432.3506, found 432.3487.
3.34 (m, 1H, H4′), 3.23 (m, H1 or 2), 3.08 (m, 1H, H2′ and
SO2CH2), 2.24 (m, 1H, cyclitol), 1.96 (m, 1H, cyclitol),
1.84–1.67 (m, 4H, SO2CH2CH2 and 2H cyclitol), 1.52–1.21
(34H, [CH2]15 and 4H cyclitol), 0.90 (t, 3H, J = 7.0 Hz,
N-(1R,2R)-2-O-(2-Azido-3,4,6-tri-O-benzyl-2-deoxy-
D-glucopyranosyl)-cyclohexyloctadecane-1-sulphonamide 35
A mixture of trichloroacetimidate 3420 (130 mg, 0.21 mmol), CH2CH3); δC (125 MHz, MeOH-d4) 97.3 (C1′), 81.4 (C1 or 2),
acceptor 33 (109 mg, 0.25 mmol) and activated 4 Å molecular 73.2 (C3′ or 5′), 70.4 (C3′ or 4′ or 5′), 70.3 (C3′ or 4′ or 5′), 60.8
sieves (200 mg) in dry CH2Cl2 (15 mL) was stirred under argon (C6′), 56.6 (C1 or 2), 54.9 (C2′), 52.6 (SO2CH2), 32.7, 32.3, 31.6,
at room temperature for 15 min. Then TMSOTf (5.3 μL, 29.4, 29.3, 29.2, 29.0, 28.8, 27.8, 24.2, 23.5, 23.3, 22.3, 13.0
0.029 mmol) was added and the solution was stirred at room (CH2CH3); HRMS (ESI) calcd for C30H61N2O7S [M + H]+
temperature for an additional 2 h. It was then percolated 593.4194, found 593.4181.
through a short column of silica gel (elution with EtOAc) and
the eluent was concentrated under reduced pressure. RBC
trans-2-(4-Methoxybenzyloxy)cyclohexyl acetate 37
(elution gradient 1 : 5 → 1 : 3 EtOAc–hexane) of the residue Cu(BF4)2·nH2O (42 mg, 0.18 mmol) was dissolved in CH2Cl2
gave the pseudodisaccharide 35 (140 mg, 75%) as an oily (20 mL) and cyclohexene oxide 36 (1.8 mL, 17.8 mmol) and
mixture of α, β anomers in the ratio of ∼1 : 1, as determined by 4-methoxybenzyl alcohol (10 mL, 80.2 mmol) were added. The
1H NMR spectroscopy; δH (500 MHz, CDCl3) 7.40–7.10 (30H, reaction mixture was stirred for 24 h, diluted with water
6 × Ph α and β), 5.62 (d, 1H, J = 2.5 Hz, NH α or β), 5.46 (s, 1H, (20 mL), and the aqueous layer extracted with CH2Cl2 (3 ×
NH α or β), 4.98 (d, 1H, J1′,2′ = 3.6 Hz, H1′α), 4.90–4.44 (m, 12H, 30 mL). The combined organic extracts were washed with
6 × CH2Ph α and β), 4.30 (d, 1H, J1′,2′ = 8.0 Hz, H1′β), 3.91 (m, brine, dried over MgSO4, filtered and the solvent was removed
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1929

View Article Online
Paper
Organic & Biomolecular Chemistry
in vacuo. The resulting crude monoprotected PMB cyclohexane- 71.6 (CH2CHvCH2), 71.0 (CH2Ar), 55.3 (OCH3), 30.33, 30.31,
diol22 was used with no further purification in the next step, 23.59, 23.58; HRMS (ESI) calcd for C17H24NaO3 [M + Na]+
whereby it was dissolved in pyridine (7.5 mL), cooled to 0 °C, 299.1618, found 299.1607.
before DMAP (3 mg, 0.26 mmol) and acetic anhydride (4.5 mL,
48.0 mmol) were added. The reaction mixture was stirred over-
night at rt, diluted with water (100 mL) and extracted with
EtOAc (3 × 100 mL). The combined organic extracts were
washed with water (100 mL), brine (100 mL), dried over
MgSO4, filtered and the solvent was removed under reduced
pressure. Column chromatography (5 : 1 hexane–Et2O) of the
ensuing residue afforded the oily product 37 (1.37 g, 55%); δH
(500 MHz, CDCl3) 7.24 (d, 2H, J = 8.7 Hz, Ph), 6.86 (d, 2H, J =
8.6 Hz, Ph), 4.83–4.79 (m, 1H, H-1), 4.56–4.49 (2 × d, 2H, J =
11.7 Hz, CH2Ar), 3.80 (s, 3H, OCH3), 3.38–3.33 (m, 1H, H-2),
2.04 (s, 3H, CH3), 2.02–1.98 (m, 2H, H-3a and 6a), 1.70–1.62
(m, 2H, H-4a and 5a), 1.43–1.18 (m, 4H, H-3b, 4b, 5b and 6b);
δC (125 MHz, CDCl3) 170.5 (CvO), 159.1, 131.0, 129.0, 114.0,
113.7, 78.4 (C-2), 75.3 (C-1), 71.0 (CH2Ar), 55.3 (OCH3), 30.0,
29.9, 23.6, 23.3, 21.4 (CH3); HRMS (ESI) calcd for C16H22NaO4
[M + Na]+ 301.1410, found 301.1397.
2-{[trans-2-((4-Methoxybenzyl)oxy)cyclohexyloxy]methyl}
oxirane 39
Compound 38 (2.22 g, 8.02 mmol) was dissolved in CH2Cl2
(30 mL) and mCPBA (4.15 g, 24.1 mmol) was added and the
reaction mixture stirred for 18 h at rt. Afterwards, the reaction
mixture was washed successively with 10% aq. sodium sulfite
(80 mL), water (100 mL), 10% aq. NaOH (80 mL) and brine
(80 mL). The organic phase was then filtered through cotton
wool and the solvent was removed in vacuo. The crude material
was passed down a short plug of silica gel (further elution
with EtOAc) and evaporated to dryness under reduced
pressure. RBC (elution gradient 1 : 1 → 2 : 1 Et2O–hexane) furn-
ished the epoxide 39 (1.50 g, 64%) as a clear oil; δH (500 MHz,
CDCl3) 7.20 (dd, 2H, J = 8.7 Hz, Ph), 6.77 (d, 2H, J = 8.7 Hz,
Ph), 4.50 (s, 2H, CH2Ar) 3.76–3.40 (5H, OCH3 and 1- or 3-CH2
propyl), 3.25 (m, 2H, H-1 and 2), 3.05 (m, 1H, H-2 propyl),
2.67–2.52 (2H, 1- or 3-CH2 propyl), 1.90 (m, 2H, cyclitol), 1.57
(m, 2H, cyclitol), 1.22–1.06 (m, 4H, cyclitol); δC (125 MHz,
CDCl3) 159.0, 131.4, 131.3, 129.14, 129.08, 113.7, 82.3 (C-1 or
2), 82.1 (C-1 or 2), 80.8 (C-1 or 2), 71.5 (CH2Ar), 71.2 (1- or
3-CH2 propyl), 70.4 (1- or 3-CH2 propyl), 55.2 (OCH3), 51.3 (C-2
propyl), 51.1 (C-2 propyl), 44.44 (1- or 3-CH2 propyl), 44.38
(1- or 3-CH2 propyl), 30.3, 30.20, 30.16, 23.6, 23.5; HRMS (ESI)
calcd for C17H24NaO4 [M + Na]+ 315.1567, found 315.1553.
1-{[trans-2-(Allyloxy)cyclohexyloxy]methyl}-4-methoxybenzene
38
The acetate 37 (938 mg, 3.37 mmol) was dissolved in MeOH
(10 mL) and NaOMe (5.4 M in MeOH, 150 µL) was added and
the solution stirred for 1 h at rt. Afterwards, TLC revealed that
there was still the presence of 37 and, thus, a further aliquot of
NaOMe (5.4 M in MeOH, 100 µL) was added and the reaction
mixture was stirred for an additional 24 h. After which, the
reaction was neutralised with Amberlite IR-120 (H+) ion-
exchange resin, filtered and the crude solution was passed
down a short plug of silica gel (elution with EtOAc) to afford
3-{[trans-2-((4-Methoxybenzyl)oxy)cyclohexyl]oxy}propane-1,2-
diol 40
the known 2-PMB protected alcohol22 as a pale yellow oil To a solution of 39 (1.50 g, 5.13 mmol) in DMSO (56.4 mL) was
which was used without further purification in the next step. added water (10.8 mL) and aq. 0.3 M KOH (2.4 mL). The reac-
To a stirred and cooled (0 °C) solution of the alcohol22 tion mixture was heated to 100 °C for 18 h, and then diluted
(886 mg, 3.75 mmol) in DMF (40 mL) was added NaH (60% with water (200 mL) followed by extraction with CH2Cl2 (3 ×
dispersion in mineral oil, 750 mg, 18.7 mmol) and the solu- 200 mL). The combined organic extracts were washed with
tion was stirred for 30 min before allyl bromide (2.92 mL, water (100 mL), brine (100 mL), dried over MgSO4, filtered and
22.8 mmol) was added dropwise. The reaction mixture was the solvent was removed under reduced pressure. The residue,
stirred under argon for a further 18 h at rt, quenched with so obtained, was percolated through a short column of silica
MeOH (50 mL), H2O (250 mL) was added and then the result- gel (further elution with EtOAc) and the subsequent eluent
ing solution was extracted with EtOAc (3 × 250 mL). The com- was concentrated under reduced pressure. RBC (elution first
bined organic extracts were washed with H2O (3 × 250 mL), with hexane → 6 : 1 EtOAc–hexane) of the residue afforded the
brine (250 mL), dried over MgSO4, filtered and the solvent was diol 40 (942 mg, 65) as a clear oil; δH (500 MHz, CDCl3) 7.28 (d,
removed under reduced pressure. The resulting residue was 2H, J = 8.6 Hz, Ph), 6.88 (dd, 2H, CH, J = 8.6, Ph), 4.59–4.50
passed down a short plug of silica gel (elution with EtOAc) (2H, CH2Ar), 3.82–3.47 (8H, OCH3, H-2 propyl, 1- and 3-CH2
and, after evaporation to dryness, purified by RBC (elution gra- propyl), 3.25 (m, 2H, H-1 and 2), 2.50 (bs, 1H, OH), 2.40 (bs,
dient hexane → 5 : 1 Et2O–hexane) to afford the allyl product 1H, OH), 2.11–2.01 (m, 2H, cyclitol), 1.68 (m, 2H, cyclitol),
38 (712 mg, 69%) as a clear oil; δH (500 MHz, CDCl3) 7.32 (d, 1.23 (m, 4H, cyclitol); δC (125 MHz, CDCl3) 159.3, 159.2, 130.5,
2H, J = 8.6 Hz, Ph), 6.90 (d, 2H, J = 8.7 Hz, Ph), 6.02–5.94 (m, 130.4, 129.5, 129.4, 113.84, 113.83, 83.40 (C-1 or 2), 82.39 (CH,
1H, CH2CHvCH2), 5.32–5.17 (4 × m, 2H, CH2CHvCH2), 4.63 C-1 or 2), 80.86 (C-1 or 2), 80.78 (C-1 or 2), 72.31 (1- or 3-CH2
(dd, 2H, J = 11.4 Hz, CH2Ar), 4.17 (m, 2H, CH2CHvCH2), 3.83 propyl), 71.37 (C-2 propyl), 71.03 (1- or 3-CH2 propyl), 70.81
(s, OCH3), 3.37–3.29 (m, 2H, H-1 and 2), 2.03–2.00 (m, 2H, (1- or 3-CH2 propyl), 70.51 (C-2 propyl), 64.20 (1- or 3-CH2
cyclitol), 1.69–1.67 (m, 2H, cyclitol), 1.39–1.20 (m, 4H, cyclitol); propyl), 63.83 (1- or 3-CH2 propyl), 55.24 (OCH3), 30.75, 30.54,
δC (125 MHz, CDCl3) 159.0, 135.8 (CH2CHvCH2), 131.5, 129.1, 29.99, 29.93, 23.78, 23.71, 23.69; HRMS (ESI) calcd for
116.1 (CH2CHvCH2), 113.7, 81.0 (C-1 or 2), 80.8 (C-1 or 2), C17H26NaO5 [M + Na]+ 333.1672, found 333.1678.
1930 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
1-[(tert-Butyldiphenylsilyl)oxy]-3-{[trans-2-((4-methoxybenzyl)-
oxy)cyclohexyl]oxy}propan-2-ol 41
C-1 or 2), 82.1 (CH, C-1 or 2), 81.9 (CH, C-1 or 2), 80.5 (CH,
C9), 71.2, 71.0, 68.9, 68.8, 63.5, 55.3 (OCH3), 38.5 (SO2CH3),
38.4, 30.0, 30.0, 29.8, 26.8, 23.5, 23.4, 19.2; HRMS (ESI) calcd
for C34H46NaO7SSi [M + Na]+ 649.2626, found 649.2628.
To a solution of the primary alcohol 40 (942 mg, 3.34 mmol)
and DIPA (5.8 mL, 3.75 mmol) in CH2Cl2 (5 mL) was added
TBDPSCl (1.04 mL, 4.00 mmol) dropwise followed by DMAP
(5 mg, 0.038 mmol) and the reaction stirred for 24 h at rt.
Afterwards, TLC revealed the presence of the starting material
40; thus an additional aliquot of TBDPSCl (0.521 mL,
2.00 mmol) was added and the reaction mixture was stirred for
a further 3 h, whereafter it was quenched with water (60 mL)
and then extracted with CH2Cl2 (3 × 60 mL). The combined
organic extracts were washed successively with water (100 mL),
brine (50 mL), filtered through cotton wool and the solvent
was removed under reduced pressure. The residue so obtained
was percolated through a short column of silica gel (further
elution with EtOAc) and the subsequent eluent was concen-
trated under reduced pressure. RBC (elution first with hexane
→ 1 : 2 EtOAc–hexane) of the residue afforded the silyl pro-
tected product 41 (1.33 g, 76%) as a pale yellow oil; δH
(500 MHz, CDCl3) 7.68–6.79 (14H, Ph), 4.56–4.44 (m, 2H,
CH2Ar), 3.90–3.51 (8H, OCH3, H-2 propyl, 1- and 3-CH2
propyl), 3.25 (m, 2H, H-1 and 2), 3.12 (bs, 1H, OH), 1.97 (m,
2H, cyclitol), 1.66 (m, 2H, cyclitol), 1.30–1.15 (m, 4H, cyclitol),
1.05 (s, 9H, 3 × CH3); δC (125 MHz, CDCl3) 159.1, 135.6, 135.6,
134.7, 133.5, 133.49, 133.42, 130.9, 130.8, 129.73, 129.72,
129.3, 129.2, 127.74, 127.71, 113.79, 113.77, 82.9 (C-1 or 2),
82.3 (C-1 or 2), 80.7 (C-1 or 2), 80.6 (C-1 or 2), 71.7, 71.6, 71.2,
71.1, 71.0, 70.4, 65.0, 64.8, 55.3, 55.2, 30.6, 30.4, 30.11, 30.06,
29.7, 26.9, 23.69, 23.66, 19.29, 19.28; HRMS (ESI) calcd for
C33H44NaO5Si [M + Na]+ 571.2856, found 571.2860.
{2-Azido-3-[(trans-2-((4-methoxybenzyl)oxy)cyclohexyl)oxy]-
propoxy} (tert-butyl)diphenylsilane 43
A solution of the mesylate 42 in DMF (10 mL) containing
sodium azide (496 mg, 7.64 mmol) was heated and stirred at
125 °C for 24 h, cooled and then poured into water (40 mL).
The resulting aqueous solution was extracted with CH2Cl2 (3 ×
40 mL) and the combined organic extracts were washed succes-
sively with water (100 mL), brine (100 mL), filtered through
cotton wool and concentrated under reduced pressure. A solu-
tion of the residue in EtOAc was percolated through a short
column of silica gel (elution with EtOAc) and the eluent con-
centrated under reduced pressure. RBC of the residue (elution
first with hexane → 1 : 1 Et2O–hexane) gave the azide 43
(1.30 g, 51%) as a clear oil; δH (500 MHz, CDCl3) 7.62–6.73
(14H, Ph), 4.44 (s, 2H, CH2Ar), 3.71–3.32 (8H, OCH3, H-2
propyl, 1- and 3-CH2 propyl), 3.22 (m, 2H, H-1 and 2), 1.86 (m,
2H, cyclitol), 1.55 (m, 2H, cyclitol), 1.25–1.04 (m, 4H, cyclitol),
0.99 (s, 9H, 3 × CH3); δC (125 MHz, CDCl3) 159.0, 135.9, 135.8,
135.6, 135.3, 134.8, 133.10, 133.06, 131.31, 131.29, 129.8,
129.7, 129.2, 129.12, 129.08, 129.0, 127.8, 127.73, 127.67,
113.74, 113.70, 82.2 (C-1 or 2), 82.0 (C-1 or 2), 80.5 (C-1 or 2),
80.4 (C-1 or 2), 71.5, 71.4, 71.19, 69.5, 69.0, 64.08, 64.05, 63.2,
63.0, 55.3 (OCH3), 30.0, 29.9, 26.9, 26.8, 26.6, 23.43, 23.40,
23.35, 19.2; HRMS (ESI) calcd for C33H43N3NaO4Si [M + Na]+
596.2915, found 596.2926.
1-[(tert-Butyldiphenylsilyl)oxy]-3-{(trans-2-[(4-methoxybenzyl)-
trans-2-{2-Azido-3-[(tert-butyldiphenylsilyl)oxy]propoxy}-
oxy)cyclohexyl]oxy}propan-2-yl methanesulfonate 42
cyclohexanol 44
To the secondary alcohol 41 (1.33 g, 2.54 mmol) in pyridine The PMB derivative 43 (131 mg, 0.240 mmol) was dissolved in
(5 mL) was added mesyl chloride (0.63 mL, 8.14 mmol) drop- a solution of 1% TFA in CH2Cl2 (8.62 mL). The reaction
wise. The reaction mixture was stirred at room temperature for mixture was stirred at room temperature for 18 h; whereafter
24 h and then quenched with saturated NaHCO3 (10 mL) fol- an additional aliquot of TFA (43 µL) was added because TLC
lowed by extraction with CH2Cl2 (3 × 20 mL). The combined indicated the presence of the PMB protected starting material
organic extracts were washed with water (60 mL), brine (43). After a further 4 h, TLC revealed the absence of any start-
(60 mL), filtered through cotton wool and then evaporated to ing material (43) and the reaction mixture was diluted with
dryness; whereafter toluene (5 mL) was added and evaporated CH2Cl2 (40 mL) and washed with saturated NaHCO3 (40 mL),
therefrom. The resulting oil was passed through a short water (40 mL), brine (40 mL) and then filtered through cotton
column of silica gel (elution with EtOAc) to give a clear yellow wool. The CH2Cl2 extract was concentrated under reduced
oil after evaporation to dryness under reduced pressure. This pressure and the residue was percolated through a short
oil was used in the following step without further purification. column of silica gel (elution with EtOAc) and concentrated to
However, a small sample of the product (100 mg) was purified dryness under reduced pressure. RBC purification of the
by RBC (1 : 1 Et2O–hexane) to afford a clear, colourless oil of 42 residue (elution first with hexane
→ 1 : 1 Et2O–hexane)
for analytical analyses; δH (500 MHz, CDCl3) 7.60–6.73 (14H, afforded the alcohol 44 (67 mg, 64%) as a clear oil; δH
Ph), 4.70 (m, 1H, H-2 propyl), 4.61–4.35 (m, 2H, CH2Ar), (500 MHz, CDCl3) 7.60–7.31 (10H, Ph), 3.74–3.11 (6H, H-1 or 2,
3.83–3.69 (m, 7H, OCH3, 1- and 3-CH2 propyl), 3.18 (m, 2H, H-2 propyl, 1- and 3-CH2 propyl), 2.99 (m, 1H, H-1 or 2), 2.60
H-1 and 2), 2.90 (s, 3H, SO2CH3), 1.90 (m, 2H, cyclitol), 1.57 (bd, 1H, OH), 1.96 (m, 2H, cyclitol), 1.65 (m, 2H, cyclitol),
(m, 2H, cyclitol), 1.24–1.08 (m, 4H, cyclitol), 0.98 (s, 9H, 3 × 1.23–1.03 (m, 4H, cyclitol), 1.00 (s, 9H, 3 × CH3); δC (125 MHz,
CH3); δC (125 MHz, CDCl3) 159.1, 135.6, 135.54, 135.52, CDCl3) 134.75, 134.58, 134.56, 132.2, 131.91, 131.87, 128.9,
132.94, 132.89, 132.8, 131.1, 131.0, 129.92, 129.90, 129.2, 126.81, 126.79, 83.6 (C-1 or 2), 83.5 (C-1 or 2), 72.8, 72.7, 67.3,
129.1, 127.9, 113.78, 113.77, 82.5 (CH, CH propyl), 82.4 (CH, 67.0, 62.8, 62.7, 62.0, 61.7, 31.02, 30.98, 28.7, 28.1, 25.9, 25.7,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1931

View Article Online
Paper
Organic & Biomolecular Chemistry
23.1, 22.9, 18.2; HRMS (ESI) calcd for C25H35N3NaO3Si Triethylammonium trans-2-(2-amino-3-hydroxypropoxy)-
[M + Na]+ 476.2340, found 476.2351.
cyclohexyl n-octadecyl phosphate 13
Pearlman’s catalyst [10–20% Pd(OH)2 on carbon, 15 mg] was
added to a solution of the azide 46 (45 mg, 0.069 mmol) in
1 : 1 THF–MeOH (10 mL) and the mixture was stirred under a
hydrogen atmosphere at rt for 2 h. Processing as described for
21 gave the amino TEA salt 13 (29 mg, 44%) as a white paste,
which did not require any chromatographic purification; δH
(500 MHz, CDCl3) 4.00–3.31 (8H, OCH2, H-1 or 2, H-2 propyl,
1- and 3-CH2 propyl), 3.10 (m, 1H, H-1 or 2), 3.02 (q, 6H, J =
7.4 Hz, 3 × CH2CH3), 2.04–1.93 (m, 2H, cyclitol), 1.63–1.50 (m,
4H, OCH2CH2 and 2H cyclitol), 1.33–1.05 (43H, [CH2]15, 3 ×
CH2CH3 and 4H cyclitol), 0.81 (t, 3H, CH3, J = 6.7 Hz,
CH2CH3); δC (125 MHz, CDCl3) 81.6 (C-1 or 2), 77.9 (C-1 or 2),
67.1, 65.9, 64.8, 53.2, 44.5 [N(CH2CH3)3], 33.0, 31.8, 30.9, 29.7,
29.6, 29.33, 29.29, 29.0, 28.70, 28.65, 28.43, 28.35, 27.9, 24.9,
24.8, 24.8, 24.7, 23.2, 23.1, 22.8, 22.7, 13.1 (CH2CH3), 7.5
[N(CH2CH3)3]; δP (202 MHz, CDCl3) −0.23 (with heteronuclear
decoupling); HRMS (ESI) calcd for C27H55NO6P [M − NEt3 − H]−
520.3772, found 520.3747.
Triethylammonium trans-2-{2-azido-3-[(tert-butyldiphenylsilyl)-
oxy]propoxy}cyclohexyl n-octadecyl phosphate 45
This compound was obtained from the alcohol 44 (280 mg,
0.62 mmol) and the hydrogenphosphonate TEA salt 1815
(537 mg, 1.23 mmol) in the presence of pivaloyl chloride
(0.48 mL, 3.86 mmol) essentially as described for the TEA salt
19. After oxidation with iodine (623 mg, 2.47 mmol) in 9 : 1
pyridine–water followed by the same aqueous workup as
described for 19, column chromatography (CH2Cl2 → 8 : 1
CH2Cl2–MeOH) of the residue afforded the octadecyl phos-
phate TEA salt 45 (276 mg, 50%) as a yellow paste; δH
(500 MHz, CDCl3) 7.63–7.30 (10H, Ph), 4.10 (m, 1H, H-1 or 2),
3.92–3.33 (7H, OCH2, H-2 propyl, 1- and 3-CH2 propyl),
3.28 (m, 1H, H-1 or 2), 3.00 (m, 6H,
3 × CH2CH3),
1.95–1.73 (m, 2H, cyclitol), 1.58–1.46 (m, 4H, OCH2CH2 and
2H cyclitol), 1.35–1.14 (43H, [CH2]15, 3 × CH2CH3 and 4H cycli-
tol), 1.73 (s, 9H, 3 × CH3), 0.80 (t, 3H, CH3, J = 6.8 Hz,
CH2CH3); δC (125 MHz, CDCl3) 134.9, 134.8, 134.6, 132.6,
132.4, 132.1, 132.0, 128.9, 128.8, 126.8, 126.7, 78.8 (C-1 or 2),
78.6 (C-1 or 2), 76.6 (C-1 or 2), 70.8, 70.1, 68.2, 67.7, 65.7, 65.6,
65.5, 63.0, 62.9, 44.5 [N(CH2CH3)3], 30.9, 29.7, 29.6, 29.5, 29.0,
28.70, 28.65, 28.4, 27.2, 26.0, 25.9, 25.7, 24.8, 24.7, 21.7, 21.1,
Biological assays
Materials
21.0, 18.2, 14.3 (CH2CH3), 8.5 [N(CH2CH3)3]; δP (202 MHz, The synthesis of 1-D-6-O-(2-amino-2-deoxy-α-D-glucopyranosyl)-
CDCl3) −1.2 (with heteronuclear decoupling); HRMS (ESI) myo-inositol 1-(octadecyl phosphate), (4, α-D-GlcpNH2-IPC18),15
calcd for C43H71N3O6PSi [M − NEt3 − H]− 784.4885, found has been described previously. The corresponding N-acetyl
784.4759.
derivate α-D-GlcpNAc-IPC18 (3) was prepared by treatment with
acetic anhydride,13 and the concentration of stock solutions
determined by measurement of the inositol content by
selected ion-monitoring GC-MS.8 Bloodstream form Trypano-
some brucei (variant MITat1.4) were isolated and membranes
(cell-free system) prepared as described previously and stored
at −80 °C.26
Triethylammonium trans-2-(2-azido-3-hydroxypropoxy)-
cyclohexyl n-octadecyl phosphate 46
Compound 45 (68 mg, 0.077 mmol) was dissolved in THF
(1 mL) and 1.0 M TBAF in THF (153 µL, 0.15 mmol) was
added. The reaction mixture was stirred at rt for 16 h and then
diluted with water (25 mL), extracted CH2Cl2 (3 × 25 mL) and
the combined organic extracts were washed with aq. 1.0 M
Activity assays
TEAB (2 × 10 mL). The organic phase was filtered through Substrate recognition assays were performed using 500 pmol
cotton wool, the solvent was removed under reduced pressure of α-D-GlcpNAc-PI (1) in incorporation buffer (25 mM Tris pH
and the resulting residue was purified by column chromato- 8.0, 50 mM KCl, 50 mM MnCl2) and varying amounts of
graphy (8 : 1 CH2Cl2–MeOH) to afford the alcohol 46 (50 mg, trypanosome cell-free system (0–15 × 106 cell equivalents per
100%) as a white paste; δH (500 MHz, CDCl3) 4.08–3.33 (8H, assay) in 96-well plates containing 100 μL final volume, and
OCH2, H-1 or 2, H-2 propyl, 1- and 3-CH2 propyl), 3.25 (m, 1H, incubated at 37 °C for 1 h. The reaction was quenched and the
H-1 or 2), 3.09 (m, 6H, 3 × CH2CH3), 2.14–1.99 (m, 2H, cycli- glycolipids enriched and analyzed by LC-MS/MS as described
tol), 1.66–1.59 (m, 4H, OCH2CH2 and 2H cyclitol), 1.30–1.19 below.
(43H, [CH2]15, 3 × CH2CH3 and 4H cyclitol), 0.88 (t, 3H, CH3,
Inhibition assays were performed in 96-well plates in
J = 6.5 Hz, CH2CH3); δC (125 MHz, CDCl3) 81.6 (C-1 or 2), 78.6 100 μL final volume, with 1% v/v DMSO with or without inhibi-
(C-1 or 2), 68.6, 68.5, 66.03, 65.99, 62.4, 61.7, 60.9, 60.6, 45.4 tor. Trypanosome cell-free system (2.5 × 106 cell equivalents
[N(CH2CH3)3], 32.2, 32.0, 31.9, 30.8, 30.74, 30.68, 29.73, 29.67, per assay) in incorporation buffer were added to wells contain-
29.42, 29.38, 25.8, 23.9, 23.8, 23.8, 23.7, 22.7, 14.1 (CH2CH3), ing 500 pmol α-D-GlcpNAc-IPC18 (3) with or without inhibitor
8.5 [N(CH2CH3)3]; δP (202 MHz, CDCl3) −1.02 (with hetero- and incubated at 37 °C for 1 h. The reaction was quenched
nuclear decoupling); HRMS (ESI) calcd for C27H53N3O6P and the glycolipids enriched and analyzed by LC-MS/MS as
[M − NEt3 − H]− 546.3677, found 546.3673.
described below.
1932 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Glycolipid enrichment
Acknowledgements
Enrichment of glycolipids was performed in a 96-well plate
format. Reactions were quenched by addition of 200 μL of 5%
propan-1-ol, 5 mM NH4OAc, and the glycolipids were bound to
C18 resin (50 mg Isolute Array cartridge), washed three times
with 200 μL 5% propan-1-ol, 5 mM NH4OAc and eluted with
100 μL 40% propan-1-ol, 5 mM NH4OAc into a 96-well collec-
tion plate.
We would like to acknowledge the Wellcome Trust (pro-
gramme grant 085622 and strategic awards 08348 and 100476)
and the MRC (studentship to ASC) for financial support.
Prior to subsequent analysis, compound 13 was dried References
under nitrogen, resuspended in MeOH (100 μL) and any free
1 M. A. J. Ferguson, J. Cell Sci., 1999, 112, 2799–2809.
amine reacted with excess d6-Ac2O (1.5 μL) in the presence of
pyridine (10 μL) for 15 min. The reaction was quenched with
water (50 μL), dried under nitrogen and resuspended in 100 μL
40% propan-1-ol, 5 mM NH4OAc.
2 D. K. Sharma, T. K. Smith, C. T. Weller, A. Crossman,
J. S. Brimacombe and M. A. J. Ferguson, Glycobiology, 1999,
9, 415–422.
3 T. K. Smith, M. J. Patterson, A. Crossman, J. S. Brimacombe
and M. A. J. Ferguson, Biochemistry, 2000, 39, 11801–11807.
4 T. K. Smith, A. Crossman, C. N. Borrissow, M. J. Paterson,
A. Dix, J. S. Brimacombe and M. A. J. Ferguson, EMBO J.,
2001, 20, 3322–3332.
Liquid chromatography – tandem mass spectrometry of
glycolipids
5 M. D. Urbaniak, A. Crossman and M. A. J. Ferguson, Chem.
Biol. Drug Des., 2008, 72, 127–132.
6 M. D. Urbaniak, D. V. Yashunsky, A. Crossman,
A. V. Nikolaev and M. A. J. Ferguson, ACS Chem. Biol., 2008,
3, 625–634.
7 N. Z. Abdelwahab, A. T. Crossman, M. D. Urbaniak and
M. A. J. Ferguson, Carbohydr. Res., 2011, 346, 708–714.
8 N. Z. Abdelwahab, A. T. Crossman, L. Sullivan,
M. A. J. Ferguson and M. D. Urbaniak, Chem. Biol. Drug
Des., 2012, 79, 270–278.
9 D. K. Sharma, T. K. Smith, A. Crossman, J. S. Brimacombe
and M. A. J. Ferguson, Biochem. J., 1997, 328, 171–177.
10 M. D. Urbaniak, A. Crossman, T. Chang, T. K. Smith,
D. M. F. v. Aalten and M. A. J. Ferguson, J. Biol. Chem.,
2005, 280, 22831–22838.
11 T. Chang, K. G. Milne, M. L. S. Guther, T. K. Smith and
M. A. J. Ferguson, J. Biol. Chem., 2002, 277, 50176–
50182.
12 N. A. Meanwell, J. Med. Chem., 2011, 54, 2529–2591.
13 T. K. Smith, S. Cottaz, J. S. Brimacombe and
M. A. J. Ferguson, J. Biol. Chem., 1996, 271, 6476–6482.
14 G. Grundler and R. R. Schmidt, Liebigs Ann. Chem., 1984,
1826–1847.
Glycolipids were analyzed by liquid chromatography coupled
to an electrospray tandem mass spectrometer (LC-MS/MS).
Samples (40 μL) were injected directly from a 96-well plate
onto a 10 × 1 mm C18 column (ACE, 5 μM) and then eluted
using a binary gradient of 5–80% propan-1-ol in 5 mM
NH4OAc (Dionex Ultimate 3000). The gradient consisted of
2 min 0% B, 2–4 min 0–100% B, 4–8 min 100%, 8–9 min
100–0% B, 9–10 min 0% B where buffer A consisted of 5%
propan-1-ol, 5 mM NH4OAc and buffer B 80% propan-1-ol,
5 mM NH4OAc. The glycolipids were analysed on an electro-
spray triple quadrapole mass spectrometer (Micromass
Quattro Ultima) in multiple reaction monitoring mode.
For each pseudodisaccharide analogue (7–12), standards of
the N-acetylated compound and corresponding free amine
were analyzed separately in order to identify unique transitions
for use in subsequent multiple reaction monitoring
experiments (Table 1). The ratio of the integrals for these tran-
sitions was used to calculate the percentage of substrate con-
version to product in a given sample. For compound 13,
standards of the N-acetylated compound and the d3-N-acetyl-
ated form were analyzed separately and found to produce a
common fragment for use in subsequent multiple reaction
monitoring experiments.
15 A. Crossman, M. J. Patterson, M. A. J. Ferguson, T. K. Smith
and J. S. Brimacombe, Carbohydr. Res., 2002, 337, 2049–
2059.
16 A. V. Nikolaev, I. A. Ivanova, V. N. Shibaev and
N. K. Kochetkov, Carbohydr. Res., 1990, 204, 65–78.
17 H. F. Russell, J. B. Bremner, J. Bushelle-Edghill,
M. R. Lewis, S. R. Thomas and F. Bates II, Tetrahedron Lett.,
2007, 48, 1637–1639.
For inhibition assays, the turnover of the substrate
α-D-GlcpNAc-IPC18 (3) was used to calculate the percentage of
substrate conversion to product in a given sample.10 Inhibitor
IC50 values were calculated using a four-parameter fit of eight-
point potency curves derived from three independent experi-
ments, and are quoted with a standard deviation.
18 Y. G. Gololobov, I. N. Zhmurova and L. F. Kasukhin, Tetra-
hedron, 1981, 37, 437–472.
Trypanosome cell-free system assays
The formation of GPI precursors is monitored by following the 19 I. Lindh and J. Stawinski, J. Org. Chem., 1989, 54, 1338–
incorporation of [3H]-mannose and then they were analysed
1342.
using high-performance liquid chromatography and fluoro- 20 S. Cottaz, J. S. Brimacombe and M. A. J. Ferguson, Carbo-
graphy as described previously.8
hydr. Res., 1995, 270, 85–91.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 1919–1934 | 1933

View Article Online
Paper
Organic & Biomolecular Chemistry
21 J. Barluenga, H. Vázquez-Villa, A. Ballesteros and 24 G. Berti, B. Macchia and F. Macchia, Tetrahedron Lett.,
J. M. Gonzalez, Org. Lett., 2002, 4, 2817–2819. 1965, 38, 3421–3427.
22 A. Tschöp, A. Marx, A. R. Sreekanth and C. Schneider, 25 M. L. S. Guther and M. A. J. Ferguson, EMBO J., 1995, 14,
Eur. J. Org. Chem., 2007, 2318–2327. 3080–3093.
23 A. S. Capes, A. T. Crossman, L. A. Webster, M. A. J. Ferguson 26 W. J. Masterson, T. L. Doering, G. W. Hart and
and I. H. Gilbert, Tetrahedron Lett., 2011, 52, 7091–7094. P. W. Englund, Cell, 1989, 62, 73–80.
1934 | Org. Biomol. Chem., 2014, 12, 1919–1934
This journal is © The Royal Society of Chemistry 2014