Synthesis of new sulfonate and phosphonate derivatives for cation-independent mannose 6-phosphate receptor targeting

Article abstract of DOI:10.1016/j.bmcl.2008.09.101

The cation-independent mannose 6-phosphate receptor (CI-M6PR) is essential for the endocytosis of proteins bearing the mannose 6-phosphate (M6P) recognition marker. This study described the synthesis of M6P and M6S analogs presenting greater affinity for

Full text of DOI:10.1016/j.bmcl.2008.09.101

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6240–6243
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of new sulfonate and phosphonate derivatives for
cation-independent mannose 6-phosphate receptor targeting
Audrey Jeanjean a, , Magali Gary-Bobo b,c,d,e, , Philippe Nirdé b,c,d,e, Simon Leiris a, Marcel Garcia b,c,d,e, Alain
Morère a,
*
a Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, CNRS, Université Montpellier 2 et 1, Bâtiment de Recherche Max Mousseron,
Ecole Nationale Supérieure de Chimie de Montpellier, 8 Rue de l’Ecole Normale, F-34296 Montpellier Cedex 05, France
b IRCM, Institut de Recherche en Cancérologie de Montpellier, Montpellier F-34298, France
c INSERM, Unité 896, Montpellier F-34298. France
d Université Montpellier 1, Montpellier F-34298, France
e CRLC Val d’Aurelle Paul Lamarque, Montpellier F-34298, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The cation-independent mannose 6-phosphate receptor (CI-M6PR) is essential for the endocytosis of pro-
teins bearing the mannose 6-phosphate (M6P) recognition marker. This study described the synthesis of
M6P and M6S analogs presenting greater affinity for CI-M6PR than their natural compounds. Moreover,
the finding of their lack of cytotoxicity for human cells and of their increased stability in human serum
supports the high potential of these isosteric derivatives in therapies requiring CI-M6PR targeting.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 1 August 2008
Revised 26 September 2008
Accepted 28 September 2008
Available online 2 October 2008
Keywords:
Mannose 6-phosphate analogs
Mannose 6-sulfate analogs
Cation-independent mannose 6-phosphate
receptor
Binding affinity
The cation-independent mannose 6-phosphate receptor (CI-
M6PR), a transmembrane glycoproteinof 300 kDa, playsvery impor-
tant roles in many biological processes.1–3 The main role of CI-M6PR
is transporting and sorting those lysosomal enzymes that contain
the mannose 6-phosphate (M6P) recognition marker in their struc-
ture from the trans-Golgi network to the lysosomes.4 CI-M6PR also
mediatestheendocytosisof extracellularligandssuch asinsulin-like
growth factor II (IGF2),5 retinoic acid6 and M6P-containing pro-
teins,7 which differ from lysosomal enzymes and are internalized
through CI-M6PR transport, include Granzyme B8, the herpes sim-
plex virus (HSV)9, leukemia inhibitory factor (LIF),10,11 and partially
renin.12 CI-M6PR also acts on molecules that do not penetrate into
cells such as the latent precursor of transforming growth factor-beta
Among the various proteins and lysosomal enzymes bearing the
M6P signal, CI-M6PR binds particular lysosomal enzymes with
high affinity, such as those of Dictyostelium discoideum, which con-
18
tains in its structure the phosphodiester Man-P-OCH3 and M6S
residues.19 This flexibility in the binding of various derivatives of
mannose, functionalized at the 6-position with phosphate or sul-
fate moieties, led us to study the impact of mannose derivatives,
modified at 6-position, on the binding to CI-M6PR. It has been
shown that an isosteric phosphonate analog of M6P binds with
high affinity to the receptor20,21 while a non-isosteric phospho-
nate20 or b-hydroxyphosphonates22 are not recognized. Moreover,
the flexibility of CI-M6PR with regard to the binding of C-6-modi-
fied mannose derivatives was confirmed by the biological study of
a series of isosteric carboxylate analogs of M6P.23–25 The aim of the
work described herein was to complete these studies with the syn-
theses of sulfonate and phosphonate analogs of M6P for subse-
quent use of these derivatives in such fields as healing
improvement26 or enzyme replacement therapy23 which main lim-
iting factor is the instability of enzyme M6P residues.
(L
-TGFb)13 and on plasminogen/plasmin conversion.14,15
The phosphate moiety as well as the hydroxyl groups on the
mannopyranosidic ring of M6P contribute to a hydrogen-bonding
network with two binding sites of CI-M6PR.16 This ability to recog-
nize two M6P residues allows CI-M6PR to bind lysosomal enzymes
with high affinity (Kd = 10À9 M).17
Two isosteric sulfonate analogs of M6S were synthesized. The
unconjugated one (Scheme 1) was prepared starting from methyl
* Corresponding author. Tel.: +33 467 14 43 45; fax: +33 467 14 43 44.
E-mail address: alain.morere@univ-montp2.fr (A. Morère), alain.morere@univ-
montp2.fr (A. Morère).
2,3,4-tri-O-benzyl-a-D
-mannopyranoside,27 which was converted
to an aldehyde by Swern oxidation followed by a Wittig–Horner
reaction using the anion of triethyl phosphonomethanesulfonate.28
These authors contribute equally to this work.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.09.101

A. Jeanjean et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6240–6243
6241
tetraethyl methylenediphosphonate. The phosphonate 731 was
then desilylated by treatment with HCl 1 M to afford 831 in 95%
yield. The last step was the deprotection of the phosphonate moi-
ety using the classical Rabinowitz procedure.33 The target unsatu-
rated phosphonate 9 was obtained in 65% yield.
O
EtO
S
OH
OH
O
OBn
O
a
O
BnO
BnO
HO
HO
The binding affinities of the isosteric analogs of M6P for CI-
M6PR were determined using the protocol we recently described.24
Briefly, increasing concentrations of the analogs were added to the
biotinylated CI-M6PR (CI-M6PRb), which was bound to pentaman-
nose 6-phosphate (PMP) adsorbed on a microtiter plate, in order to
displace the receptor from its ligand (PMP). The remaining bound
CI-M6PRb was quantified using the streptavidin/peroxidase couple
and o-phenylenediamine (OPD) substrate by optical density mea-
surements. From these experiments, dose-inhibition curves were
drawn (Fig. 1) and used to calculate the IC50 values for each isoster-
ic analog of M6P (Table 1). In a previous paper we demonstrated
that M6S shows high affinity for CI-M6PR.25 Indeed, as it is the case
with M6P itself, 2.5 mM of the M6S was demonstrated to be suffi-
cient to displace the CI-M6PR retained on the PMP-sepharose col-
umn. This result was confirmed by determination of the binding
affinity (Table 1), which was in good agreement with the literature
data. Therefore, M6S bound to CI-M6PR 10-fold less well than M6P.
Contrary to M6S, the sulfonates were more effectively bound to CI-
M6PR. Indeed, unsaturated sulfonate 6 and saturated sulfonate 3
had affinities for CI-M6PR, respectively, 3.2- and 1.5-fold higher
than this of M6S. The finding that compound 6 has higher affinity
for CI-M6PR than 3 was not surprising. Indeed, we recently re-
ported that the unsaturated carboxylate M6C-u displayed 2-fold
higher affinity for CI-M6PR than the saturated carboxylate M6C-
s.24 We assume that in both cases, the unsaturated carboxylate
and sulfonate derivatives better matched the binding pocket of
CI-M6PR than the saturated analogs. Therefore, in order to com-
plete this study, we extended the comparison to phosphonate
derivatives, which are well known to be good bioisosteres of
phosphates.34 The binding affinity for CI-M6PR of the saturated
phosphonate (M6Pn-s) isosteric analog of M6P was found to be
1.8-fold higher than that of M6P (Table 1) and this result was in
very good accordance with the literature data.21 Surprisingly, the
unsaturated phosphonate 9 bound to CI-M6PR less well than the
M6Pn-s (Table 1). Indeed, the binding affinity of 9 for CI-M6PR
was 2-fold lower than that of M6Pn-s. However, it is noteworthy
that the conjugated phosphonate 9 still binds to CI-M6PR as well
as M6P itself (Table 1). This suggests that, contrary to the sulfonate
OCH3
OCH3
1
b
O
O
nBu4NO
O
NaO
S
S
c
O
OBn
O
OH
O
BnO
BnO
HO
HO
OCH3
OCH3
3
2
Scheme 1. Reagents and conditions: (a) i—(COCl)2, DMSO, iPr2NEt, À60 °C; ii—
EtO3SCH2PO(OEt)2, nBuLi, À68 °C, THF, 55%; (b) nBu4NI, acetone, reflux, 15 h, 81%;
(c) iv—DOWEX Na+, H2O/MeOH (8/2), 1 h; v—H2/Pd/C, EtOH/AcOEt (7/3), 4 h, 82%.
Deprotection of the sulfonate moiety of 1 was then performed by
an SN2 procedure29,30 using tetrabutylammonium iodide to lead
to 2 in 81% yield. Complete debenzylation and double bond reduc-
tion of 2 was achieved by catalytic hydrogenation (H2/Pd/C) to af-
ford the saturated sulfonate 3.
The conjugated sulfonate (Scheme 2) was prepared starting from
methyl
2,3,4,6-tetra-O-trimethylsilyl-a-D
-mannopyranoside,31
which was directly converted to an aldehyde at 6-position in a
one-step procedure.32 Compound 4 was obtained by addition of
the triethyl phosphonomethanesulfonate anion to this aldehyde.
Deprotection of the sulfonate moiety was performed by treatment
with tetrabutylammonium iodide, as described above. Compound
5, thus obtained, was converted to the target unsaturated sulfonate
6 by a desilylation method using a catalytic amount of ceric ammo-
nium nitrate (CAN) in a mixture of acetonitrile and water (95/5, v/v).
The first step in the preparation of unsaturated phosphonate
was performed similarly to 6 (Scheme 2). It consisted of the direct
oxidation of methyl 2,3,4,6-tetra-O-trimethylsilyl-a-D-mannopy-
ranoside31 in aldehyde using the Collins oxidation procedure, fol-
lowed by the Wittig–Horner reaction using the anion of
R1
OTMS
OR2
OTMS
a
O
R2O
R2O
TMSO
TMSO
O
OCH3
OCH3
4
5
7
R1
;
R2
b
d
4 = SO3Et ; TMS
8
5 = SO3NBu4 ; TMS
7 = PO(OEt)2 ; TMS
8 = PO(OEt)2 ; H
c
e
O
O
NaO
NaO
P
S
NaO
O
OH
O
OH
O
HO
HO
HO
HO
OCH3
OCH3
9
6
Figure 1. Dose-dependent inhibition of CI-M6PR binding to PMP by M6P analogs.
The binding of CI-M6PRb to PMP was inhibited by: (A) increasing concentrations of
M6P, saturated (M6Pn-s) or unsaturated (9) M6-phosphonates, and (B) increasing
concentrations of M6S or saturated (3) or unsaturated (6) M6-sulfonates. The data
represent the mean of triplicates from a typical experiment and were confirmed in
two additive experiments.
Scheme 2. Reagents and conditions: (a) i—CrO3/pyridine, CH2Cl2, 0 °C, 1 h; ii—4:
(EtO)2OPCH2SO3Et, nBuLi, THF, À68 °C, 57%; 7: ((EtO)2OP)2CH2, NaH, THF, 20 °C,
30 min, 49%; (b) nBu4NI/acetone, reflux, 15 h, 95%; (c) CAN, CH3CN/H2O (5/5), 2 h
then DOWEX Na+, 4 h, 98%; (d) HCl 1 M, THF, 20 °C, 10 min, 95%; (e) TMSBr,
pyridine, 2 h then DOWEX Na+ 5 h, 65%.

6242
A. Jeanjean et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6240–6243
Table 1
Binding affinities for CI-M6PR
posed of six residues that stabilized the ligand (Fig. 2B). In
contrast, the sulfonate analog 3 (Fig. 2C) was associated with a
marked decrease in the binding affinity, and only five residues sta-
bilized this ligand. However, after minimization, moderate orienta-
tions of the side chains could have occurred and the strong
interaction of residue S386, which is involved in ligand stability
as revealed by single amino acid substitutions,35 with mannose
was reduced for this analog 3. Taken together, these structural
parameters might explain the low binding affinity of this analog
for CI-M6PR.
RBAb
a
Compound
IC50 (M)
M6P
2.3 Â 10À5 ( 1.0 Â 10À5
1.3 Â 10À5 ( 0.2 Â 10À5
2.2 Â 10À5 ( 0.3 Â 10À5
2.3 Â 10À4 ( 0.9 Â 10À4
1.5 Â 10À4 ( 0.6 Â 10À4
7.1 Â 10À5 ( 1.1 Â 10À5
)
)
)
)
)
)
1.0
1.8
1.05
0.1
0.15
0.3
M6Pn-s20,21
9
M6S25
3
6
a
IC50 were obtained from 3 to 5 independent experiments ( SD).
Relative binding affinity = RBA = IC50(M6P)/IC50(Analogue)
b
.
Since we previously showed that M6P is unstable in human ser-
um,24 we compared the stability of the analogs 3, 6, 9, M6Pn-s and
M6S to M6P. After a 2 day-incubation in human serum at 37 °C the
binding capacity of all analogs to CI-M6PR was maintained while
the affinity of M6P decreased about 5-fold (Fig. 3). This indicates
that, in contrast to M6P, all the analogs tested were stable after a
long exposure in human serum.
We also analysed the cytotoxicity of these analogs on human
fibroblast cell line and two breast cancer cell lines, MCF7 and
MDA-MB-231 (Figs. 4 and 5). Increasing concentrations (up to
and carboxylate series, the double bond in 9 leads to a partial
inhibitory effect on ligand binding compared with M6Pn-s.
We constructed the analogs based on the X-ray structure of the
M6P (Fig. 2A) liganded into the CI-M6PR (PDB entry 1SZ0). The
M6Pn-s displayed more hydrogen bonds with the essential resi-
dues in the binding pocket than the M6P counterpart. This better
stability of the ligand might explain the better affinity of the phos-
phonate for the receptor. The phosphonate binding site was com-
100 lM) of the compounds 3, 6, 9, M6Pn-s and M6S were incu-
bated for 4 days in culture medium. The growth of all cell lines
0 day serum
2 days 37ºC
2 days 37ºC + serum
140
120
100
80
b
60
40
20
0
M6P
5.10-4
M6Pn-s
5.10-4
(9)
5.10-4
M6S
2.10-4
(3)
2.10-4
(6)
10-4
Analogues (M)
Figure 3. Stability of M6P analogs in human serum. The competitive inhibition of
CI-M6PRb binding to PMP by M6P and its analogs was assayed in the presence or
absence of serum (0 day serum). These inhibitions were considered as controls
and expressed as 100% (grey). The inhibitions obtained after incubation of M6P and
M6P analogs for 2 days at 37 °C in the absence (white) or presence of 75% of human
serum (black) were compared to the respective controls. Values represent the
mean standard deviations of triplicates from a typical experiment and confirmed
in two additive experiments.
10-4
10-5
10-6
M
M
M
120
100
80
60
40
20
0
Control
M6P M6Pn-s
(9)
M6S
(3)
(6)
Analogues
Figure 4. Growth of human fibroblasts in presence of M6P or M6P analogs. Cells
were treated for 4 days in absence (control) or in presence of M6P or M6P analogs as
saturated or unsaturated M6-phosphonates or -sulfonates or with M6S. After
treatment, cell proliferation was measured by MTT assay as described in the
Figure 2. Ligand docking in the CI-M6PR ligand binding pocket. (A) Close-up view
of M6P taken as a reference.16 (B and C) Close-up views of the hydrogen bonds
between M6Pn-s (B) and 3 (C) and the residues Tyr 421, Gln 348, Tyr 324, Ser 386,
and Arg 391.
Supporting informations. The data represent
a typical experiment and were
confirmed in two additive experiments. All standard deviations from triplicate
were 65%.

A. Jeanjean et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6240–6243
6243
MCF7
MDA
References and notes
1. Dahms, N. M.; Hancock, M. K. Biochim. Biophys. Acta 2002, 1572, 317.
2. Ghosh, P.; Dahms, N. M.; Kornfeld, S. Nat. Rev. Mol. Cell Biol. 2003, 4, 202.
3. Hawkes, C.; Kar, S. Brain Res. Rev. 2004, 44, 117.
100
50
0
4. Kornfeld, S. Annu. Rev. Biochem. 1992, 61, 307.
5. Oka, Y.; Rozek, L. M.; Czech, M. P. J. Biol. Chem. 1985, 260, 9435.
6. Kang, J. X.; Li, Y.; Leaf, A. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13671.
7. Kornfeld, S.; Mellman, I. Annu. Rev. Cell Biol. 1989, 5, 483.
8. Wowk, M. E.; Trapani, J. A. Microbes Infect. 2004, 6, 752.
9. Brunetti, C. R.; Dingwell, K. S.; Wale, C.; Graham, F. L.; Johnson, D. C. J. Virol.
1998, 72, 3330.
10. Blanchard, F.; Raher, S.; Duplomb, L.; Vusio, P.; Pitard, V.; Taupin, J. L.; Moreau,
J. F.; Hoflack, B.; Minvielle, S.; Jacques, Y.; Godard, A. J. Biol. Chem. 1998, 273,
20886.
-4 -5 -6 -4 -5 -6 -4 -5 -6 -4 -5 -6 -4 -5 -6
M6Pn-s (9) M6S (3)
PBS
11. Metcalf, D. Stem Cells 2003, 21, 5.
(6)
12. Saris, J. J.; van den Eijnden, M. M.; Lamers, J. M.; Saxena, P. R.; Schalekamp, M.
A.; Danser, A. H. Hypertension 2002, 39, 573.
Analogue (log M)
13. Yang, L.; Tredget, E. E.; Ghahary, A. Wound Repair Regen. 2000, 8, 538.
14. Godar, S.; Horejsi, V.; Weidle, U. H.; Binder, B. R.; Hansmann, C.; Stockinger, H.
Eur. J. Immunol. 1999, 29, 1004.
15. Nykjaer, A.; Christensen, E. I.; Vorum, H.; Hager, H.; Petersen, C. M.; Roigaard,
H.; Min, H. Y.; Vilhardt, F.; Moller, L. B.; Kornfeld, S.; Gliemann, J. J. Cell Biol.
1998, 141, 815.
Figure 5. Growth of MCF7 or MDA-MB-231 human breast cancer cell lines in the
presence of M6P analogs. The growth was evaluated in the presence of M6P analogs
as described in Figure 3. The data represent
a typical experiment and were
confirmed in two additive experiments. All standard deviations from triplicate were
65%.
16. Olson, L. J.; Dahms, N. M.; Kim, J. J. J. Biol. Chem. 2004, 279, 34000.
17. Tong, P. Y.; Gregory, W.; Kornfeld, S. J. Biol. Chem. 1989, 264, 7962.
18. Gabel, C. A.; Costello, C. E.; Reinhold, V. N.; Kurz, L.; Kornfeld, S. J. Biol. Chem.
1984, 259, 13762.
19. Freeze, H. H. Arch. Biochem. Biophys. 1985, 243, 690.
20. Vidil, C.; Morère, A.; Garcia, M.; Barragan, V.; Hamdaoui, B.; Rochefort, H.;
Montero, J.-L. Eur. J. Org. Chem. 1999, 447.
was unaffected by these compounds indicating that their binding
to CI-M6PR, even for a long time period, is neither cytostatic nor
cytotoxic. So the replacement of phosphate by a phosphonate or
sulfonate group did not cause any toxicity.
In conclusion, chemical modifications on M6P or M6S induced a
gain in efficiency resulting in better binding affinity for CI-M6PR
and, importantly, greater stability of these compounds in human
serum. This study underlines the fact that CI-M6PR can bind a
broad range of M6P analogs, and the great diversity of ligands
bearing a M6P signal suggests the numerous potential therapeutic
applications offered by CI-M6PR. For example, it has already been
proposed that exogenous M6P could inhibit the activation of L-
TGF-b.36 This prevention of TGFb1 and TGFb2 activation at the
wound site results in markedly improved healing and less scar-
ring.37,38 This type of competitive inhibition could be applied to
other pathological processes that are mediated through CI-M6PR
interactions, such as the activation of complex structures in plas-
minogen activation, growth-regulating proteins (i.e., LIF, Granzyme
B) or viral proteins such as HSV glycoprotein D.
21. Berkowitz, D. B.; Maiti, G.; Charette, B. D.; Dreis, C. D.; MacDonald, R. G. Org.
Lett. 2004, 6, 4921.
22. Vidal, C.; Vidil, S.; Morère, A.; Garcia, M.; Montero, J.-L. Eur. J. Org. Chem. 2000,
3433.
23. Gary-Bobo, M.; Nirdé, P.; Jeanjean, A.; Morère, A.; Garcia, M. Curr. Med. Chem.
2007, 14, 2945.
24. Jeanjean, A.; Garcia, M.; Leydet, A.; Montero, J. L.; Morère, A. Bioorg. Med. Chem.
2006, 14, 3575.
25. Vidal, S.; Garcia, M.; Montero, J. L.; Morère, A. Bioorg. Med. Chem. 2002, 10,
4051.
26. Dennis, P. A.; Rifkin, D. B. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 580.
27. Boren, H. B.; Eklind, K.; Garegg, P. J.; Lindberg, B.; Pilotti, A. Acta Chem. Scand.
1972, 26, 4143.
28. Carretero, J. C.; Demillequand, M.; Ghosez, L. Tetrahedron 1987, 43, 5125.
29. Borbas, A.; Szabovik, G.; Antal, Z.; Herczegh, P.; Agocs, A.; Liptak, A. Tetrahedron
Lett. 1999, 40, 3639.
30. Roush, W. R.; Gwaltney, S. L.; Cheng, J.; Scheidt, K. A.; McKerrow, J. H.; Hansell,
E. J. Am. Chem. Soc. 1998, 120, 10994.
31. Vidil, C.; Vidal, S.; Morère, A.; Montero, J.-L. Phosphorus, Sulfur Silicon Relat.
Elem. 2000, 158, 125.
32. Mahrwald, R.; Theil, F.; Schick, H.; Schwarz, S.; Palme, H.-J.; Weber, G. J. Prakt.
Chem. 1986, 328, 777.
Acknowledgments
33. Ben-Bari, M.; Dewynter, G.; Aymard, C.; Jei, T.; Montero, J.-L. Phosphorus, Sulfur
Silicon Relat. Elem. 1995, 105, 129.
34. Engel, R. Chem. Rev. 1977, 77, 349–367.
35. Hancock, M. K.; Haskins, D. J.; Sun, G.; Dahms, N. M. J. Biol. Chem. 2002, 277,
11255.
36. Purchio, A. F.; Cooper, J. A.; Brunner, A. M.; Lioubin, M. N.; Gentry, L. E.;
Kovacina, K. S.; Roth, R. A.; Marquardt, H. J. Biol. Chem. 1988, 263, 14211.
37. Shah, M.; Foreman, D. M.; Ferguson, M. W. Lancet 1992, 339, 213.
38. Shah, M.; Foreman, D. M.; Ferguson, M. W. J. Cell Sci. 1994, 107, 1137.
We thank Michel Gleizes for technical assistance. We also wish
to thank the organisations ‘Vaincre les Maladies Lysosomales
(VML)’ for the grant assigned to Magali Gary-Bobo and ‘Association
pour la Recherche sur le Cancer (ARC)’ for financial support.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.09.101.