A multivalent ligand for the mannose-6-phosphate receptor for endolysosomal targeting of an activity-based probe

Article abstract of DOI:10.1002/anie.201406842

The ubiquitously expressed mannose-6-phosphate receptors (MPRs) are a promising class of receptors for targeted compound delivery into the endolysosomal compartments of a variety of cell types. The development of a synthetic multivalent mannose-6-phosphat

Full text of DOI:10.1002/anie.201406842

Angewandte
Chemie
DOI: 10.1002/anie.201406842
Receptor Targeting
A Multivalent Ligand for the Mannose-6-Phosphate Receptor for
Endolysosomal Targeting of an Activity-Based Probe**
Sascha Hoogendoorn, Gijs H. M. van Puijvelde, Johan Kuiper, Gijs A. van der Marel, and
Herman S. Overkleeft*
Abstract: The ubiquitously expressed mannose-6-phosphate
receptors (MPRs) are a promising class of receptors for
targeted compound delivery into the endolysosomal compart-
ments of a variety of cell types. The development of a synthetic,
multivalent, mannose-6-phosphate (M6P) glycopeptide-based
MPR ligand is described. The conjugation of this ligand to
fluorescent DCG-04, an activity-based probe for cysteine
cathepsins, enabled fluorescent readout of its receptor-targeting
properties. The resulting M6P-cluster–BODIPY–DCG-04
probe was shown to efficiently label cathepsins in cell lysates
as well as in live cells. Furthermore, the introduction of the 6-O-
phosphates leads to a completely altered uptake profile in COS
and dendritic cells compared to a mannose-containing ligand.
Competition with mannose-6-phosphate abolished all uptake
of the probe in COS cells, and we conclude that the mannose-6-
phosphate cluster targets the MPR and ensures the targeted
delivery of cargo bound to the cluster into the endolysosomal
pathway.
enzymes recombinantly expressed in CHO cells and mem-
brane-bound CI-MPRs.[3] The internalization of recombinant
enzyme bound to the CI-MPR ensures delivery into the
endolysosomal compartments of cells deficient in the lyso-
somal enzyme. Structural and biochemical studies have
revealed multiple M6P-binding domains in the CI-MPR[4]
and the formation of receptor dimers for both types of
MPR.[5] Furthermore, the CI-MPR has a relatively low
affinity for M6P (Kd = 7 mm) compared to glycoproteins that
contain multiple phosphorylated mannose residues (Kd = 2–
20 nm).[6] The notion that multivalency increases the affinity
of M6P-containing ligands towards the receptor has prompted
several groups to investigate synthetic ligands since these
enable the creation of defined structures for both the glycan
chain and the spacing between the phosphate residues.[7] We
previously reported a synthetic mannose cluster[8] that con-
tains six flexibly spaced mannose units, and which functions as
a ligand for mannose-binding lectins on dendritic cells.
Herein, we describe the synthesis and biological evaluation
of an analogous mannose-6-phosphate cluster (M6PC, 10;
Figure 1) with the aim of targeting the mannose-6-phosphate
receptor. The flexible nature of the cluster may enable
multivalent binding interactions, thereby increasing potency
and possibly receptor internalization.[9] To investigate the
MPR-targeting properties of this ligand, we conjugated the
cluster to fluorescent BODIPY–DCG-04, an activity-based
probe for cysteine cathepsins, a family of proteases that
mostly reside in the endolysosomal compartments.[10] Cathep-
sins play important roles in both health and disease[11] and
activity-based probes are powerful tools to visualize and
modulate their activity.[12] We demonstrate the ability of the
M6PC–BODIPY–DCG-04 probe (11; Figure 1) to label
cathepsins in cell and tissue lysates. We further report an
in situ analysis of the uptake and trafficking of 11 and its
inhibition of cathepsins in both COS cells, a fibroblast-like
cell line, and dendritic cells. We finally show that M6PC–
BODIPY–DCG-04 (11) differs in its targeting ability from its
nonphosphorylated counterpart.
T
he 300 kDa cation-independent mannose-6-phosphate
receptor (CI-MPR/IGFII-R) and its smaller 46 kDa homo-
logue, the cation-dependent mannose 6-phosphate receptor
(CD-MPR), are essential for the correct trafficking of newly
synthesized lysosomal enzymes from the trans-Golgi network
to the lysosomes. The receptor population is largely localized
intracellularly but a small (ca. 10%) fraction of the receptors
shuttles continuously between the cell membrane and the
endocytic pathway.[1] The endocytic nature of the receptor,
combined with its ubiquitous expression on fibroblast cells of
various tissues,[2] makes it a valid candidate for drug targeting.
Indeed, current enzyme replacement therapies for the
lysosomal storage disorders Fabry disease and Pompe disease
make use of the binding interaction between mannose-6-
phosphate (M6P) residues on the complex N-glycan chains of
[*] Dr. S. Hoogendoorn, Prof. Dr. G. A. van der Marel,
Prof. Dr. H. S. Overkleeft
Bio-organic Synthesis, Leiden Institute of Chemistry
Leiden University
Einsteinweg 55, 2300 RA, Leiden (The Netherlands)
E-mail: h.s.overkleeft@chem.leidenuniv.nl
Peptide scaffold 7 was used as a starting point for the
solid-phase peptide synthesis (SPPS) approach to mannose-6-
phosphate cluster 10 (Figure 1).[8] We envisaged that a triazole
moiety could be used to connect the sugar to the peptide
backbone since it has been shown that the aglycone part of the
mannoside does not interfere with receptor binding.[7c] For
this purpose, a propargyl mannose-6-phosphate building
block was designed and synthesized that contains acid-labile
tert-butyl protective groups on the phosphate group and base-
labile benzoyl groups on the hydroxy groups (compound 6,
Figure 1, and Scheme S1 in the Supporting Information),
Dr. G. H. M. van Puijvelde, Prof. Dr. J. Kuiper
Dept of Biopharmaceutics, LACDR, Leiden (The Netherlands)
[**] We thank H. van den Elst for assisting with HPLC–MS purifications
and C. Erkelens, F. Lefeber, and K. B. S. S. Gupta for NMR
measurements. This research was funded by the Netherlands
Organization for Scientific Research (NWO-CW) and the European
Research Council.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201406842.
Angew. Chem. Int. Ed. 2014, 53, 10975 –10978
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10975

.
Angewandte
Communications
geneous salt form of 10 as seen by a sharp
phosphate peak in the NMR spectrum. How-
ever, the final compound M6PC–BODIPY–
DCG-04 (11) was only successfully synthesized
+
by using the heterogeneous NH4 phosphate
cluster because of solubility issues with either
the starting material or the product. The use of
the more reactive propiolic acid handle instead
of the previously employed pentynoic acid[8,14]
resulted in a drastic improvement of the click
reaction (1 h, room temperature) between the
mannose-6-phosphate
cluster
10
and
BODIPY(TMR)–DCG-04 (12; Figure S1 in
the Supporting Information). Subsequent pu-
rification by HPLC–MS under NH4OAc con-
ditions provided the (mostly) ammonium salt
form of probe 11, which was used without
further cation-exchange procedures.
As shown in Figure 2, we first determined
the ability of compound 11 to function as an
activity-based probe for cathepsins by per-
forming lysate labeling experiments. For this,
increasing concentrations of the probe were
added to lysates of different origins (mouse
liver, mouse dendritic cells, or monkey COS
cells) at pH 5.5, the optimal pH for most of the
cathepsin activities.[10] Concentration-depen-
dent labeling of active cathepsins was observed
in each lysate, with a shift in molecular weight
corresponding to the mass of the probe
(Figure 2, compare lane 5 (1 mm 11) to lane 11
(1 mm 14, BODIPY(FL)-DCG-04; Figure S1).
Inactivation of the cathepsins by preincubation
with 1 or 10 mm of the known cathepsin
inhibitors AS44[8] or various DCG-04 deriva-
Figure 1. Design and synthesis of the fluorescent MPR-targeted activity-based cathepsin
probe 11. Reagents and conditions: a) Mannose-6-phosphate (6), sodium ascorbate,
CuSO4, CH2Cl2/H2O, RT, 72 h; b) 1. 20% piperidine/NMP, RT, 15 min; 2. propiolic acid,
EEDQ, CH2Cl2, RT, 1.5 h; c) TFA/TIS (98/2, v/v), 3ꢀ15 min; d) 1. K2CO3, MeOH, RT,
24 h; 2. Citric acid (aq); 3. RP-HPLC. R=NH4+ (used in the next step) or R=Na+ (after
Chelex-Na+ ion exchange); e) BODIPY(TMR)–DCG-04 (12), sodium ascorbate, CuSO4,
DMF/H2O (26%). NMP=N-methyl-2-pyrrolidinone, EEDQ=2-ethoxy-1-ethoxycarbonyl-
1,2-dihydroxyquinoline, DMF=dimethylformamide.
protecting groups that are compatible with SPPS conditions.
Protected propargyl mannose-6-phosphate 6 was attached to
the six azide groups in peptide 7 on solid support by CuI-
catalyzed Huisgen 1,3-cycloaddition.[13] Next, the Fmoc group
was removed and propiolic acid was coupled to the primary
amine with the use of EEDQ. The peptide was cleaved from
the resin by using acidic conditions, which also ensured
complete deprotection of the phosphate groups. We consider
this protecting-group strategy to be a considerable improve-
ment compared to previously reported methods towards M6P
derivatives. These rely on either 2,2,2-trichloroethyl protec-
tion of the phosphate group, which requires harsh solution-
phase conditions for cleavage,[7a] or the use of unprotected
phosphate, which results in electronic repulsion and coupling
difficulties.[7b] Global deprotection was accomplished by using
Figure 2. Lysate cathepsin labeling. Various cell lysates (10–20 mg total
protein) were incubated (1 h, 378C) with increasing concentrations of
probe 11 at pH 5.5. Alternatively, lysates were incubated (1 h, 378C)
with azido-DCG-04 (13, 1 or 10 mm), AS44 (15, 10 mm) or BODI-
PY(FL)–DCG-04 (14, 1 or 10 mm), before treatment with 11 (1 mm, 1 h,
378C). Proteins were resolved by 12.5% SDS-PAGE, followed by
fluorescence scanning [Cy2: BODIPY(FL), Cy3: BODIPY(TMR)] and
total protein staining with coomassie brilliant blue (CBB).
a saturated solution of K2CO3 in methanol, followed by
quenching with citric acid. The glycopeptide was purified
under neutral NH4OAc conditions and lyophilized to give the
ammonium salt of compound 10. 31P NMR analysis showed
a very broad peak, which is indicative of multiple salt forms of
the peptide. The peptide was ion-exchanged to the sodium
form by using Chelex-Na+ resin, thereby resulting in a homo-
10976 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 10975 –10978

Angewandte
Chemie
tives[8,15] (for structures, see Figure S1) resulted in loss of
labeling with probe 11, thus showing that it competes for the
same set of cathepsins.
Having established that attachment of the mannose-6-
phosphate cluster does not interfere with cathepsin profiling,
we compared the new probe head-to-head with the previously
reported mannose-cluster–BODIPY–DCG-04 probe (16;[8]
Figure S1). To elucidate whether our probe is taken up by
MPR-mediated endocytosis, we used COS cells, an African
green monkey fibroblast-like cell line.
Cells were incubated for 16 h with 0.5 mm of the MPR-
targeted probe 11, nontargeted fluorescent BODIPY(TMR)-
DCG-04 (12), or the mannose receptor (MR)-targeted probe
16 and subsequently analyzed by flow cytometry. As shown in
Figure 3A, only cells that were treated with probes 11 or 12
showed a shift in the population fluorescence indicative of
probe binding/uptake. Cellular fluorescence was time-depen-
dent in case of probe 11 (50% positive after 6 h compared to
74% after 16 h of incubation) and was almost completely
inhibited in the presence of 1 mm M6P. COS cells treated for
6 h with 0.5 mm of probe 11 and imaged by confocal
fluorescence microscopy showed bright intracellular fluores-
cent vesicles, thus establishing that the compound was taken
up into the endocytic pathway (Figure 3B). Next, as addi-
tional proof for MPR-dependent uptake and trafficking of the
probe into the lysosomes, we determined the cathepsin
labeling. Cells were treated under various conditions,
washed, lysed, and analyzed by 12.5% SDS-PAGE followed
by in-gel fluorescence scanning. In agreement with the flow
cytometry data, no labeling was observed for mannose-
cluster–BODIPY–DCG-04 (16), a result in accordance with
the selectivity of this probe for mannose-binding lectins. The
M6P probe 11, on the other hand, showed both concentration-
and time-dependent labeling of cathepsins (Figure 3C,D). As
shown in Figure 3E, cathepsin labeling was inhibited when
using increasing concentrations of M6P, with almost complete
inhibition at 1 mm M6P. Even when taking into account that
the probe contains six M6P residues, a large molar excess of
monovalent M6P is needed to inhibit its uptake. Somewhat
surprisingly, pretreatment of the cells with nonfluorescent
azido-DCG-04 (13) did not compete with cathepsin labeling
by the targeted probe. To clarify this, we used green
fluorescent nontargeted BODIPY(FL)-DCG-04 (14) in
COS cathepsin labeling studies and indeed observed only
background bands and not the expected labeling as seen in
lysates (Figure 2 and Figure S2). Presumably, diffusion of the
nontargeted probes into the endolysosomal compartments of
COS cells is ineffective. Together, these findings highlight the
efficiency with which the targeted probe 11 is taken up by
MPR-mediated endocytosis into COS cells and trafficked to
the endolysosomal compartments, thereby leading to the
inactivation of cathepsins.
In a next set of experiments, we treated live immature
mouse dendritic cells (DCs) with varying concentrations of
either probe. This resulted in similar concentration-depen-
dent labeling profiles, which are completely absent after
preincubation with unlabeled azido-DCG-04 (13; Fig-
ure S3).[16] Confocal fluorescence microscopy and SDS-
PAGE analysis confirmed that blockage of the mannose-
binding lectins present on DCs with the yeast oligomannoside
mannan (3 mgmLÀ1) results in inhibition of the uptake of the
mannose-cluster probe 16.[8,14] By contrast, the uptake and
Figure 3. M6PC–BDP–DCG-04 (11) uptake and cathepsin labeling in live COS cells. a) Flow cytometry analysis of COS cells that were treated with
0.5 mmM6PC–BDP–DCG-04 (11, 6 h or 16 h), MC–BDP–DCG-04 (16, 16 h), the nontargeted probe BODIPY(TMR)–DCG-04 (12, 16 h), or first with
1 mmM6P (16 h) followed by treatment with probe 11 for 6 h. Mean percentages of cells in the negative (left) and positive (right) populations
from three independent experiments are given. b) Representative confocal microscopy image of live COS cells treated with probe 11 (0.5 mm) for
6 h, showing fluorescently labeled intracellular compartments. Scale bar: 7.5 mm. c,d) COS cells were treated with increasing concentrations of
probe 11 or 16 for 16 h (c) or with a 0.5 mm concentration of 11 for different periods of time (d) before being washed, lysed, and analyzed by
12.5% SDS-PAGE. Representative fluorescence scans (Cy3) and total protein stains (CBB) are depicted. e) Competition experiments between
different concentrations of M6P or azido-DCG-04 (13; 16 h, 378C) and 11 (0.5 mm, 6 h, 378C). After treatment, cells were washed with PBS, lysed,
and resolved on a 12.5% SDS-PAGE gel. BDP=BODIPY, M6P(C)= mannose-6-phosphate (cluster), MC=mannose cluster, PBS=phosphate-
buffered saline.
Angew. Chem. Int. Ed. 2014, 53, 10975 –10978
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10977

.
Angewandte
Communications
[5] a) D. L. Roberts, D. J. Weix, N. M. Dahms, J.-J. P. Kim, Cell 1998,
93, 639 – 648; b) L. J. Olson, J. Zhang, N. M. Dahms, J.-J. P. Kim,
J. Biol. Chem. 2002, 277, 10156 – 10161; c) J. C. Byrd, J. H. Park,
B. S. Schaffer, F. Garmroudi, R. G. MacDonald, J. Biol. Chem.
2000, 275, 18647 – 18656.
cathepsin binding of phosphorylated probe 11 is not inhibited
by mannan, thus indicating that the phosphorylated mannose
residues are unable to bind to the MR and that the uptake of
this probe occurs through other means. An attempt at uptake
inhibition through the addition of 1 mm of M6P to the
medium did not succeed. Since DCs are highly endocytoti-
cally active, other processes besides receptor-mediated endo-
cytosis (such as macropinocytosis) may be responsible for
probe uptake in these cells.[17]
In conclusion, we have shown that peptide-based target-
ing devices bearing 6-O-phosphorylated mannose units have
completely altered targeting properties compared to non-
phosphorylated mannose-bearing peptide conjugates. Since
the MPR is widely expressed on a variety of cells and tissues,
we envisage that our artificial M6P cluster may provide an
effective means of broad targeted delivery. Cathepsins are
efficiently inhibited by the targeted probe, a result that has
wider implications given that these enzymes are associated
with various diseases.[11b] The ability to monitor and perturb
cathepsin activities in a variety of cell types or tissues,
depending on the choice of ligand, could thus be greatly
advantageous. The conjugation of our M6P cluster to other
biological entities, such as recombinant lysosomal enzymes,
may thus be a suitable alternative for their targeted delivery
into cells affected by lysosomal storage disorders such as
Pompe or Fabry disease.
[6] P. Y. Tong, W. Gregory, S. Kornfeld, J. Biol. Chem. 1989, 264,
7962 – 7969.
[7] a) M. K. Christensen, M. Meldal, K. Bock, H. Cordes, S.
Mouritsen, H. Elsner, J. Chem. Soc. Perkin Trans. 1 1994,
1299 – 1310; b) H. Franzyk, M. K. Christensen, R. M. Jorgensen,
M. Meldal, H. Cordes, S. Mouritsen, K. Bock, Bioorg. Med.
Chem. 1997, 5, 21 – 40; c) J. J. Distler, J. Guo, G. W. Jourdian,
O. P. Srivastava, O. Hindsgaul, J. Biol. Chem. 1991, 266, 21687 –
21692.
[8] U. Hillaert, M. Verdoes, B. I. Florea, A. Saragliadis, K. L. L.
Habets, J. Kuiper, S. Van Calenbergh, F. Ossendorp, G. A.
van der Marel, C. Driessen, H. S. Overkleeft, Angew. Chem.
Int. Ed. 2009, 48, 1629 – 1632; Angew. Chem. 2009, 121, 1657 –
1660.
[9] S. J. York, L. S. Arneson, W. T. Gregory, N. M. Dahms, S.
Kornfeld, J. Biol. Chem. 1999, 274, 1164 – 1171.
[10] K. Brix, A. Dunkhorst, K. Mayer, S. Jordans, Biochimie 2008, 90,
194 – 207.
[11] a) M. M. Mohamed, B. F. Sloane, Nat. Rev. Cancer 2006, 6, 764 –
775; b) J. Reiser, B. Adair, T. Reinheckel, J. Clin. Invest. 2010,
120, 3421 – 3431.
[12] a) M. Verdoes, K. Oresic Bender, E. Segal, W. A. van der Lin-
den, S. Syed, N. P. Withana, L. E. Sanman, M. Bogyo, J. Am.
Chem. Soc. 2013, 135, 14726 – 14730; b) S. D. Wiedner, L. N.
Anderson, N. C. Sadler, W. B. Chrisler, V. K. Kodali, R. D.
Smith, A. T. Wright, Angew. Chem. Int. Ed. 2014, 53, 2919 –
2922; Angew. Chem. 2014, 126, 2963 – 2966; c) J. A. Joyce, A.
Baruch, K. Chehade, N. Meyer-Morse, E. Giraudo, F.-Y. Tsai,
D. C. Greenbaum, J. H. Hager, M. Bogyo, D. Hanahan, Cancer
Cell 2004, 5, 443 – 453.
[13] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002,
67, 3057 – 3064; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin,
K. B. Sharpless, Angew. Chem. Int. Ed. 2002, 41, 2596 – 2599;
Angew. Chem. 2002, 114, 2708 – 2711.
Received: July 3, 2014
Published online: August 22, 2014
Keywords: activity-based protein profiling · drug delivery ·
.
endolysosomal pathway · fluorescent probes ·
mannose-6-phosphate
[14] S. Hoogendoorn, K. L. Habets, S. Passemard, J. Kuiper, G. A.
van der Marel, B. I. Florea, H. S. Overkleeft, Chem. Commun.
2011, 47, 9363 – 9365.
[1] P. Ghosh, N. M. Dahms, S. Kornfeld, Nat. Rev. Mol. Cell Biol.
2003, 4, 202 – 213.
[2] M. R. Natowicz, M. M.-Y. Chi, O. H. Lowry, W. S. Sly, Proc. Natl.
Acad. Sci. USA 1979, 76, 4322 – 4326.
[15] D. Greenbaum, K. F. Medzihradszky, A. Burlingame, M. Bogyo,
Chem. Biol. 2000, 7, 569 – 581.
[3] a) Y. A. Ioannou, D. F. Bishop, R. J. Desnick, J. Cell Biol. 1992,
119, 1137 – 1150; b) J. van Hove, H. W. Yang, J.-Y. Wu, R. O.
Brady, Y.-T. Chen, Proc. Natl. Acad. Sci. USA 1996, 93, 65 – 70.
[4] M. K. Hancock, D. J. Haskins, G. Sun, N. M. Dahms, J. Biol.
Chem. 2002, 277, 11255 – 11264.
[16] D. Greenbaum, A. Baruch, L. Hayrapetian, Z. Darula, A.
Burlingame, K. F. Medzihradszky, M. Bogyo, Mol. Cell. Proteo-
mics 2002, 1, 60 – 68.
[17] F. Sallusto, M. Cella, C. Danieli, A. Lanzavecchia, J. Exp. Med.
1995, 182, 389 – 400.
10978 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 10975 –10978