A selective inhibitor Gal-PUGNAc of human lysosomal β-hexosaminidases modulates levels of the ganglioside GM2 in neuroblastoma cells

Article abstract of DOI:10.1002/anie.200804583

β-hexosaminidases HEXA and HEXB is cell-permeable and modulates the acitivity of HEXA and HEXB in tissue culture, increasing ganglioside GM2 levels. Gal-PUGNAc should allow the role of these enzymes to be studied at the cellular level without generating a

Full text of DOI:10.1002/anie.200804583

Communications
DOI: 10.1002/anie.200804583
Enzyme Inhibitors
A Selective Inhibitor Gal-PUGNAc of Human Lysosomal
b-Hexosaminidases Modulates Levels of the Ganglioside GM2 in
Neuroblastoma Cells**
Keith A. Stubbs,* Matthew S. Macauley, and David J. Vocadlo*
Imbalances in the levels of many glycoconjugates have been
implicated in human diseases. Many inhibitors of the large
super family of glycoside hydrolases, which are the enzymes
that break down these glycoconjugates, are known. However,
few of these inhibitors have been developed and character-
ized as being able to selectively target the enzyme of interest
over other functionally related glycoside hydrolases. Such
highly selective inhibitors are critical tools to gain insights
into the roles of specific glycoconjugates in disease etiology
and normal cellular function.
In humans, there are three enzymes that cleave terminal
b-linked N-acetylglucosamine residues from glycoconjugates.
The two dimeric isozymes, hexosaminidase A and B (HEXA
and HEXB),[1] share high sequence identity and are members
of the family 20 glycoside hydrolases (GH20). HEXA and
HEXB are localized in the lysosome, where they hydrolyze
terminal N-acetylglucosamine and N-acetylgalactosamine
residues of glycosphingolipids such as the ganglioside GM2
Figure 1. a) Structure of ganglioside GM2 and b) O-GlcNAc, and c) the
compounds used herein.
(Figure 1a), which contains a tetrasaccharide terminating at
its non-reducing end with an N-acetylgalactosamine residue.[2]
Inheritable mutations impairing either enzyme result in the
accumulation of GM2 in the brain, causing the neurode-
generative disorders known as Tay-Sachs and Sandhoff
diseases.[3] Deleterious effects arising from the accumulation
of gangliosides are still being uncovered.[4] Interest in the
biological roles played by gangliosides have increased as their
roles in the modulation of cellular signaling have become
apparent.[5,6]
The other human glycoside hydrolase that cleaves termi-
nal b-linked N-acetylglucosamine residues is a family 84
enzyme termed O-GlcNAcase (OGA). This enzyme acts on
post-translationally modified nucleocytoplasmic glycopro-
teins that bear 2-acetamido-2-deoxy-b-d-glucopyranoside
residues linked to serine or threonine (O-GlcNAc, Fig-
ure 1b).[7,8] The consequences of impairing O-GlcNAcase in
mammals are unknown; however, this enzyme has been
implicated in Alzheimer disease and Type II Diabetes.[8]
Although O-GlcNAcase and the lysosomal b-hexosami-
nidases act on different substrates and are located in different
cellular compartments, both of these enzymes use a catalytic
mechanism involving nucleophilic participation of the 2-
acetamido group of the substrate.[9–11] This commonality of
the catalytic mechanism delayed the development of selective
O-GlcNAcase inhibitors.[11] However, various inhibitors of O-
GlcNAcase[12–15] are proving to be critical tools for evaluating
the roles played by O-GlcNAc.[16] In contrast, no inhibitor has
yet been described that inhibits b-hexosaminidases in a
cellular setting to increase levels of GM2. Such a small
molecule probe would however be a powerful tool for
defining the roles played by gangliosides such as GM2 or
for creating readily titratable models of certain disease states.
[*] Dr. K. A. Stubbs,[+] M. S. Macauley, Prof. D. J. Vocadlo
Department of Chemistry, Simon Fraser University
8888 University Drive, Burnaby, British Columbia V5A 1S6 (Canada)
Fax: (+1)604-291-3765
E-mail: dvocadlo@sfu.ca
[+] Current address: Chemistry M313, School of Biomedical, Biomo-
lecular and Chemical Sciences, University of Western Australia
35 Stirling Highway, Crawley, Western Australia 6009 (Australia)
Fax: (+618)6488-1005
E-mail: kstubbs@cyllene.uwa.edu.au
[**] K.A.S. and M.S.M. contributed equally to this work. We thank Don
Mahuran and Michael Tropak for their generous gift of human
HEXA and HEXB, Rosemary Cornell for the HPTLC plates, and Cara-
Lynne Schengrund for useful discussions on the use of SK-N-SH
cells and HPTLC of gangliosides. This work was supported by a
grant from the Natural Sciences and Engineering Research Council
of Canada (NSERC). D.J.V. is a scholar of the Michael Smith
Foundation of Health Research (MSFHR) and a Tier II Canada
Research Chair in Chemical Glycobiology. K.A.S. is supported by a
Discovery project grant from the Australian Research Council.
M.S.M. is a recipient of a senior scholarship from the NSERC and
MSFHR.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804583.
1300
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1300 –1303

Angewandte
Chemie
Despite a number of inhibitors having been generated for
these three enzymes (HEXA, HEXB, and O-GlcNAcase), the
number of synthetic steps and overall yield often limit their
use in biological studies, as large quantities of these inhibitors
are often required. Furthermore, most of these inhibitors
concomitantly inhibit both family 20 (GH 20) and family 84
(GH 84) enzymes and are thus less useful for studying the role
of one particular enzyme in a biological setting as they may
elicit a complex phenotype. Accordingly, there is a need for
potent inhibitors that are selective for lysosomal b-hexos-
aminidases over O-GlcNAcase, and are able to elevate levels
of ganglioside GM2 in cells.
One inhibitor of all three enzymes is O-(2-acetamido-2-
deoxy-d-glucopyranosylidene)-amino-N-phenylcarbamate
PUGNAc.[17] Despite its benefits, one limitation of PUGNAc
is its indiscriminate inhibition of both family 20 lysosomal b-
hexosaminidases[11,18] and family 84 O-GlcNAcase.[11,19] Given
that HEXA and HEXB can process both gluco- and galacto-
configured hexosaminides,[1] while O-GlcNAcase is selective
for gluco-configured substrates,[19] a galacto-configured C-4
epimer of PUGNAc appeared to be a simple route to ensure
selectivity by exploiting these differences in substrate specif-
icities. We therefore synthesized Galactose-PUGNAc 1 (Gal-
PUGNAc), and investigated its activity toward human b-
hexosaminidase in enzyme assays and in cultured human
neuronal cells.
PUGNAc 1 is therefore comparable to the most potent
inhibitors of family 20 b-hexosaminidases that have Ki values
in the low nanomolar range (see Supporting Informa-
tion).[11,20,21]
We evaluated the selectivity of Gal-PUGNAc 1 and found
that at a concentration 1 mm, Gal-PUGNAc 1 showed less
than 1% inhibition of O-GlcNAcase, which is consistent with
the inability of O-GlcNAcase to process galactose-configured
saccharides.[19] The Ki value of Gal-PUGNAc 1 against O-
GlcNAcase must therefore be greater than 10 mm, making it
the most selective inhibitor for lysosomal b-hexosaminidases
over O-GlcNAcase (over 500000-fold; see Supporting Infor-
mation). a-N-Acetylgalactosaminidases are functionally
related to b-hexosaminidases in that they also process
terminal N-acetylgalactosamine residues, albeit only those
having an a-glycosidic linkage. Therefore, to further inves-
tigate the selectivity of Gal-PUGNAc 1 we evaluated its
potency towards the G. gallus GH 27 a-N-acetylgalactosami-
nidase, which shares 50% identity with the human homologue
(NAGA) including completely conserved active site resi-
dues[22] and determined a Ki value of 140 Æ 10 mm. Gal-
PUGNAc 1 is therefore 2800-fold more selective for b-
hexosaminidase over the GH 27 a-N-acetylgalactosamini-
dase. Gal-PUGNAc 1 also binds over 40 times more tightly
than another galactose-configured inhibitor of lysosomal b-
hexosaminidases, Gal-NAG-thiazoline (Figure 1c), which
was reported to have a Ki of 820 nm against HEXA.[23] Gal-
NAG-thiazoline, however, has not been tested against O-
GlcNAcase nor used in cell culture. We therefore prepared
this compound and found that, like Gal-PUGNAc 1, it is also
a poor inhibitor of O-GlcNAcase.
Having demonstrated the selectivity of Gal-PUGNAc 1
and Gal-NAG-thiazoline for the lysosomal b-hexosamini-
dases in vitro, we evaluated these inhibitors in cultured SK-N-
SH cells. These cells are a well-characterized human neuro-
blastoma cell line that has appreciable levels of GM2.[24] Cells
were treated with a panel of inhibitors of HEXA, HEXB, and
O-GlcNAcase; this panel was comprised of Gal-PUGNAc 1,
PUGNAc,[12,17,25] NAG-thiazoline,[9,11] Gal-NAG-thiazo-
line,[23] and NButGT.[11] After nine days in the presence of
inhibitors, glycosphingolipids were extracted from the cells
and ganglioside levels were assessed by high-performance
thin-layer chromatography using resorcinol staining. Levels
of GM2 were significantly elevated (30–40%) in cells treated
with Gal-PUGNAc 1, PUGNAc, NAG-thiazoline, and Gal-
NAG-thiazoline (Figure 3). Notably, NButGT, the O-GlcNA-
case specific inhibitor,[11] did not elevate cellular GM2 levels
(Figure 3d). This result is consistent with NButGT being a
poor inhibitor of HEXB (Ki = 340 mm)[11] and highlights the
selectivity of this inhibitor towards O-GlcNAcase.
Starting from galactosamine hydrochloride, Gal-
PUGNAc 1 was prepared in eight steps, with an overall
yield of nearly 20% (see Supporting Information), using a
route modified from that previously described for synthesis of
PUGNAc.[12] Gal-PUGNAc 1 was found to be a potent and
competitive inhibitor of HEXA and HEXB, with an inhib-
ition constant (Ki) of 18 Æ 1 nm against HEXB (Figure 2) and
51 Æ 3 nm against HEXA (see Supporting Information). Gal-
To further investigate the value and specificity of these
inhibitors we examined their effect on cellular O-GlcNAc
levels. We treated SK-N-SH cells overnight with the panel of
inhibitors used above and analyzed cell lysates for levels of O-
GlcNAc by Western blot. As expected, known O-GlcNAcase
inhibitors used previously in cells, including PUGNAc, NAG-
thiazoline, and NButGT, all significantly increased cellular O-
GlcNAc levels. In contrast, neither Gal-PUGNAc 1 nor Gal-
NAG-thiazoline increased cellular O-GlcNAc levels
Figure 2. Inhibition of human HEXB catalyzed hydrolysis of pNP-
GlcNAc by Gal-PUGNAc 1 shows a pattern of competitive inhibition.
~
The concentrations of Gal-PUGNAc 1 (mm) used were 0.15 ( ), 0.10
&
^
*
*
( ), 0.050 ( ), 0.020 (&), 0.010 ( ), and 0.00 units( ). Inset: graph-
ical analysis of Ki from plotting KM apparent against the concentration
of Gal-PUGNAc 1. Km,app is the apparent Km value obtained at each
inhibitor concentration.
Angew. Chem. Int. Ed. 2009, 48, 1300 –1303
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1301

Communications
Figure 4. Gal-PUGNAc 1 and Gal-NAG-thiazoline do not increase
levels O-GlcNAc-modified proteins. SK-N-SH cells were treated over-
night with 50 mm of each of the compounds and analyzed for levels of
O-GlcNAc-modified proteins by Western blot (upper panel). Western
blot of cellular b-actin levels demonstrates equal loading for each lane
(lower panel).
potency of Gal-PUGNAc 1 over Gal-NAG-thiazoline, Gal-
PUGNAc 1 offers advantages for studying lysosomal b-
hexosaminidase inhibition and ganglioside GM2 in a cellular
context without generating a complex chemical phenotype
stemming from concomitant inhibition of O-GlcNAcase. We
expect these compounds will prove to be valuable tools for
researchers interested in the roles of GM2 in cells and in vivo.
Figure 3. Small-molecule inhibitors of HEXA and HEXB increase levels
of GM2 in SK-N-SH cells. HPTLC analysis of lipid extracts of cells
treated with 200 mm of each compound for nine days (left panels). In
each experiment, control cells were grown parallel to treated cells.
Magnification of the band representing GM2 is shown on the right
panels along with densitometric analyses of the intensities of the GM2
bands.
Received: September 17, 2008
Published online: January 7, 2009
(Figure 4) highlighting their selectivity for HEXA and
HEXB.
Keywords: carbohydrates · gangliosides · glycosidases ·
One important observation made here for cell biologists
using non-selective O-GlcNAcase inhibitors to uncover the
biological roles of O-GlcNAc is that PUGNAc and NAG-
thiazoline perturb both O-GlcNAc and ganglioside levels. A
further cautionary note is that PUGNAc has also been
recently shown to inhibit an a-N-acetylglucosamindase bac-
terial homologue from GH 89 that has high active site
similarity with human NAGLU.[26] Therefore, results obtained
using PUGNAc in cells should be interpreted with these off-
target effects in mind, as levels of multiple glycoconjugates
will be perturbed.
.
inhibitors
[1] D. Mahuran, R. A. Gravel, Adv. Genet. 2001, 44, 145 – 163.
[2] E. Klenk, Hoppe Seylers Z. Physiol. Chem. 1942, 273, 76 – 86.
[3] B. Triggs-Raine, D. J. Mahuran, R. A. Gravel, Adv. Genet. 2001,
44, 199 – 224.
[4] D. Zhou, J. Mattner, C. Cantu III, N. Schrantz, N. Yin, Y. Gao, Y.
Sagiv, K. Hudspeth, Y. P. Wu, T. Yamashita, S. Teneberg, D.
Wang, R. L. Proia, S. B. Levery, P. B. Savage, L. Teyton, A.
Bendelac, Science 2004, 306, 1786 – 1789.
In summary, Gal-PUGNAc 1 is a readily accessed and
selective inhibitor of the human lysosomal b-hexosamini-
dases. This selectivity is manifested in the ability of Gal-
PUGNAc 1 to increase GM2 ganglioside levels in cultured
cells, but to not affect O-GlcNAc levels. We find that Gal-
NAG-thiazoline affords similar control over GM2 levels and
expect other b-hexosaminidase inhibitors that permeate into
cells, including those that are non-carbohydrate-based,[27] to
act similarly. Owing to the greater stability and 40-fold higher
[5] A. R. Todeschini, S. Hakomori, Biochim. Biophys. Acta 2008,
1780, 421 – 433.
[6] A. Sasaki, K. Hata, S. Suzuki, M. Sawada, T. Wada, K.
Yamaguchi, M. Obinata, H. Tateno, H. Suzuki, T. Miyagi, J.
Biol. Chem. 2003, 278, 27896 – 27902.
[7] C. R. Torres, G. W. Hart, J. Biol. Chem. 1984, 259, 3308 – 3317.
[8] G. W. Hart, M. P. Housley, C. Slawson, Nature 2007, 446, 1017 –
1022.
[9] S. Knapp, D. Vocadlo, Z. N. Gao, B. Kirk, J. P. Lou, S. G. Withers,
J. Am. Chem. Soc. 1996, 118, 6804 – 6805.
1302
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1300 –1303

Angewandte
Chemie
[10] I. Tews, A. Perrakis, A. Oppenheim, Z. Dauter, K. S. Wilson,
C. E. Vorgias, Nat. Struct. Biol. 1996, 3, 638 – 648.
[18] M. Horsch, L. Hoesch, A. Vasella, D. M. Rast, Eur. J. Biochem.
1991, 197, 815 – 818.
[11] M. S. Macauley, G. E. Whitworth, A. W. Debowski, D. Chin,
D. J. Vocadlo, J. Biol. Chem. 2005, 280, 25313 – 25322.
[12] K. A. Stubbs, N. Zhang, D. J. Vocadlo, Org. Biomol. Chem. 2006,
4, 839 – 845.
[13] H. C. Dorfmueller, V. S. Borodkin, M. Schimpl, S. M. Shepherd,
N. A. Shpiro, D. M. F. van Aalten, J. Am. Chem. Soc. 2006, 128,
16484 – 16485.
[19] D. L. Dong, G. W. Hart, J. Biol. Chem. 1994, 269, 19321 – 19330.
[20] M. Terinek, A. Vasella, Helv. Chim. Acta 2005, 88, 10 – 22.
[21] E. Kappes, G. Legler, J. Carbohydr. Chem. 1989, 8, 371 – 388.
[22] S. C. Garman, L. Hannick, A. Zhu, D. N. Garboczi, Structure
2002, 10, 425 – 434.
[23] B. Amorelli, C. Yang, B. Rempel, S. G. Withers, S. Knapp,
Bioorg. Med. Chem. Lett. 2008, 18, 2944 – 2947.
[14] S. Knapp, M. Abdo, K. Ajayi, R. A. Huhn, T. J. Emge, E. J. Kim,
J. A. Hanover, Org. Lett. 2007, 9, 2321 – 2324.
[24] B. C. Yowler, R. D. Kensinger, C. L. Schengrund, J. Biol. Chem.
2002, 277, 32815 – 32819.
[15] S. A. Yuzwa, M. S. Macauley, J. E. Henioen, X. Shan, R. J.
Dennis, Y. He, G. E. Whitworth, K. A. Stubbs, E. J. McEachern,
G. J. Davies, D. J. Vocadlo, Nat. Chem. Biol. 2008, 4, 483 – 490.
[16] M. S. Macauley, A. K. Bubb, C. Martinez-Fleites, G. J. Davies,
D. J. Vocadlo, J. Biol. Chem. 2008, 283, 34687 – 34695.
[17] D. Beer, J. L. Maloisel, D. M. Rast, A. Vasella, Helv. Chim. Acta
1990, 73, 1918 – 1922.
[25] H. Mohan, A. Vasella, Helv. Chim. Acta 2000, 83, 114 – 118.
[26] E. Ficko-Blean, K. A. Stubbs, O. Nemirovsky, D. J. Vocadlo,
A. B. Boraston, Proc. Natl. Acad. Sci. USA 2008, 105, 6560 –
6565.
[27] M. B. Tropak, J. E. Blanchard, S. G. Withers, E. D. Brown, D.
Mahuran, Chem. Biol. 2007, 14, 153 – 164.
Angew. Chem. Int. Ed. 2009, 48, 1300 –1303
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1303