Carbohydrate Bis-acetal-Based Substrates as Tunable Fluorescence-Quenched Probes for Monitoring exo-Glycosidase Activity

Article abstract of DOI:10.1021/jacs.7b01948

Tunable F?rster resonance energy transfer (FRET)-quenched substrates are useful for monitoring the activity of various enzymes within their relevant physiological environments. Development of FRET-quenched substrates for exo-glycosidases, however, has been hindered by their constrained pocket-shaped active sites. Here we report the design of a new class of substrate that overcomes this problem. These Bis-Acetal-Based Substrates (BABS) bear a hemiacetal aglycon leaving group that tethers fluorochromes in close proximity, also positioning them distant from the active site pocket. Following cleavage of the glycosidic bond, the liberated hemiacetal spontaneously breaks down, leading to separation of the fluorophore and quencher. We detail the synthesis and characterization of GlcNAc-BABS, revealing a striking 99.9% quenching efficiency. These substrates are efficiently turned over by the human exo-glycosidase O-GlcNAcase (OGA). We find the hemiacetal leaving group rapidly breaks down, enabling quantitative monitoring of OGA activity. We expect this strategy to be broadly useful for the development of substrate probes for monitoring exo-glycosidases, as well as a range of other enzymes having constrained pocket-shaped active sites.

Full text of DOI:10.1021/jacs.7b01948

Communication
pubs.acs.org/JACS
Carbohydrate Bis-acetal-Based Substrates as Tunable Fluorescence-
Quenched Probes for Monitoring exo-Glycosidase Activity
†
,†,‡
Samy Cecioni and David J. Vocadlo*
†
Department of Chemistry, Simon Fraser University, Burnaby, British Columbia, Canada
‡
Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, British Columbia, Canada
*
S Supporting Information
Efforts to create substrate probes that circumvent the steric
constraints of the active site of exo-glycosidases have been made
ABSTRACT: Tunable Fo
̈
rster resonance energy transfer
12,13
(
FRET)-quenched substrates are useful for monitoring the
by using fluorogenic phenolate-based glycosides,
reactive
or spirocyclic rhodol and
While important tools, these
1
4−16
activity of various enzymes within their relevant physio-
logical environments. Development of FRET-quenched
substrates for exo-glycosidases, however, has been hindered
by their constrained pocket-shaped active sites. Here we
report the design of a new class of substrate that
overcomes this problem. These Bis-Acetal-Based Sub-
strates (BABS) bear a hemiacetal aglycon leaving group
that tethers fluorochromes in close proximity, also
positioning them distant from the active site pocket.
Following cleavage of the glycosidic bond, the liberated
hemiacetal spontaneously breaks down, leading to
separation of the fluorophore and quencher. We detail
the synthesis and characterization of GlcNAc-BABS,
revealing a striking 99.9% quenching efficiency. These
substrates are efficiently turned over by the human exo-
glycosidase O-GlcNAcase (OGA). We find the hemiacetal
leaving group rapidly breaks down, enabling quantitative
monitoring of OGA activity. We expect this strategy to be
broadly useful for the development of substrate probes for
monitoring exo-glycosidases, as well as a range of other
enzymes having constrained pocket-shaped active sites.
quinone methide-based probes,
rhodamine-based glycosides.
1
7,18
probes may exhibit limitations in certain applications, such as the
study of the regulation of enzyme activity in different cellular
compartments. For example, the pH-dependent fluorescence of
spirocyclic fluorochromes and phenolate-based substrates may
be problematic in cellular compartments of different pH. Also,
the potential inactivation of enzymes by reactive electrophiles
may present challenges when aiming to quantitate enzyme
activity. Tuning the photophysical properties of fluorogenic
phenolate-based glycosides can also be challenging. Recently, a
different approach has been pursued for quantitating human
19
glucocerebrosidase activity in cells. This “dark-to-bright”
fluorescence-quenched substrate relied on modifying the
primary hydroxyl group of the pyranose. While effective for
glucocerebrosidase, this approach has not proven generally
applicable since hydroxyl groups are often critical recognition
20,21
elements for glycoside hydrolases.
We were therefore interested in overcoming the steric
limitations of exo-processing enzymes while preserving the
carbohydrate recognition motif. We felt this could be
accomplished by positioning both fluorochromes outside of
the active site pocket, where they would not interfere with
substrate recognition. We were inspired by the cyanogenic β-
he creation of fluorescence-quenched fluorescent sub-
strates has delivered valuable probes with tunable
fluorescent properties. Various enzyme classes including, most
notably, proteases have had such substrates deployed for a range
22
T
glucosidase, which cleaves cyanogenic glycosides to form a
cyanohydrin, which subsequently breaks down to liberate
cyanide. We envisioned that a bis-acetal glycoside would position
the two fluorochromes in close proximity and that exo-
glycosidase catalyzed cleavage of the substrate would yield a
hemiacetal that would in turn spontaneously breakdown to
liberate two fragments (Figure 1). If each fragment could bear
one fluorochrome, the net consequence would be that catalytic
activity would lead to separation of these FRET partners and
result in a convenient “dark-to-bright” change in fluorescence.
One concern, however, is the known lability of acetals under
acidic conditions that has been exploited in pH-sensitive
1
−4
of applications both in cells and in vivo. Successful targeting of
proteases likely stems from their cleft-like active site architecture,
which accommodates positioning the requisite pair of
fluorochromes at the N- and C-terminus of peptide-based
substrates. Similarly, FRET-based substrates for endo-glycosi-
dases, which cleave in the middle of glycan chains, have been
5
,6
generated. More problematic has been the creation of such
substrate probes for exo-acting enzymes, which have pocket-
shaped active sites that are sterically demanding.
We were attracted to this problem by the nearly 100 glycoside
hydrolases encoded by the human genome. Most are exo-
4
23,24
prodrugs or linkers for drug delivery.
We recognized that
7
carbohydrate anomeric acetals are chemically distinct from
regular acetals. Due to stabilization by the endocyclic oxygen, we
expected these bis-acetals would show stability across a range of
physiological pH values.
glycosidases, and they are receiving increasing interest owing to
their emerging roles in human health. Deficiencies in the
activities of several exo-glycosidases found in lysosomes are well8-
known to be involved in the set of lysosomal storage diseases.
Exo-glycosidases have also emerged as therapeutic targets for
Received: February 24, 2017
9
10
11
diseases such as Parkinson’s, Alzheimer’s, and cancers.
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b01948
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 2. Synthesis of Bifunctional Bis-acetal Intermediates
Figure 1. Enzymatic hydrolysis of fluorescence-quenched GlcNAc Bis-
Acetal-Based Substrates (GlcNAc-BABS) releases a hemiacetal which in
turn spontaneously decomposes, leading to separation of the
fluorophore/quencher pair. Bis-acetal is shown in the orange oval, and
hemiacetal motif in blue oval.
Here, we detail a generally applicable synthesis of carbohydrate
bis-acetal-based substrates (BABS). This approach yields
efficiently quenched substrates that enable “dark-to-bright”
conversion upon enzymatic processing. We exemplify the
approach using the exo-acting human O-GlcNAcase (hOGA),
which cleaves O-linked β-N-acetylglucosamine (O-GlcNAc)
units from serine and threonine residues of nuclear and
25
cytoplasmic proteins. We selected hOGA for study because it
has emerged as an enzyme of high interest owing to its potential
10
as a therapeutic target for tauopathies. Furthermore, the
complex biology of OGA creates a need for better probes that
could help understand its regulation.
To test our approach, we set out to prepare GlcNAc bis-
acetals. Our synthesis was motivated by Stoodley et al., who used
a glucose pendant group as a chiral auxiliary to explore
stereoselective functionalization of vinylogous carbonates.
We first generated β-formyl glycoside 1 (Scheme 1) using silver
bis-acetals 6 and 8, respectively (Scheme 2). We then developed
a one-pot, three-step, protecting group manipulation that
provided us with carboxylic acids 7 and 9. GlcNAc bis-acetals 7
and 9 are flexible advanced intermediates for the late stage
conjugation of fluorophores and quenchers. We decided to test
our concept of fluorescence-quenched GlcNAc-BABS with the
commonly used EDANS fluorophore and DACBYL quencher
pair (Scheme 3). We therefore used a two-step coupling
26−28
Scheme 1. Synthesis of GlcNAc Vinylogous Carbonate
procedure first reacting EDANS-NH with carboxylic acids 7
2
and 9 using HBTU and DIPEA in DMF. The second step
consisted of copper-catalyzed azide−alkyne cycloaddition
(
CuAAC) with DABCYL-alkyne to yield the protected
nitrate and formic acid. Wittig-type olefination of the formyl
glycoside using 2-tributylphosphorane-methylpropionate deliv-
ered vinylogous carbonates 2-E and 2-Z. Epoxidation of
intermediate 2-E using mCPBA provided epoxide 3 with
complete stereoselectivity (Scheme 2). In model reactions
using methanol, we tested both the acid-catalyzed epoxide
opening of compound 3 using catalytic CSA and the
bromoalkoxylation of compound 2-E using NBS. Both
approaches yielded GlcNAc bis-acetals with complete regiose-
lectivity. The crystalline nature of these products allowed us to
confirm the stereochemical outcome of these routes by X-ray
diffraction of 4 and 5 (Table S1).
GlcNAc-BABS 10 and 12. De-O-acetylation using potassium
carbonate in methanol provided us with the target molecules
GlcNAc-Br-BABS 11 and GlcNAc-OH-BABS 13.
Notably, we found no detectable hydrolysis of the bis-acetal
motif at pH values as low as 2 and as high as 10 (Figure S1). This
remarkable stability likely stems from the participation of the
endocyclic oxygen in the bis-acetal motif. We also evaluated the
quenching efficiency of GlcNAc-BABS, and we found 99.93%
and 99.89% quenching efficiencies for 11 and 13 respectively
(Figure S2). This remarkable 1000-fold fluorescence quenching
highlights that this design positions the fluorophore and
quencher in close proximity, potentially enabling contact
quenching. Interestingly, at higher concentrations of GlcNAc-
BABS, we observed an inner-filter effect (IFE) (Figure S3) that is
presumably due to the fluorescence signal being screened
We next applied this methodology to the generation of
bifunctional GlcNAc bis-acetals by bromoalkoxylation of 2-E and
epoxide opening of 3 using 2-(2-azidoethoxy)ethanol to afford
B
DOI: 10.1021/jacs.7b01948
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 3. Conjugation of Fluorophore and Quencher Enables Generation of GlcNAc-BABS
intermolecularly by nearby GlcNAc-BABS. This effect is
commonly observed and corrected for when using fluores-
29
cence-quenched substrates.
We next set out to test whether hOGA could hydrolyze these
GlcNAc-BABS and whether the intermediate hemiacetal would
quickly breakdown to produce measurable fluorescence on a
suitable time scale. Incubation of GlcNAc-BABS 11 and 13 with
hOGA showed increasing rates with increasing GlcNAc-BABS
concentrations (Figure 2a). We measured these steady-state
rates for various concentrations of GlcNAc-BABS and, after
correction for IFE, determined the second-order rate constants
for each substrate probe (Figure 2b, Figure S4). hOGA catalyzed
hydrolysis of GlcNAc-Br-BABS 11, and GlcNAc-OH-BABS 13
−1
−1
revealed k /K values of 263 and 519 M s , respectively.
cat
m
Interestingly, these values are comparable to the second-order
rate constant measured for methyl β-D-N-acetylgluco-
3
0
−1 −1
pyranoside (440 M s ) and about 50-fold higher than the
rate constants measured for O-GlcNAc-modified protein
3
1
substrates. We also confirmed that the observed rates were
linearly dependent on hOGA concentrations (Figure 2c).
Accordingly, both GlcNAc-BABS 11 and 13 are competent
substrates that are processed at comparable rates as the natural
substrates of hOGA.
During these assays, we noticed a short lag phase (∼1 min) in
our initial rate experiments. The initial velocities gradually
reached linear rates (Figure 2a). We speculated that this lag
stemmed from an increase in the concentration of a hemiacetal
(
HA) intermediate to reach a steady-state concentration. To test
Figure 2. GlcNAc-BABS probes are turned over by hOGA. (a)
Evolution of fluorescence (RFU) for different concentrations of
GlcNAc-BABS in the presence of hOGA. Dotted lines represent the
linear rates reached at steady state. (b) Rates of hOGA catalyzed
hydrolysis of different concentrations of GlcNAc-BABS. (c) Rates of
hydrolysis of GlcNAc-BABS depend on the concentrations of hOGA.
this idea and determine the first-order rate constant for
decomposition of these hemiacetals (HA), we developed an
assay in which we can rapidly halt enzyme activity (Figure 2d).
The assay was initiated by addition of hOGA, allowed to proceed
for 1 min, and then the enzymatic reaction was rapidly stopped
by addition of a high concentration of the tight binding hOGA
(
d) Hemiacetal breakdown assay design containing high hOGA
concentration. Stopped after 1 min by addition of Thiamet-G. (e)
Evolution of fluorescence for the hemiacetal breakdown assay with
different concentrations of GlcNAc-OH-BABS. (f) Measurement of
first-order rate constants for the breakdown of hemiacetal after addition
of Thiamet-G.
32
inhibitor Thiamet-G (100 μM; 50 000 × K ; K = 2 nM). We
i
i
could then monitor the decomposition of any accumulated
hemiacetal (HA) (Figure 2e−f). Interestingly, we found that the
hemiacetal (HA) generated from GlcNAc-OH-BABS 13 broke
down 4.5 times faster than the one generated from GlcNAc-Br-
BABS 11, with an observed first-order rate constant (k ) of 4.4 ×
3
13 as compared to 11 (Figure 2a). To evaluate the compatibility
of the BABS design with complex cellular extracts, we also
measured the turnover of GlcNAc-Br-BABS in SK-N-SH cell
lysate in the presence of increasing concentrations of hOGA
(Figure S5). This substrate is efficiently processed within lysates,
at rates comparable to that observed in buffer. Taken together,
these results support that the steady-state initial rates reliably
report on the kinetics of the enzymatic cleavage of these GlcNAc-
BABS probes.
−2
−2 −1
10
and 1 × 10
s respectively. These results are comparable
to reported rate constants for the hydrolysis of acetaldehyde
−
2
−1
33
methoxyethyl hemiacetal (k = ∼5 × 10 s , pH = 7.5). In
0
addition, the stabilizing effect of the electron-withdrawing
bromine atom is in good agreement with the electronic effects
of substituents on hemiacetal breakdown. Notably, these rate
constants are also consistent with the more rapid approach to
steady state upon hOGA-catalyzed hydrolysis of GlcNAc-BABS
3
4
C
DOI: 10.1021/jacs.7b01948
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Collectively these data validate the proof of concept for the
Glyco-BABS design as efficient fluorescence-quenched sub-
strates and suggest that Glyco-BABS probes may prove to be
valuable for monitoring glycosidase activities in cells. Cell active
fluorescence-quenched Glyco-BABS could provide useful
information regarding target engagement by inhibitors directly
in live cells as well as permit understanding of endogenous factors
that regulate enzyme activity. Accordingly, these substrate probes
should prove complementary to covalent activity-based probes
(4) Evans, M. J.; Cravatt, B. F. Chem. Rev. 2006, 106, 3279−3301.
(5) Cottaz, S.; Brasme, B.; Driguez, H. Eur. J. Biochem. 2000, 267,
5
(
593−5600.
6) Yang, G.-Y.; Li, C.; Fischer, M.; Cairo, C. W.; Feng, Y.; Withers, S.
G. Angew. Chem., Int. Ed. 2015, 54, 5389−5393.
7) Lombard, V.; Golaconda Ramulu, H.; Drula, E.; Coutinho, P. M.;
Henrissat, B. Nucleic Acids Res. 2014, 42, D490−D495.
8) Boyd, R. E.; Lee, G.; Rybczynski, P.; Benjamin, E. R.; Khanna, R.;
(
(
Wustman, B. A.; Valenzano, K. J. J. Med. Chem. 2013, 56, 2705−2725.
(9) Sidransky, E.; Lopez, G. Lancet Neurol. 2012, 11, 986−998.
(10) Yuzwa, S. A.; Vocadlo, D. J. Chem. Soc. Rev. 2014, 43, 6839−6858.
(11) He, L.; Fan, C.; Kapoor, A.; Ingram, A. J.; Rybak, A. P.; Austin, R.
C.; Dickhout, J.; Cutz, J.-C.; Scholey, J.; Tang, D. Nat. Commun. 2011, 2,
1
,4
14,35
(ABP) for exo-glycosidases,
which report on the levels of
active enzyme in cells. The modular approach used here will
enable late-stage incorporation of bright fluorophores, including
cell-compatible red-shifted fluorophores to help move toward
quantitative monitoring of hOGA activity in cells. Such cell active
3
(
07.
12) Burke, H. M.; Gunnlaugsson, T.; Scanlan, E. M. Chem. Commun.
2015, 51, 10576−10588.
13) Harlan, F. K.; Lusk, J. S.; Mohr, B. M.; Guzikowski, A. P.;
Batchelor, R. H.; Jiang, Y.; Naleway, J. J. PLoS One 2016, 11, e0156312.
14) Komatsu, T.; Kikuchi, K.; Takakusa, H.; Hanaoka, K.; Ueno, T.;
12
Glyco-BABS should complement other substrate probes and
could be particularly valuable in cases where these other substrate
designs show limitations. For example, Glyco-BABS derivatives
targeting lysosomal and endolysosomal enzymes may prove
useful, especially because pH-insensitive fluorochromes can be
installed. Indeed, one of the main advantages of this design is its
versatility, as developing a probe for another member of the large
class of exo-glycosidases may simply require another core
carbohydrate residue. Finally, we also expect this acetal and
bis-acetal concept could be applied to other classes of hydrolases
including, for example, sulfatases and phosphatases which have
similarly constrained active sites.
(
(
Kamiya, M.; Urano, Y.; Nagano, T. J. Am. Chem. Soc. 2006, 128, 15946−
15947.
(15) Han, J.; Han, M. S.; Tung, C.-H. Mol. BioSyst. 2013, 9, 3001−
3008.
(
16) Ho, N.-H.; Weissleder, R.; Tung, C.-H. ChemBioChem 2007, 8,
60−566.
17) Kamiya, M.; Asanuma, D.; Kuranaga, E.; Takeishi, A.; Sakabe, M.;
Miura, M.; Nagano, T.; Urano, Y. J. Am. Chem. Soc. 2011, 133, 12960−
2963.
18) Sakabe, M.; Asanuma, D.; Kamiya, M.; Iwatate, R. J.; Hanaoka, K.;
5
(
1
(
Terai, T.; Nagano, T.; Urano, Y. J. Am. Chem. Soc. 2013, 135, 409−414.
(19) Yadav, A. K.; Shen, D. L.; Shan, X.; He, X.; Kermode, A. R.;
Vocadlo, D. J. J. Am. Chem. Soc. 2015, 137, 1181−1189.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b01948.
■
*
S
(20) McCarter, J. D.; Adam, M. J.; Withers, S. G. Biochem. J. 1992, 286,
7
21−727.
21) Namchuk, M. N.; Withers, S. G. Biochemistry 1995, 34, 16194−
6202.
(22) Barrett, T.; Suresh, C. G.; Tolley, S. P.; Dodson, E. J.; Hughes, M.
A. Structure 1995, 3, 951−960.
23) Gillies, E. R.; Goodwin, A. P.; Frec
004, 15, 1254−1263.
24) Liu, B.; Thayumanavan, S. J. Am. Chem. Soc. 2017, 139, 2306−
317.
(25) Vocadlo, D. J. Curr. Opin. Chem. Biol. 2012, 16, 488−497.
26) Larsen, D. S.; Stoodley, R. J. J. Chem. Soc., Perkin Trans. 1 1989,
841−1852.
27) Bhatia, G. S.; Lowe, R. F.; Pritchard, R. G.; Stoodley, R. J. Chem.
(
1
Methods, compound characterization (PDF)
Crystallographic data (CIF)
(
́
het, J. M. J. Bioconjugate Chem.
AUTHOR INFORMATION
Corresponding Author
dvocadlo@sfu.ca.
ORCID
David J. Vocadlo: 0000-0001-6897-5558
Notes
The authors declare the following competing financial
interest(s): The authors note they have filed a provisional patent
application pertaining to this research.
■
*
2
(
2
(
1
(
Commun. 1997, 1981−1982.
(28) Idris, M. S.; Larsen, D. S.; Schofield, A.; Stoodley, R. J.; Tiffin, P.
D. Tetrahedron Lett. 1995, 36, 3251−3254.
(29) Matayoshi, E.; Wang, G.; Krafft, G.; Erickson, J. Science 1990, 247,
ACKNOWLEDGMENTS
This work was supported by grants from the Natural Sciences
and Engineering Research Council of Canada (RGPIN/-2015-
5426) and the Canadian Institutes of Health Research (MOP-
02756). D.J.V. thanks the Canada Research Chairs Program for
support as a Tier I CRC in Chemical Biology. S.C. thanks the
CIHR for a postdoctoral fellowship. We thank Dr. John R.
Thompson for assistance with X-ray data collection and analysis.
954−958.
30) Greig, I. R.; Macauley, M. S.; Williams, I. H.; Vocadlo, D. J. J. Am.
Chem. Soc. 2009, 131, 13415−13422.
31) Shen, D. L.; Gloster, T. M.; Yuzwa, S. A.; Vocadlo, D. J. J. Biol.
■
(
(
0
1
Chem. 2012, 287, 15395−15408.
(32) Yuzwa, S. A.; Macauley, M. S.; Heinonen, J. E.; Shan, X.; Dennis,
R. J.; He, Y.; Whitworth, G. E.; Stubbs, K. A.; McEachern, E. J.; Davies,
G. J.; Vocadlo, D. J. Nat. Chem. Biol. 2008, 4, 483−490.
(
33) Sorensen, P. E.; Jencks, W. P. J. Am. Chem. Soc. 1987, 109, 4675−
4
690.
(
(
34) Przystas, T. J.; Fife, T. H. J. Am. Chem. Soc. 1981, 103, 4884−4890.
REFERENCES
■
35) Witte, M. D.; Kallemeijn, W. W.; Aten, J.; Li, K.-Y.; Strijland, A.;
(
1
1) Garland, M.; Yim; Joshua, J.; Bogyo, M. Cell Chem. Biol. 2016, 23,
Donker-Koopman, W. E.; van den Nieuwendijk, A. M. C. H.;
Bleijlevens, B.; Kramer, G.; Florea, B. I.; Hooibrink, B.; Hollak, C. E.
M.; Ottenhoff, R.; Boot, R. G.; van der Marel, G. A.; Overkleeft, H. S.;
Aerts, J. M. F. G. Nat. Chem. Biol. 2010, 6, 907−913.
22−136.
(
2) Hu, H.-Y.; Vats, D.; Vizovisek, M.; Kramer, L.; Germanier, C.;
Wendt, K. U.; Rudin, M.; Turk, B.; Plettenburg, O.; Schultz, C. Angew.
Chem., Int. Ed. 2014, 53, 7669−7673.
3) Whitney, M.; Savariar, E. N.; Friedman, B.; Levin, R. A.; Crisp, J. L.;
(
Glasgow, H. L.; Lefkowitz, R.; Adams, S. R.; Steinbach, P.; Nashi, N.;
Nguyen, Q. T.; Tsien, R. Y. Angew. Chem., Int. Ed. 2013, 52, 325−330.
D
DOI: 10.1021/jacs.7b01948
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX