A universal flow cytometry assay for screening carbohydrate-active enzymes using glycan microspheres

Article abstract of DOI:10.1039/c3cc39155h

We describe a simple, multiplexed assay that integrates glycan synthesis, bioconjugation to microspheres, fluorescent chemical/biochemical detection and multiparameter flow cytometric analysis to screen activities of different families of carbohydrate-active enzymes.

Full text of DOI:10.1039/c3cc39155h

Chemical Communications
www.rsc.org/chemcomm
Volume 49 | Number 48 | 18 June 2013 | Pages 5431–5518
ISSN 1359-7345
COMMUNICATION
Aarthi Chandrasekaran, Anup K. Singh et al.
A universal flow cytometry assay for screening carbohydrate-active enzymes
using glycan microspheres
1359-7345(2013)49:48;1-X

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A universal flow cytometry assay for screening
carbohydrate-active enzymes using glycan
microspheres†
Aarthi Chandrasekaran,*ab Kai Deng,ab Chung-Yan Koh,b Taichi Takasuka,c
Lai F. Bergeman,c Brian G. Fox,c Paul D. Adamsad and Anup K. Singh*ab
Cite this: Chem. Commun., 2013,
49, 5441
Received 21st December 2012,
Accepted 11th March 2013
DOI: 10.1039/c3cc39155h
www.rsc.org/chemcomm
We describe a simple, multiplexed assay that integrates glycan
synthesis, bioconjugation to microspheres, fluorescent chemical/
biochemical detection and multiparameter flow cytometric analysis to
screen activities of different families of carbohydrate-active enzymes.
Glycosyltransferases (GTs) and glycoside hydrolases (GHs) are
carbohydrate-active enzymes responsible for the biosynthesis
and breakdown of glycans in animals, plants and bacteria. By
controlling the diversity of glycans, GTs and GHs regulate a
number of cellular and disease processes in humans.1 Given their
diverse biological roles, several small molecule inhibitors of GTs
and GHs are in clinical use for treatment of diseases such as
tuberculosis, lysosomal storage diseases and influenza.2,3 Further,
these enzymes are being utilized for synthesis of novel carbo-
hydrates for drug/vaccine development and diagnostics.4,5 More
recently, recombinant GTs and GHs have been gaining promi-
nence in the biotechnology industry for their roles in the produc-
tion of liquid biofuels and chemicals from plant-derived cellulosic
Fig. 1 Glycan microsphere-based flow cytometry assay. (A) Substrates for GT
assay and schematic of representative GT assay using lactose-conjugated micro-
spheres. (B) Substrates for GH assay and schematic of a representative GH assay
using N3-cellotetraose-conjugated microspheres.
biomass.6,7 Manipulating cell-wall composition (by regulating GT To overcome these limitations and provide a universal screening
activities) and increasing the efficiency of the enzymatic hydrolysis platform for both GHs and GTs, we have developed a simple flow
of cell-wall polysaccharides (by engineering improved GHs) are cytometry assay using glycan microspheres. Conventionally,
critical for lowering costs and improving production of biofuels.7–9 microspheres have been extensively utilized for immunoassay
Current methods to screen GTs and GHs are plagued with development10, and more recently, employed in applications such
limitations such as requirement for specialized equipment and as genotyping11 and enzyme assays.12 In this study, we demon-
training (in the case of mass spectrometry, HPLC, CE assays), strate for the first time the potential of an assay that combines
special sample and waste handling (radiolabeled assays) and covalent coupling of synthetic glycan substrates with micro-
are currently available only for specific enzyme families (in the spheres with standard flow cytometry technologies to provide a
case of assays with colorimetric/fluorescent substrates).3,7,9 general screening platform that can be easily adapted to any
laboratory setting (Fig. 1 and Fig. S1, ESI†). Further, our approach
permits easy multiplexing and is amenable to high throughput
(HTP) screening of large enzyme libraries.
a Technology Division, Joint BioEnergy Institute, Emeryville, CA, USA.
E-mail: achandrasekaran@lbl.gov, aksingh@sandia.gov; Fax: +1-925-294-3020;
For GT assays, the glycan-functionalized microspheres are incu-
Tel: +1-925-294-1260
b Biotechnology and Bioengineering Department, Sandia National Laboratories,
bated with the GTs and an appropriate activated sugar nucleotide
donor. GTs transfer the monosaccharide from the activated donor
onto the acceptor sugar (coupled with microspheres) thereby
generating a new glycan product on the microspheres. GT activity is
tested using specific fluorescent lectins or antibodies that recognize
the newly formed glycosidic linkages. The labeled microspheres are
then analyzed by flow cytometry. The formation of the new
Livermore, CA, USA
c Great Lakes Bioenergy Research Center, University of Wisconsin, Madison,
WI, USA
d Physical Biosciences Division, Lawrence Berkeley National Laboratory, Berkeley,
CA, USA
† Electronic supplementary information (ESI) available: Additional data, glycan
substrate synthesis, methods. See DOI: 10.1039/c3cc39155h
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5441--5443 5441

View Article Online
Communication
ChemComm
glycosidic linkage leads to an increase in lectin binding indicated linkage is detected by adding fluorescein-SNA lectin. Flow cytometry
by an increase in fluorescence in the flow cytometry assay (Fig. 1a). analysis demonstrated a significant increase (B10-fold) in fluores-
Numerous lectins and antiglycan antibodies are currently available cence of microspheres treated with 6-ST and CMP-sialic acid
for characterizing a wide variety of plant and animal glycan compared with the control (no enzyme) microspheres (Fig. 2a).
products.13,14 In addition, the assay can be adapted to screen Our assay is extremely sensitive and can detect as low as
GTs that accept unnatural activated monosaccharide donors 25 fmol of 6-ST enzyme (Fig. S3, ESI†). The second enzyme tested,
containing chemical handles (e.g., azides, biotin) that can GalT, transfers galactose from the donor sugar (UDP-galactose) to
subsequently be detected by labelled probes.15,16
the acceptor amine-modified glucose 2, to form lactose (Fig. S2,
For GH assays, we use glycans modified with a bio-orthogonal ESI†). Detection using fluorescein-RCA-I lectin (specific to the
chemical handle, azide, on the non-reducing end as substrates. newly formed Gal-b-1,4-Glc linkage) shows a significant increase
The azide-modified substrate coupled with the microspheres is in fluorescence compared to the control microspheres (Fig. 2b).
detected by Staudinger ligation using a fluorescent phosphine Using 6-ST as an example, we further extend the assay to monitor
probe.17 In the presence of active GHs, the glycosidic bonds GT reaction kinetics. We observe 6-ST activity within 5 minutes
within the attached substrate are cleaved resulting in a loss of and saturation after B30 minutes (Fig. 2c).
fluorescence due to loss of the azide from the microsphere
surface, which is quantified by flow cytometry (Fig. 1b).
To validate the GH assay, we chose a cellulase–cellulose
oligosaccharide system as a model GH enzyme–substrate system.
In this study, a panel of glycan substrates were synthesized We chemically synthesized substrate 3, a bifunctionalized cello-
for screening representative GTs and GHs (Fig. 1, compounds 1, tetraose substrate with a C5-amine linker on the reducing end and
2, 3 and 4). The reducing ends of all glycans are functionalized an azide handle on the non-reducing end (Fig. S7 and S8, ESI†).
with a linker containing a terminal amine group (a C5–NH2 Following conjugation of substrates (Fig. S4B, ESI†), we incubated
linker). These amine-modified glycans are coupled with carboxyl the microspheres with four different purified recombinant
polystyrene microspheres using standard carbodiimide-based Clostridium thermocellum cellulase enzymes to initiate the hydrolysis
peptide coupling protocols (see Fig. S6 and methods, ESI†).
of substrate 3 attached to the microsphere. These cellulases release
The GT assay was validated by characterizing the activity of two the products, azide–glucose, azide–cellobiose or azide–cellotriose,
recombinant GT enzymes: bacterial Photobacterium damsela a-2,6- from the microsphere surface (Fig. S5, ESI†). Any remaining azides
sialyltransferase (6-ST) and human b-1,4-galactosyltransferase on the microsphere are fluorescently labeled by Staudinger ligation
(GalT). The 6-ST enzyme transfers sialic acid (Neu5Ac) from the using Dylight 488-phosphine reagent and quantified using flow
donor (CMP-sialic acid) to the amine modified-lactose substrate 1, cytometry. As shown in Fig. 2d, incubation with all the four
coupled with the microspheres, to form the product a-2,6-sialyl- cellulases resulted in cleavage of substrate 3 and subsequent loss
lactose (Fig. S2 and S4A, ESI†). The newly formed Neu5Ac-a-2,6-Gal of fluorescence compared to microsphere samples not incubated
Fig. 2 Glycan microsphere-based GT and GH assay. (A) Analysis of GT activity of 6-ST using lactose-microspheres, (B) GalT activity using glucose-microspheres,
(C) reaction time course for 6-ST activity. Data in 2A–C are representative of three separate experiments. (D) Flow cytometry analysis of GH activities of four
recombinant cellulases using N3-cellotetraose (N3-DP4) conjugated microspheres: CelAcc_CBM3a; CelKcc; CelLcc_CBM3a; CelRcc_CBM3a. Data are representative of at
least two separate experiments. (E) Comparison of GH activities of the four cellulases. Values are quantified by calculating the median of fluorescence values for
approximately 40,000 events per sample and then normalizing the median values to that of the no enzyme control. Bars represent percentage with respect to control.
Data show the average of two separate experiments Æ SD. (F) Reaction time course for a representative enzyme, CelAcc_CBM3a. Data are representative of three
separate experiments. See ESI† for detailed methods.
c
5442 Chem. Commun., 2013, 49, 5441--5443
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
(e.g., Staudinger ligation17 or ‘click-chemistry’ Huisgen cyclo-
addition reactions18,19) methods for detecting the glycan pro-
ducts formed from GT/GH activities. Recent advances in HTP
combinatorial glycan synthesis strategies20 and in multipara-
meter, microsphere-based flow cytometry assays (e.g., Luminex
platform)10 enable massive multiplexing of hundreds of dis-
crete assays with a single sample. By integrating our assay with
such technologies, one can envision a ‘glycan library-on-a-bead’
strategy with the ability to screen activities of large libraries of
different GT and GH subfamilies using hundreds of unique
glycan substrates. Such assays will be invaluable for advance-
ments in GT and GH research for applications in drug and
biomarker discovery, diagnostics and biofuels development.
The authors thank Dr. Rajiv Bharadwaj for insightful scien-
tific discussions and Prof. Richmond Sarpong of UC Berkeley
for using the instruments in his group. Work at the DOE
Joint BioEnergy Institute and DOE Great Lakes Bioenergy
Fig. 3 Multiplexed enzymatic assay. (A) Dot plot (forward scatter vs. side scatter)
depicts distinct regions for each microsphere population (1 mm versus 6 mm) based
on microsphere size. (B) Multiplexed analysis of BG and EC activities of CTEC2.
Values are quantified by calculating the median of fluorescence values for
approximately 40,000 events per sample and then normalizing the median values
to the value of the no CTEC2 control. Bars represent percentage with respect to
control (100%). Data show the average of three separate experiments Æ SD.
with any enzymes. Different enzymes showed different levels of loss Research Center (GLBRC) is supported by the US DOE, Office
of fluorescence indicating different hydrolytic activities for each of Science, Office of Biological and Environmental Research,
enzyme with this substrate (Fig. 2e). For instance, while the cellulase through contracts DE-AC02-05 CH11231 (LBNL) and DE-FC02-
CelAcc_CBM3a showed maximum activity, CelRcc_CBM3a showed 07ER64494 (GLBRC). Sandia is a multiprogram laboratory
minimal activity with substrate 3. Using a representative cellulase, operated by Sandia Corporation, a Lockheed Martin Company,
CelAcc_CBM3a, we demonstrate that our assay can be extended to for US DOE’s Nuclear Security Administration under contract
study reaction time course of cellulase activity. We observe a DE-AC04-94AL85000.
significant loss of fluorescence in as little as 2 minutes after enzyme
addition (B50% of initial value) with values dropping to 15–20% of
the initial value by 20 minutes (Fig. 2f).
A key feature of our assay is the ability to multiplex, where
multiple enzyme activities can be analyzed simultaneously.
Notes and references
1 K. Ohtsubo and J. D. Marth, Cell, 2006, 126, 855–867.
2 J. R. Brown, B. E. Crawford and J. D. Esko, Crit. Rev. Biochem. Mol.
Biol., 2007, 42, 481–515.
Multiplexing can be achieved using microspheres of different
sizes or fluorescent properties where each individual micro-
sphere population is coupled with a specific substrate. We
demonstrate a proof-of-principle multiplexed assay by characteri-
zing the cellulolytic activities of a cellulase cocktail (CTEC2) on
two different substrates in a single experiment. CTEC2 contains a
mixture of enzymes including beta-glucosidases (BGs) and exo/
endocellulases (ECs). ECs do not work on cellobiose substrates
and require longer cellulose oligomers as substrates while BGs
typically hydrolyze cellobiose to glucose with decreasing affinities
for longer cellulose oligomers. To profile the cellulolytic activities
of CTEC2, we developed a multiplexed assay by combining
microspheres of different sizes (1 mm and 6 mm). 1 mm micro-
spheres were conjugated with a bifunctionalized cellobiose sub-
strate, 4 (Fig. S7 and S9, ESI†) and the 6 mm microspheres were
conjugated with substrate 3. These two sizes were chosen because
the individual microsphere populations cluster into two separate
and distinct regions on a dot plot with minimal overlap between
the clusters (Fig. 3a). Our assay shows that CTEC2 causes B65%
loss of fluorescence with the microspheres that contain 4 and
B45% loss of fluorescence signal with microspheres that contain
3 G. K. Wagner and T. Pesnot, ChemBioChem, 2010, 11, 1939–1949.
4 B. Lepenies, J. Yin and P. H. Seeberger, Curr. Opin. Chem. Biol., 2010,
14, 404–411.
5 P. H. Liang, C. Y. Wu, W. A. Greenberg and C. H. Wong, Curr. Opin.
Chem. Biol., 2008, 12, 86–92.
6 W. R. Scheible and M. Pauly, Curr. Opin. Plant Biol., 2004, 7, 285–295.
7 Y. H. P. Zhang, M. E. Himmel and J. R. Mielenz, Biotechnol. Adv.,
2006, 24, 452–481.
8 M. Anbar and E. A. Bayer, Methods Enzymol., 2012, 510, 261–271.
9 C. Brown, F. Leijon and V. Bulone, Nat. Protocols, 2012, 7,
1634–1650.
10 V. V. Krishhan, I. H. Khan and P. A. Luciw, Crit. Rev. Biotechnol.,
2009, 29, 29–43.
11 S. Bruse, M. Moreau, M. Azaro, R. Zimmerman and L. Brzustowicz,
BioTechniques, 2008, 45, 559–571.
12 M. J. Saunders, H. Kim, T. A. Woods, J. P. Nolan, L. A. Sklar,
B. S. Edwards and S. W. Graves, Cytometry, Part A, 2006, 69, 342–352.
13 R. D. Cummings and M. E. Etzler, in Essentials of Glycobiology, ed.
A. Varki, R. D. Cummings, J. D. Esko, H. H. Freeze, P. Stanley,
C. R. Bertozzi, G. W. Hart and M. E. Etzler, Cold Spring Harbor, NY,
2nd edn, 2009.
14 S. Pattathil, U. Avci, D. Baldwin, A. G. Swennes, J. A. McGill,
Z. Popper, T. Bootten, A. Albert, R. H. Davis, C. Chennareddy,
R. Dong, B. O’Shea, R. Rossi, C. Leoff, G. Freshour, R. Narra,
M. O’Neil, W. S. York and M. G. Hahn, Plant Physiol., 2010, 153,
514–525.
15 O. Blixt, K. Allin, O. Bohorov, X. Liu, H. Andersson-Sand,
J. Hoffmann and N. Razi, Glycoconjugate J., 2008, 25, 59–68.
3 when compared with microspheres that were incubated only 16 H. C. Hang, C. Yu, M. R. Pratt and C. R. Bertozzi, J. Am. Chem. Soc.,
2004, 126, 6–7.
with buffer (no CTEC2 control) (Fig. 3b).
17 E. Saxon and C. R. Bertozzi, Science, 2000, 287, 2007–2010.
18 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, Angew.
In conclusion, we demonstrate a simple, multiplexed,
glycan microsphere-based bioassay for rapid screening of
carbohydrate-active enzymes using standard flow cytometry
technologies. Our assay is compatible with a number of bio-
Chem., Int. Ed., 2002, 41, 2596–2599.
19 J. M. Baskin, J. A. Prescher, S. T. Laughlin, N. J. Agard, P. V. Chang,
I. A. Miller, A. Lo, J. A. Codelli and C. R. Bertozzi, Proc. Natl. Acad.
Sci. U. S. A., 2007, 104, 16793–16797.
chemical (fluorescent lectins and antibodies) and chemical 20 O. Blixt and N. Razi, Methods Enzymol., 2006, 415, 137–153.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5441--5443 5443