Magnetic glyco-nanoparticles: A unique tool for rapid pathogen detection, decontamination, and strain differentiation

Article abstract of DOI:10.1021/ja076086e

Many pathogens use mammalian cell surface carbohydrates as anchors for attachments, which subsequently results in infection. The unique combination of magnetic nanoparticles and diverse carbohydrate bioactivities prompts us to develop a magnetic glyco-nanoparticle (MGNP)-based system to rapidly detect Escherichia coli (E. coli) in just 5 min. High capture efficiencies up to 88% were achieved. Moreover, to accurately differentiate three E. coli strains, response patterns of MGNPs were utilized to decipher the pathogen identity. This appealing approach may have clinical applications since the virulence of many pathogens can be correlated with carbohydrate binding specificity. To the best of our knowledge, this is the first time that MGNPs have been used to detect, quantify, and differentiate E. coli cells, which can provide an exciting new avenue for decontamination and diagnostic applications. Copyright

Full text of DOI:10.1021/ja076086e

Published on Web 10/12/2007
Magnetic Glyco-nanoparticles: A Unique Tool for Rapid Pathogen Detection,
Decontamination, and Strain Differentiation
†
‡
,†
Kheireddine El-Boubbou, Cyndee Gruden, and Xuefei Huang*
Department of Chemistry, The UniVersity of Toledo, 2801 West Bancroft Street, MS 602, Toledo, Ohio 43606,
Department of CiVil Engineering, The UniVersity of Toledo, 2801 West Bancroft Street, MS 307,
Toledo, Ohio 43606
Received August 13, 2007; E-mail: xuefei.huang@utoledo.edu
The appealing mechanisms by which nanocomposites knit
biomolecules not only lend credence in designing novel nanosensors
but considerably advance medical applications. Recent headline
Scheme 1
1
news about Escherichia coli (E. coli) contamination in produce and
Bacillus anthracis attacks pinpoint the urgent need for an effective
method for microbial decontamination and rapid detection without
time-consuming cell culturing. It is known that many bacteria use
mammalian cell surface carbohydrates as anchors for attachments,
2
which subsequently results in infection. The unique combination
of magnetic nanocomposites and diverse carbohydrate bioactivities
prompts us to embark on a biosensing research program. Herein,
we report a magnetic glyco-nanoparticle (MGNP)-based system to
not only detect E. coli within 5 min, but also remove up to 88% of
the target bacteria from the medium. Furthermore, the identities of
three different E. coli strains were easily determined on the basis
of the response patterns to two MGNPs highlighting their potential
in biosensing.
Huisgen reaction with immobilized alkynes.5 NP 1 without
carbohydrates (control) did not remove any Con A, proving that
the separation of Con A is due to its interaction with carbohydrates,
rather than the nonspecific absorption to NP surface. Incubation of
100 mM mannose with MGNP 3 (2 mg/mL) and Con A did not
increase the intensity of residual emission of the supernatant after
magnetic separation, suggesting that the avidity of MGNP 3 to Con
A is at least 200-fold higher than the affinity of the monomeric
mannose. These results reveal the multivalent appeal of MGNP
with ligand clustering leading to strong binding.
It is advantageous to use magnetic nanoparticles (NPs) for
detection. The high surface/volume ratio offers more contact surface
3
area for attaching carbohydrates and for capturing pathogens. The
sizes of NPs are typically about 2 orders of magnitude smaller than
a bacterium, which allows the attachment of multiple NPs onto a
bacterial cell rendering easy magnet-mediated separation.3 More-
over, the small NPs have faster kinetics in solution as compared to
their micrometer-sized counterparts, which can result in fast
detection.
The above experiments set the stage for detection of bacterium
E. coli. After incubating MGNP 3 (2 mg/mL) with solutions of an
,4
3
7
E. coli strain ORN178 (10 -10 cells/mL in PBS buffer) for a few
minutes, a magnetic field was applied separating MGNP/E. coli
aggregates (Figure 1). The supernatants were carefully removed
and the remaining aggregates were washed thoroughly, stained with
a fluoresecent dye (PicoGreen), transferred to a glass slide, and
imaged. Fluorescent microscopic imaging showed that E. coli can
Our journey commenced with functionalization of silica-coated
magnetite NP (NP 1) with D-mannose (Man) through either a
triazole linker (MGNP 2) formed by the [2+3] Huisgen reaction5
or an amide linkage (MGNP 3) (Schemes 1 and S1). With our
4
be reliably detected with a limit of 10 cells/mL (Figure 2a). With
NP 1, no bacteria were observed on the slides.
6
covalent approach, all carbohydrates are uniformly oriented on the
Following the same protocol, we enumerated bacteria in the
aggregate and supernatant using the fluorescent microscope. The
capture efficiency was calculated by dividing the number of E. coli
in MGNP aggregate over the total number of cells in both the
supernatant and the aggregate. High capture efficiencies up to 88%
can be achieved with 45 min incubation (Table S1). Next, the effect
of incubation time was examined. Interestingly, even in just 5 min,
E. coli can be detected with a high capture efficiency of 65% (Table
S2). MGNP/E. coli complexes were then imaged by TEM with
MGNP aggregates observed on the surface, at the lateral ends and
NP surface, which is crucial for high performances in cell-capturing
studies.7 All MGNPs were characterized by X-ray diffraction,
transmission electron microscopy (TEM), thermogravimetric analy-
ses, and IR spectroscopy (Figures S1-4).
To ensure that carbohydrates on MGNP retain their binding
abilities, the interaction between various MGNPs with a mannose
8
binding lectin, concanavalin A (Con A), was first investigated.
9
Carbohydrate-lectin interaction is central in devising our bio-
sensor. After mixing NPs with fluorescein-labeled Con A, a
magnetic field was applied to the mixture through a handheld
magnet inducing aggregation of magnetic NPs on the side of the
vial. The residual fluorescence of supernatants was then recorded
(Figure S5). With MGNP 3, the emission intensity of the supernatant
decreased 87% indicating that most Con A was removed by 3.
Triazole linked MGNP 2 was less efficient accounting for a 60%
emission decrease probably because of the low efficiency of the
†
Department of Chemistry.
Department of Civil Engineering.
‡
Figure 1. Schematic demonstration of pathogen detection by MGNPs.
13392
9
J. AM. CHEM. SOC. 2007, 129, 13392-13393
10.1021/ja076086e CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. E. coli strain differentiation by MGNPs 3 and 4.
In conclusion, we demonstrate the potential of sugar-coated
magnetic nanoparticles for fast bacterial detection and removal,
which provides an attractive avenue for pathogen decontamination
and diagnostic applications.
Figure 2. (a) Representative fluorescence microscopic images of captured
E. coli. The concentration (cells/mL) of bacteria incubated with MGNP 3
is indicated on each image (see Supporting Information for experimental
details). (b-d) TEM images of MGNP 3/E. coli complexes.
Acknowledgment. We are grateful for DARPA (Grant HR0011-
7-1-0003), the American Heart Association (0715118B), and
0
University of Toledo for financial support. We thank Profs. Terry
Bigioni and William Gunning (University of Toledo) for TEM
training, Prof. Paul Orndorff (North Carolina State University) for
providing us the E. coli ORN178 and ORN208 strains, and Mr.
Vivek Tewari and Guang Cai for their help in research.
along the pili10 of E. coli cells (Figure 2b-d, Figure S4). To the
best of our knowledge, although glyco-nanoparticles have been
studied as bioprobes for pathogens,11 this is the first time that
MGNPs have been used for bacterium detection and decontamina-
Supporting Information Available: Procedures for MGNP fab-
rication and characterization; procedures for pathogen detection and
differentiation; selected NMR spectra. This material is available free
of charge via the Internet at http://pubs.acs.org.
tion.1
2,13
The capture efficiency using MGNP 3 is much higher than the
0∼30% range typically observed with antibody or lectin func-
1
14,15
tionalized magnetic particles,
which can be difficult to fabricate
because of challenges in immobilizing biomacromolecules.15 Fur-
References
thermore, the orientation of the antibody/lectin on a NP surface is
(1) (a) Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2004, 42, 6042. (b) de la
Fuente, J. M.; Penades, S. Biochim. Biophys. Acta 2006, 1760, 636.
difficult to control, which may affect their binding capacities.16
(
(
2) Lingwood, C. A. Curr. Opin. Chem. Biol. 1998, 2, 695.
3) Gu, H.; Xu, K.; Xu, C.; Xu, B. Chem. Commun. 2006, 941.
To further demonstrate the advantages of MGNPs, we explore
the possibility of bacterium differentiation. It is known that several
bacteria may bind with the same carbohydrate albeit with various
affinities.17 This provides a unique opportunity to use a MGNP
array system, where the selective binding of a microbe to various
carbohydrates will lead to different responses. The resulting
characteristic response patterns18 will then allow differentiation of
bacteria. As a proof-of-principle, we synthesized galactose (Gal)
functionalized MGNP 4 in a similar manner as Man-MGNP 3 and
investigated the usage of these two MGNPs to rapidly detect and
differentiate three E. coli strains: ORN178, ORN208 a mutant strain
with greatly reduced mannose binding affinity,19 and an environ-
mental strain (ES) isolated from Lake Erie with unknown carbo-
hydrate binding specificity. While 65% of ORN178 was captured
by MGNP 3, only 15% was caught by MGNP 4 (Table S3). The
mutant strain ORN208 can still be trapped by both MGNPs,
although at lower levels. With the ES strain, high capture efficien-
cies of 70% and 75% were achieved by MGNPs 3 and 4,
respectively, suggesting its strong binding with both mannose and
galactose.
(4) Zhao, X.; Hilliard, L. R.; Mechery, S. J.; Wang, Y.; Bagwe, R. P.; Jin,
S.; Tan, W. Proc. Nat. Acad. Sci., U.S.A. 2004, 101, 15027.
(
5) Lin, P.-C.; Ueng, S.-H.; Yu, S.-C.; Jan, M.-D.; Adak, A. K.; Yu, C.-C.;
Lin, C.-C. Org. Lett. 2007, 9, 2131.
(6) For a random absorption approach, see: Hor a´ k, D.; Babi cˇ , M.; Jendelov a´ ,
P.; Herynek, V.; Trchov a´ , M.; Pientka, Z.; Pollert, E.; H a´ jek, M.; Sykov a´ ,
E. Bioconjugate Chem. 2007, 18, 635.
(
7) Kell, A. J.; Simard, B. Chem. Commun. 2007, 1227.
(8) Lis, H.; Sharon, N. Chem. ReV. 1998, 98, 637.
9) Jelinek, R.; Kolusheva, S. Chem. ReV. 2004, 104, 5987.
10) Kim, B.-S.; Yang, W.-Y.; Ryu, J.-H.; Yoo, Y.-S.; Lee, M. Chem. Commun.
005, 2035 and references cited therein.
(
(
2
(11) Schofield, C. L.; Field, R. A.; Russell, D. A. Anal. Chem. 2007, 79, 1356
and references cited therein.
(
12) For recent examples of other approaches using mannose for pathogen
detection, see refs 10, 13, and the following: Shen, Z.; Huang, M.; Xiao,
C.; Zhang, Y.; Zeng, X.; Wang, P. G. Anal. Chem. 2007, 79, 2312.
13) Disney, M. D.; Seeberger, P. H. Chem. Biol. 2004, 11, 1701.
14) Chang, S.-C.; Adriaens, P. EnV. Eng. Sci. 2007, 24, 58.
15) Porter, J.; Robinson, J.; Pickup, R.; Edwards, C. J. Appl. Microbiol. 1998,
84, 722.
(
(
(
(
16) Luk, Y.-Y.; Tingey, M. L.; Dickson, K. A.; Raines, R. T.; Abbott, N. L.
J. Am. Chem. Soc. 2004, 126, 9024.
(17) Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int. Ed.
998, 37, 2754.
1
(
18) (a) Wright, A. T.; Anslyn, E. V. Chem. Soc. ReV. 2006, 35, 14. (b) Albert,
K. J.; Lewis, N. S.; Schauer, C. L.; Sotzing, G. A.; Stitzel, S. E.; Vaid, T.
P.; Walt, D. R. Chem. ReV. 2000, 100, 2595. (c) For a recent application
of pattern recognition in array sensing, see: Miranda, O. R.; You, C.-C.;
Phillips, R.; Kim, I.-B.; Ghosh, P. S.; Bunz, U. H. F.; Rotello, V. M. J.
Am. Chem. Soc. 2007, 129, 9856.
The response patterns of the three E. coli strains to Man-MGNP
3
and Gal-MGNP 4 allow us to easily determine the microbial
identity: ORN178 (Man strong, Gal weak), ORN208 (Man weak,
Gal weak), and ES (Man strong, Gal strong) (Figure 3). The ability
to distinguish pathogen strains can have clinical applications since
the virulence of many pathogens can be correlated with carbohydrate
binding specificity.20 Moreover, the nondestructive nature of MGNP
binding can allow the concentration and recovery of pathogens and
further analysis by other techniques.1
(
(
(
19) Harris, S. L.; Spears, P. A.; Havell, E. A.; Hamrick, T. S.; Horton, J. R.;
Orndorff, P. E. J. Bacteriol. 2001, 183, 4099.
20) Stevens, J.; Blixt, O.; Tumpey, T. M.; Taubenberger, J. K.; Paulson, J.
C.; Wilson, I. A. Science 2006, 312, 404.
21) Tang, Y.-W.; Procop, G. W.; Persing, D. H. Clin. Chem. 1997, 43, 2021.
3,21
JA076086E
J. AM. CHEM. SOC.
9
VOL. 129, NO. 44, 2007 13393