87

SEMICONDUCTOR QUANTUM DOTS AS HIGHLY EFFECTIVE BIOSENSING TOOLS

Raluca Buiculescu1, Nikos A Chaniotakis1,2

1Department of Chemistry, University of Crete, Iraklion, 71003 Crete, GREECE

2Chemistry Department, TUFTS University, Medford MA 02556 Web: http://www.analytical-chemistry.uoc.gr Fax: +30 2810 545165; Tel: +30 2810 545018

Abstract–In this paper, highly luminescent core/shell CdSe/ZnS quantum dots have been used in the construction of two sensing systems for the detection of oligonucleotide hybridization and enzymatic reaction by covalent binding the biomolecules directly to the surface of the quantum dots. It is shown that the oligonucleotide-modified quantum dots have proven to be able to detect large sequences of mismatched bases through the hybridization process, while the system based on biosilica encapsulated enzyme-modified quantum dots was found to be suitable for monitoring low substrate concentrations in solution and with a storage lifetime of more than 2 months.

1. INTRODUCTION

Quantum dots (QDs) are colloidal semiconductor nanocrystals with sizes in the range of nanometers that possess unique optoelectronic properties due to quantum confinement effects. Their broad excitation and narrow size-tunable emission spectrum, negligible photobleaching, and high photochemical stability [1] render them suitable for use in a large number of bioanalytical applications. In addition, their surface can be modified by conjugation with a wide range of functional molecules (e.g., small biomolecules, ions, nucleic acids or proteins) thus expanding their use in a wide variety of optical sensing experiments [2].

The aim of the present study was to develop simple photoluminescent biosensor systems for the detection of oligonucleotide hybridization and enzymatic reaction. In order to achieve this goal, the oligonucleotide and enzyme of choice were covalently immobilized onto the surface of highly luminescent core/shell semiconductor CdSe/ZnS quantum dots. The QD-oligonucleotide conjugates were subsequently used for hybridization experiments, while the QD-enzyme conjugates

were subjected to encapsulation into a bioinspired silica shell for increasing the stability of the system. The pores of the grafted silica matrix allow the transport of the substrate to the entrapped QD-enzyme conjugates and its subsequent conversion to products.

2. METHODS

CdSe/ZnS QDs have been conjugated with the Mycobacterium tuberculosis 16S rDNA oligonucleotide [3] 5’ - CGT CAT GAA AGT CGG TAA CAC CCG AAG CCA - 3’or with the Acetylcholinesterase (AChE) enzyme, through the coupling reaction of N-ethyl-N'-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysulfosuccinimide sodium salt (sulfo-NHS). The obtained conjugates were separated from any possible unbound oligonucleotide/enzyme or unconjugated QDs by centrifugation/washing (5000 rpm, 60 min) with buffer on Millipore Amicon Ultra filters with 10,000 and 100,000 respectively molecular weight cutoff (MWCO). The conjugates stock solutions were stored at 4oC. The QD-oligonucleotide conjugates were further subjected to hybridization with complementary target (named 100%) and with targets with 1, 2, 6 or 12 mismatched bases (named 1m, 2m, 6m and 12m respectively) by heating for 5 min at 95oC. For the bio-inspired silicification reaction of the QD-enzyme conjugates, a poly-L-lysine (PLL) solution was added to the conjugates in order to produce a large number of amine groups on their surface which facilitates the biosilicification process. Subsequently a freshly prepared solution of silicic acid was added. After mixing in a vortex for approx 4 hours at room temperature, the nano-complexes were cleaned by successive centrifugation/wash cycles. The cleaned silica

978-1-4673-0738-3/12/$31.00 © 2012 IEEE

88

biomimetic composites were stored at 4oC until use.

3. RESULTS AND DISCUSSION

3.1. QD-based Oligonucleotide Detection

For the QD-oligonucleotide experiments we have chosen oligonucleotides from Mycobacterium tuberculosis, a pathogenic bacterium with unusual high lipid content in its wall. This makes the cells impervious to Gram staining and thus difficult to identify and diagnose with standard staining methods.

In Figure 1 the fluorescence signal of the QD-oligonucleotide conjugates before and after hybridization with complementary and mismatched targets is seen. It can be observed that formation of the double stranded DNA sequence, results in an enhancement of the fluorescence fingerprint.

600 620 640 660 680 7000

1

2

3

4

5

6

7

8

Fluo

resc

ence

inte

nsity

Wavelength, nm

QD-probe QD-probe+Target 100% QD-probe+Target 12m

Fig. 1. QD-oligonucleotide probe before (solid) and after hybridization with 100% (dash) and 12m (dot) targets.

The increase in the fluorescence signal

intensity was obvious and of the same order of magnitude for the 100%, 1m, 2m and 6m target oligonucleotides. As far as the 12m target is concerned, an increase of the fluorescence signal can also be observed, but to a smaller extent. When heating the bare QDs at 95oC for 5 minutes (data not presented), their fluorescence signal also records an increase, in the same order of magnitude as the QD-probe. The fact that an increase in the fluorescence signal was also recorded in the case of the bare QDs, suggests that this phenomenon is due to modifications that appear in the QDs structure and not as a result of conjugation with the oligonucleotide. As known, the photoluminescence of core/shell quantum dots is indeed temperature dependent,

and the increase in the QDs fluorescence upon heating has been attributed to delocalization of the carriers localized at the core – shell interface states induced by atom diffusion or lattice strain. [4]

These results indicate that our system which is composed of highly luminescent core/shell CdSe/ZnS quantum dots conjugated with oligonucleotides is a suitable methodology for the identification of bacterial DNA in solution. Based on this method, we should be able to detect any bacterium that can not be identified with standard staining methods. 3.2. QD-based Sensor for the Detection

of Enzymatic Reaction

The efficient entrapment of QD-AChE conjugates within the silica architecture was examined by ATR-FT-IR spectroscopy.

Fig. 2. ATR-FT-IR spectra of bio-inspired silica nanocomposites: PLL-silica, AChE-PLL-silica

and QD-AChE-PLL-silica.

The spectrum of the QD-AChE-PLL-silica nanocomposites was compared to those of PLL-silica nanocomposites and of AChE-PLL-silica nanocomposites. The ATR-FT-IR spectra present the characteristic peaks of the amide I and amide II bands in the region 1400–1700 cm-1 as well as the biosilica peaks (A, B and C), in the region 760–1300 cm-1 (Figure 2), proving the successful formation of the silica structure in all cases.

The amide I and II vibrations of the polypeptide backbone are very sensitive to changes that might appear in the secondary conformations [5, 6] and thus they are closely monitored. As can be seen from Figure 3, the

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm-1)

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm )

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm-1)

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm )

% R

efle

ctan

ce

I IIamide

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm-1)

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm )

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm-1)

% R

efle

ctan

ce

I IIamide

A

B

C

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

100

90

95

85

80

70

75

60

65

551600 1400 1200 1000 800

Wavenumbers (cm )

% R

efle

ctan

ce

I IIamide

A

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

QD/AChE /PLL/silicaAChE /PLL/silicaPLL/silica

89

amide I peak of all poly-L-lysine templated silica nanocomposites (PLL-silica, AChE-PLL-silica and QD-AChE-PLL-silica respectively) is shifted in comparison to the ones of the solid poly-L-lysine and that of the enzyme AChE. Among these, the QD-AChE-PLL-silica nanocomposites display the highest amide I shift toward lower wavenumbers confirming the interaction of poly-L-lysine and protein residues with the biosilica network.

Fig. 3. Amide I band of AChE, PLL and bio-inspired silica nanocomposites (PLL-silica, AChE-PLL-silica

and QD-AChE-PLL-silica).

The monitoring of an enzymatic reaction usually takes place by tracing its products. When these products are either bases, or acids, a change in the solution pH is observed. The enzyme used in this study, AChE, hydrolyses the substrate acetylcholine producing choline and acetic acid. The surface of the QDs is covered with mercaptoundecanoic acid chains which are pH sensitive.

The response of the QD-AChE-PLL-silica nanocomposites to the substrate acetylcholine chloride was evaluated by fluorescence means. Upon addition of substrate, the acetic acid produced by the enzymatic reaction lowered the pH of the surrounding environment of the QDs, which lead to the quenching of their photoluminescence. As a result, the higher the substrate concentration is, the higher the quantity of acetic acid to be produced and so the larger the decrease in the observed photoluminescence.

Figure 4 shows the response of the QDs based system to a large scale of substrate concentrations and a linear relationship between the concentration of the substrate and

the enzyme activity has been found in the range of 100 to 1000 μM (detailed in the inset) with a detection limit of 1μM.

0.0 2.0x103 4.0x103 6.0x103 8.0x103 1.0x1040.0

0.1

0.2

0.3

0.4

0.5

0 200 400 600 800 10000.15

0.20

0.25

0.30

0.35

0.40

Enzy

me

activ

ity

[AChCl], nM

Enzy

mat

ic a

ctiv

ity

[Acetylcholine chloride], ì M

Fig. 4. Calibration curve of the QD/AChE/PLL/silica biosensor. The inset represents the linear

range of response.

The stabilizing effect that the biomimetically synthesized silica has on the QD-AChE complex was evaluated (Figure 5), as compared to the free complex in solution. The QD-AChE-PLL-silica nanocomposites present a far better time stability, with a remaining activity of 65% after 45 days of measurements, while the activity of the QD-AChE conjugates not encapsulated into bio-silica reached 50% of their initial activity after only 20 days of measurements. It is thus evident that the biomimetically synthesized silica provides an environment within which the relatively unstable AChE is stabilized.

0 10 20 30 40 5040

50

60

70

80

90

100

110

Rem

aini

ng a

ctiv

ity, %

Time, days

QD-AChE-silica QD-AChE

Fig. 5. Stability study of the QD/AChE entraped into the bio-inspired silica nanocomposites compared to the

QD/AChE complex free in solution.

4. CONCLUSIONS

From the data presented we can conclude that semiconductor quantum dots are useful

100

90

95

85

80

70

75

60

65

55

45

50

40

35

1700 1650 1600

Wavenumbers (cm-1)

% R

efle

ctan

ce

B

QD/AChE/PLL/silicaAChE/PLL/silicaPLL/silicaAChEPLL

100

90

95

85

80

70

75

60

65

55

45

50

40

35

1700 1650 1600

Wavenumbers (cm-1)

% R

efle

ctan

ce

B

QD/AChE/PLL/silicaAChE/PLL/silicaPLL/silicaAChEPLL

100

90

95

85

80

70

75

60

65

55

45

50

40

35

1700 1650 1600

Wavenumbers (cm-1)

% R

efle

ctan

ce

B

QD/AChE/PLL/silicaAChE/PLL/silicaPLL/silicaAChEPLL

100

90

95

85

80

70

75

60

65

55

45

50

40

35

1700 1650 1600

Wavenumbers (cm-1)

% R

efle

ctan

ce

B

QD/AChE/PLL/silicaAChE/PLL/silicaPLL/silicaAChEPLL

90

tools that can be used in the construction of novel biosensing systems. The oligonucleotide-modified quantum dots have proven to be able to detect large sequences of mismatched bases through the hybridization process, thus finding possible applications in detecting mutations of DNA sequences. The exact mechanisms that lye behind this phenomenon are still not well understood. Further experiments will be undertaken leading to a better understanding of the phenomenon and improving of the method. In addition, these QDs were the basis for the development of a new optical biosensor system for the detection of enzymatic activity. The enzyme was successfully immobilized onto the QDs and then stabilized by formation of the outer nanoporous silica thin shell. We have shown that the pores of the bio-inspired silica nanocomposites do not impede the transport of the substrate to the enzyme and the optically active biosensor was found to be suitable for monitoring low substrate concentrations in solution. The biodetection system proposed is shown to be stable with storage lifetime of more than 2 months. These QD based biosensors can provide stable and sensitive nano-biosensor platforms for the detection of the enzymatic activity.

Acknowledgments–This work was supported by the program NANOMYC (Contract No. 036812) of the European Commission and the GGET11 ROM23_4_ET29. References [1] M.F. Frasco and N. Chaniotakis, “Semiconductor

Quantum Dots in Chemical Sensors and Biosensors”, Sensors, 9, pp. 7266–7286, 2009.

[2] R. Gill, M. Zayats, I. Willner, “Semiconductor quantum dots for bioanalysis”, Angew. Chem. Int. Ed., 47, pp. 7602–7625, 2008.

[3] R. Frothingham, H.G. Hills and K.H. Wilson, Extensive DNA sequence conservation throughout the Mycobacterium tuberculosis complex, J. Clin. Microbiol. 1994, 32(7), 1639–1643.

[4] P. Jing, J. Zheng, M. Ikezawa, X. Liu, S. Lv, X. Kong, J. Zhao, Y. Masumoto, “Temperature-dependent photoluminescence of CdSe-Core CdS/CdZnS/ZnS-multishell quantum dots”, J. Phys. Chem. C, 113, pp. 13545–13550, 2009.

[5] S. Krimm, J. Bandekar, “Vibrational spectroscopy and conformation of peptides, polypeptides and proteins”, Adv. Prot. Chem., 38, pp. 181–364, 1986.

[6] J. Bandekar, “Amide modes and protein conformation”, Biochim. Biophys. Acta, 1120, pp. 123–143, 1992.