Cross-Linker Mediated Biofunctionalization of Single Wall Carbon Nanotubes with Glucose Oxidase

Biosensors are among the most anticipated devices for the development of advanced diagnostic tools to meet the current challenges in biomedical research [1, 2]. A number of physical and biological materials such as gold nanoparticles and protein, [3-8] are potential candidates to meet the desired need of research and development. Since the discovery of carbon nanotubes (CNTs) by Iijima in 1991 [9], due to their high aspect ratio, CNTs were expected to possess interesting electronic, mechanical and molecular properties. The ascertainment of the nanoscale properties of carbon nanotubes has impelled their progression in the biosensor field [7, 10]. Functionalized CNTs can act as potent surfaces for linking of a variety of important biomolecules such as peptides, proteins and nucleic acids [11, 6]. Functionalized carbon nanotubes exhibit improved properties with respect to solubility, ease of dispersion and cytotoxicity. Hence they can be easily manipulated and processed for various applications. Enzyme functionalized single walled carbon nanotubes (SWCNTs) provide a basis for constructing biosensors, biomedical devices and bioreactors [12, 13].

The conducting nature of SWCNTs can be successfully exploited for bioanalytical applications by coupling sensing biomolecules such as enzymes to the carboxyl groups of the SWCNTs. Enzymes have been linked to SWCNTs via diimide activated amidation and through non-covalent adsorption [14, 15]. In order to functionalize SWCNTs with enzymes; they are subjected to various chemical reactions (carboxylation, acylation and amidation) and physical processes (sonication, washing, filtering, centrifugation, drying). The multiple steps involved in functionalization of SWCNTs, leads to structural aberration in SWCNTs and also there is considerable loss of CNTs at each step. However, some of the above mentioned physical and chemical steps are unavoidable; a simpler and diminutive chemical reaction procedure for bio-molecules immobilization would be of immense importance.

The present study demonstrates a simple two step reaction for the covalent attachment of glucose oxidase (GOD) on SWCNTs. Unlike conventional procedures, the present procedure only involves carboxylation and amidation. The carboxylation of SWCNTs was achieved by a sonochemical method and eventually the amidation of SWCNTs was done with GOD via cross-linkers 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) & N-hydrosuccinimide (NHS). Carboxylation. The amidation of SWCNT was characterized by Fourier transform infra red spectroscopy (FTIR), ultraviolet-visible spectroscopy (UV-Vis), atomic force microscopy (AFM) and elemental analysis. Biofunctionalization of SWCNT with GOD further opens possibilities for their applications in biosensor field and chemically modulated nanoelectronic based device development.

Materials and Methods

SWCNTs prepared by LASER ablation, were procured from Nanostructures and Amorphous Materials, USA. The length, diameter and purity of the SWCNTs were 0.5-100 µm, 1-2 nm and 90%, respectively. The SWCNTs were used without any further purification. GOD was procured from Sigma USA. Thionyl chloride (SOCl₂) and N, N-dimethyl formamide (DMF) were procured from Sigma, Germany. EDC and NHS were purchased from Pierce, USA. Doubled distilled deionized water obtained from ELGA PURELAB Ultra purification system with resistivity >18 MO was used throughout the course of study.

In the present study biofunctionalization of SWCNT with GOD was achieved by both two-step and conventional approaches for comparison.

Two Step Approach

Immobilization of GOD on surface of SWCNT involves carboxylation and amidation through cross linkers as shown in figure 1.

Figure 1. Biofunctionalization of SWCNTs by carboxylation and amidation employing cross-linkers.

Carboxylation

Pristine SWCNTs (30 mg) were sonicated in 3:1 solution of concentrated H2SO4 and HNO3 at room temperature for 1-2 hours. Then 1 M HCl was added to the mixture and was again sonicated for 30 minutes.

Carboxylated SWCNTs were filtered and washed thoroughly with deionized water and were finally air dried.

Amidation

Carboxylated SWCNTs (20 mg) were sonicated in DMF for 30 minutes. NHS (35 mg) and EDC (58 mg) were added and stirred with sonicated SWCNTs for 24 hrs at 37°C. GOD (35 mg) was then added to the above mixture which was then kept for incubation at room temperature for five days with constant stirring. Then subsequently the mixture was washed with DMF, phosphate buffer (pH 7.4) and deionized water to remove unreacted GOD. Finally product was dried overnight in desiccator.

Conventional Approach

The conventional procedure for the immobilization of GOD on SWCNT surface involves carboxylation, acylation and amidation as demonstrated in figure 2. The same carboxylation procedure was adopted as was followed in the two step approach.

Figure 2. Biofunctionalization of SWCNTs by carboxylation, acylation & amidation.

Acylation

Carboxylated SWCNTs (20 mg) were stirred in 15 ml of 20:1 mixture of thionyl chloride and DMF at 700°C for 24 hrs. Acyl-chlorinated SWCNT were then filtered and washed with anhydrous THF and dried under vacuum at room temperature for 20 mins.

Amidation

Acyl-chlorinated SWCNTs were reacted with GOD in DMF at room temperature for a period of five days. The reaction mixture was then washed subsequently with DMF, phosphate buffer (pH 7.4) and deionized water to remove unreacted GOD. The product obtained was dried overnight in a desiccator.

Characterization

The chemical functionalization was identified by FTIR spectra (Thermo Electric Corp. Nicolet Model 470). UV-Vis absorption spectra were recorded on Perkin Elmer Lambda 15 spectrometer; in the range of 170-650 nm for all solutions. Elemental analysis was done by gravimetry method on Flash EA 1112 Series (Thermo Electron Corporation) elemental analyzer. An AFM was used in noncontact mode (Veeco USA) for obtaining topological views.

Enzyme Activity Measurement

The reaction mixture was prepared by adding 0.50 ml glucose (10 % w/v) and 2.40 ml o-dianisidine dihydrochloride solution in 0.1 M potassium phosphate buffer (pH 7.0). Peroxidase (0.1 ml) was added to the solution, mixed and left to equilibrate at 25°C for 3-5 minutes. Biofunctionalized SWCNTs having linked GOD (0.10 ml) were added to the final solution.

Results and Discussions

FTIR Spectra

The FTIR spectra of carboxylated, acylated and amidated SWCNTs obtained for conventional and two-step approaches are shown in Figure 3 [16, 17]. FTIR spectra of pristine SWCNT and pure GOD were recorded as controls (figure 3a & 3b). FTIR spectra of pristine SWCNT show characteristic peaks at 3420.7 cm^-1 for hydroxyl group (-OH), 1560.1 cm^-1 for nitro group (- NO₂), 1375.7 cm^-1 for -C=C bond and 1051.7 cm^-1 for -C-C bond (figure 3a). FTIR spectra of GOD shows prominent peaks at 3402.2 cm^-1 for -N-H stretching, 2927.4 cm^-1 for C-H stretching, 1654.5 cm^-1 for NH⁺₃ asymmetric bending, 1544.7 cm^-1 for N-H bending and 1220.4 cm^-1 for C-N stretching (figure 3b). FTIR spectra of carboxylated and acylated SWCNT shows peaks at 2913.0 cm^-1 for -OH stretching of -COOH group (figure 3c) and 3425.1 cm^-1 for -OH stretching (figure 3d) due to moisture as evidenced in the pristine SWCNT spectra (figure 3a). In carboxylated and acylated SWCNTs spectra, peaks were observed at 1467 and 1400.2 cm^-1, respectively for C-H bending (figure 3c & 3d). The peak at 1700.1 cm^-1 in acylated SWCNTs spectra, clearly signifies the -C=O stretching vibration of the COCl group (figure 3d). A prominent peak at 616.2 cm^-1 was observed in the acylated SWCNT spectra (figure 3d) which is attributed to C-Cl stretching of -COCl group but was absent in carboxylated SWCNT spectra (figure 3c) suggesting acylation of SWCNTs. FTIR spectra of amidated SWCNTs by two-step and conventional approaches shows noteworthy peaks at 3406.4 and 3373.0 cm^-1, respectively for N-H stretching due to primary amide (figure 3e & figure 3f). The peak at 2974 cm^-1 for -C-H stretching due to vibration in enzyme group was observed in amidated spectra of conventional procedure (figure 3e). Significant peaks at 1640.9 and 1634.1 cm^-1 were observed in amidation spectra of the two step method and conventional procedure, respectively which are assigned to -C=O stretching due to newly formed amide bonds (figure 3e & 3f).

Peaks for -N-H, -C-H and -C=O groups were observed in amidated spectra of both two step and conventional approaches suggesting functionalization of CNTs with GOD. A downshift of peak was observed for -C=O stretching in the amidation spectra of conventionally prepared SWCNTs at 1634.1 cm^-1 due to coupling between closely placed enzyme molecules at the end of SWCNTs. Such downshift is not evidenced in the two step approach as it avoids steric hindrance. Hence it can be cited as an advantage over the conventional approach. The peak at 876.9 cm^-1 for amidation in the conventional method corresponds to C-N stretching (figure 3e). Other peaks were also observed such as 2974 cm^-1 for -C-H stretching in enzyme molecules, 1708 cm^-1 due to free -C-O group stretching and 1564 cm^-1 for free -N-H group of the enzyme in the spectra of two step amidation method.

Figure 3. FTIR spectra of: (a) Pristine SWCNTs (b) Pure GOD (c) Carboxylated - SWCNTs (d) Acylated - SWCNTs (e) Amidated - SWCNTs (via cross-linker) (f) Amidated - SWCNTs (Conventional method)

UV-Vis Spectra

UV-Vis spectroscopy serves as a method for quantitative measurement of degree of functionalization on SWCNTs. UV-Vis spectra obtained for functionalized SWCNTs indicate disruption in one-dimensional electronic structure of SWCNT as a result of functionalization [15, 18 & 19]. The Π electrons present in the molecular orbital of the carbon atoms participate in new bond formation. Functionalization results in engagement of these electrons in bond formation hence these electrons are no longer available.

Figure 4. UV-Vis spectra of SWCNTs (a) Pristine (b) Carboxylated (c) Acylated (d) Amidated

As a result of free Π electrons, pristine SWCNTs show absorbance in the UV-Vis range. The peaks obtained at 230, 210 & 274 nm in subsequent spectra are due to functional groups acquainted on SWCNT (figure 4).

Pristine SWCNT shows a characteristic peak at 170 nm imputable to Π-Π* transitions due to Π electrons of the double bonds in the SWCNT (figure 4a). A prominent peak at 230 nm is observed for carboxylated SWCNT indicating a transition due to an unshared pair of electrons of the -C=O bond in the carbonyl group (-COO) (figure 4b). In case of acylation, there is a chloride substitution in carbonyl group, carrying a lone pair of electron resulting in Π-Π* transition. Chloride withdraws an electron from carbon due to inductive effect causing a lone pair of electron to be held more firmly. Therefore, greater energy is needed for Π-Π* transition due to which peak was shifted to shorter wavelength i.e. 210 nm (figure 4a). A significant peak at 274 nm describes amidation of SWCNTs (figure 4d). Enzymes being protein usually show absorbance at 280 nm [20] but a shift is observed at 274 nm in absorption spectra of amidated SWCNTs due to formation of new amide bond between GOD & SWCNT.

Figure 5. Histogram showing absorbance of various functionalized SWCNTs.

Elemental Analysis

Quantitative analysis for elements namely carbon (C), nitrogen (N), hydrogen (H), oxygen (O) and sulphur (S) was done to reveal SWCNT functionalization in accordance to the change in quantity of particular element during subsequent steps of functionalization.

Table 1. Table showing Elemental Analysis of SWCNTs. A) Pristine B) Carboxylated C)

Elemental analysis reveals the difference in surface chemical functionalization between pristine, carboxylated and amidated SWCNT. Pristine SWCNT shows mainly C (89.68%) and also traces of H (0.24%) and O (1.56%) which may be attributed to the presence of moisture. Besides this, small amount of N (0.58%) was also observed which might have been doped during the chemical process involved in SWCNT synthesis (Table 1A). Increase in H (0.36%) and O (3.68%) underlies carboxylation (Table 1B). Increase in N (0.75%) was also ascertained in carboxylated SWCNTs (figure 1B). No trace of sulphur was observed in either pristine or carboxylated SWCNTs. Also there was no significant change observed in carbon percentage for pristine and carboxylated SWCNTs. But for amidated SWCNTs, a decrease in carbon (50.71%) and increase in H (3.94%) and O (6.89%) was observed compared to carboxylated SWCNTs which may be due to the attachment of enzyme molecules (Table 1C). It was further confirmed by the presence of increased nitrogen (8.71 %) and sulphur (0.19 %) in amidated SWCNTs which can be attributed to the attachment of the GOD molecules.

AFM (Atomic Force Microscopy)

The height of pristine SWCNTs was measured to be 22-24 nm using the line analysis tool in the AFM for surface and modified CNTs (Fig 6). Occurrence of bundles of 10-12 nanotubes is assumed to be present considering the average diameter of single SWCNT as 2 nm. The height of carboxylated carbon nanotubes was found to be 51.3 nm, which is higher than reported for pristine SWCNT i.e. 22 nm. This could be probably due to clumped CNTs. A height of 66.0 nm was observed for acylated SWCNTs. The significant increase in the height of amidated SWCNT in conventional (177.6 nm) and two step approach (237.0 nm) suggests the presence of immobilized GOD on the SWCNT surface (Fig 7).

Figure 6. AFM topography of SWCNT: a) pristine b) carboxylated c) acylated d) amidated (conventional method) and e) amidated (two-step method).

Figure 7. Height comparison of GOD modified SWCNTs in two approaches after common modification.

GOD molecules attached height in two step approach is higher than in conventional approach as shown in figure 7. The tentative number of GOD molecules on SWCNT in the conventional and two-step approach was determined to be 17 and 23, respectively. This was ascertained by increased height of SWCNT i.e. Pristine, carboxylated and amidated SWCNT [19]. However, this deviation does not signify low efficiency of amidation as the size of the GOD molecule is 10 nm. Moreover, GOD binding efficiency was found to be more in the present approach as compared to the conventional approach. SWCNTs were found to be of rougher texture in conventional approach as compared to the present approach as determined by roughness analysis. More roughness in the conventional approach is attributed to harsh chemical treatment in subsequent steps. Such harsh treatment further suggests damage to intrinsic properties of SWCNT.

Enzyme Activity Measurement

The intensity of color produced was monitored at 436 nm in the spectrophotometer. The Glucose oxidase activity calculated for conventional method was found to be 0.11u/mg and 0.17u/mg for two step method. This experiment reveals the bioactivity of enzyme indicating that it retains its activity in immobilized state. The data also shows the better efficiency of the two step method.

Conclusions

Carbon nanotubes have shown immense potential for the development of nanoscale devices. The multistep procedure for biomolecules interfacing with SWCNTs has widely registered its application in biosensor development. Simple procedures with fewer steps can be more useful and applicable in an ever-developing scenario. This paper presents a simple and convenient two-step procedure for immobilization of GOD on SWCNT surfaces. It is inferred from FTIR studies that the present approach is as efficient as earlier reported methods for immobilization of GOD on SWCNTs. FTIR peaks at 1700.1 and 1640 cm^-1 show carboxylation and amidation, respectively. In the present investigation, a significant peak at 274 nm in UV-Vis studies in conventional and two step approaches further affirms the results of FTIR studies. Increase in nitrogen and sulphur percentage in elemental analysis during amidation in both the procedures suggests successful amidation. Topographical studies by AFM for structural changes in amidated SWCNTs clearly shows immobilization of GOD onto SWCNTs. Significant differences in the heights of amidated SWCNTs demonstrates advantage of present approach over the conventional method. Roughness analysis by AFM further signifies that the present two step methodology prevents structural damage to SWCNTs. Interfacing of biomolecules with SWCNTs by a two-step procedure may provide insights into structural and functional properties of CNTs and biomolecules for the development of advanced biosensors.

Acknowledgements

This work was supported by Department of Information technology. Authors are thankful to Mr. Ashwani Kumar for his kind assistance during the experimental work. We are also thankful to Dr. Amit Sharma and Dr. Inderpreet Kaur for their valuable guidance and suggestions.

References

1. Azamian BR, Davis JJ, Coleman KS, Bagshaw CB, Green MLH. Bioelectrochemical single-walled carbon nanotubes. J Am Chem Soc 2002; 124(43):12664-12665
2. Veetil JV, Ye K. Development of Immunosensors Using Carbon Nanotubes. Biotechnol Prog 2007; 23:517-531
3. Kaur H, Das T, Kumar R, Ajore R, Bharadwaj LM. Covalent attachment of actin filaments to Tween 80 coated polystyrene beads for cargo transportation. Article in press.
4. Bhalla V, Bajpai RP, Bharadwaj LM. DNA Electronics. EMBO Reports 2003; 4(5):442-225
5. Ajore R, Kumar R, Kaur I, Sobti RC, Bharadwaj LM. DNA immobilization chemical interference due to aggregates study by Dip and Drop approach. JBBM 2007; 70:779-785
6. Daniel S , Rao TP, Rao KS, Rani SK, Naidu GRK , Lee HY, et al. A review of DNA functionalized/grafted carbon nanotubes and their characterization. Sensors and Actuators B Chemical 2007; 122:672-682
7. Liang W, Zhuobin Y. Direct electrochemistry of Glucose Oxidase at a gold electrode modified with single-Wall carbon nanotubes. Sensors 2003; 3:544-554.
8. Kumar S, Kumar R, Jindal VK, Bharadwaj LM. Immobilization of single walled carbon nanotubes on glass surface. Materials Letters 2008; 62:731-734.
9. Iijima S. Helical microtubules of graphitic carbon. Nature 1991; 354:56-58.
10. Wang J Carbon-Nanotube Based Electrochemical Biosensors: A Review Electroanalysis 2004; 17(1): 7 - 14.
11. Kam NWS, Liu Z, Dai H. Carbon nanotubes as intracellular transporters for proteins and DNA: An investigation of the uptake mechanism and pathway. Angewandte Chemie 2006; 118: 591-595.
12. Wang SG, Zhang Q, Wang R, Yoon SF, Ahn J, Yang DJ, et al. Multi-walled carbon nanotubes for the immobilization of enzyme in glucose biosensors. Electrochemistry Communications 2003; 5:800-803.
13. Sinha N, Ma J, Yeow T J. Carbon Nanotube-Based Sensors JNN 2006; 6:573-590.
14. Robert J. Chen, Yuegang Zhang, Dunwei Wang, Hongjie Dai Noncovalent Sidewall Functionalization of Single-Walled Carbon Nanotubes for Protein Immobilization J.Chem.Soc. 2001; 123: 3838-3839.
15. Wang Y, Iqbal Z, Malhotra SV. Functionalization of carbon nanotubes with amines and enzymes. Chem Phy Letters 2005; 402:96-101.
16. Silverstein RM, Webster FX. Spectrometric identification of organic compounds. 6th edition. New York: Wiley India edition; reprint 2006:71-143.
17. Pavia DL, Lampman GM, Kriz GS. Introduction to spectroscopy-A guide for students of organic chemistry, Washington: Harcourt College publishers; reprint: 13-84, 353-389.
18. Saini RK, Chiang WI, Peng H, Smalley RE, Billups WB, Hauge RH, et al. Covalent sidewall functionalization of single wall carbon nanotubes. J Am Chem Soc 2003; 125:3617-3621.
19. Besteman K, Lee JO, Wiertz FGM, Heering HA, Dekker C. Enzyme-coated carbon nanotubes as single-molecule biosensors. Nano letters 2003; 3 (6): 727-730
20. Waddell W J. A simple UV spectrophotometric method for the determination of protein. J Lab Clin Med 1956; 48:311-314.