Heparin-immobilized gold-assisted controlled release of growth factors via electrochemical modulation

Abstract

We developed a versatile platform for the electrochemical release of growth factors assisted by a heparin-immobilized gold surface. The controlled release of basic fibroblast growth factor (bFGF) from heparinized gold could be effectively modulated by biphasic electrochemical stimulation which actively controlled specific interactions between heparin and bFGF in a remote manner so that the released bFGF maintained its bioactivity.

---

Controlled release via external stimuli has been actively studied for the delivery of active biomolecules such as proteins, DNA, peptides, and drugs. Common external stimuli include pH changes,^1,2 photoactivation,³ temperature changes,⁴ magnetic modulation,⁵ and electrochemical stimulation.⁶ Among them, electrochemical stimulation can be employed only at a designated region with a predetermined time profile, allowing for spatio-temporal modulation. In addition, topical electrochemical stimulation typically does not hamper the biological activities of biomolecules, cells, and organs.⁷ Therefore, the electrochemical activation within a certain range of voltage and current can be considered relatively biocompatible.

Thompson et al. reported electrochemically controlled release of charged molecules from conducting polymer-modified electrodes with two distinctive redox states, where one state was suitable for binding to target molecules and the other state accelerated their release.⁸ However, the uncertain biocompatibility of synthetic conducting polymers might limit their applications. In another case, the electrochemical desorption of thiols from gold by applying a negative potential⁹ was demonstrated for the controlled release of immobilized materials (e.g., biomolecules,¹⁰ channel compounds,¹¹ DNAs¹²). Fukuda et al. demonstrated the detachment of cells along with the reductive desorption of self-assembled monolayers on gold electrode.¹³ However, only one-time burst release of immobilized molecules was possible while the remaining alkane thiols along with the released molecules might cause problems in further usage. In parallel, the electrochemical modulation of polyelectrolyte multilayers (PEMs) from a conductive substrate has also been applied. Schmidt et al. described the electrochemical release of drug molecules incorporated in PEMs.¹⁴ Graf et al. developed an electrochemically stimulated release system from PEM-liposomes on an electrode.¹⁵ Our group also reported the retrieval of cell-embedded bioactive microgels by electrochemical release of PEMs from micropatterned ITO.¹⁶ All of these studies were based on layer-by-layer polymer coating the conducting surface and, therefore, require tedious deposition steps. More recently, Servant et al. reported a carbon nanotube-containing electro-responsive polymer hydrogel for controlled drug release with electrical biases.¹⁷ Nonetheless, this system has an inherent issue of temperature rise via resistive heating.

Here, we report that the heparin-immobilized gold surface can serve as a repeatedly-usable platform for controlled release of bioactive growth factors via electrochemical stimulation, where biocompatible materials and simple surface immobilization were employed. The electrochemical release of growth factors could be effectively modulated via active control of specific interactions between heparin and growth factors without undesired side effects such as toxic polymer delamination or temperature rise during electrochemical stimulation.

Thiol-functionalized heparin (Hep-SH)¹⁸ was synthesized to allow heparin to be chemisorbed on gold surfaces and bFGF, a model growth factor, was loaded on heparin-immobilized gold (Fig. 1A). First, the contact angle of the gold surface decreased from 80° to 26° after chemisorption of hydrophilic Hep-SH, and then increased again to 50° after loading bFGF, more hydrophobic than heparin because of several hydrophobic residues.¹⁹ Additionally, cyclic voltammetry (CV) measurements showed that the chemisorption of Hep-SH significantly changed the shape of the CV curve, but bFGF loading did not affect the CV curve, indicating that no changes occurred in the electrochemical properties by physical bFGF loading (Fig. S1, ESI†). Furthermore, the chemisorption of Hep-SH on gold surfaces and the loading of bFGF on heparin-immobilized gold were quantitatively analyzed by quartz crystal microbalance (QCM) (Fig. 1B). An immediate decrease in frequency (thus immediate increase in mass) was observed upon the injection of the Hep-SH solution. Subsequently, the frequency was stabilized within 10 min, showing the fast chemisorption of Hep-SH on gold. The amount of chemisorbed Hep-SH was ∼530 ng cm⁻², and the adsorbed Hep-SH did not decrease after washing, indicating the chemisorption of Hep-SH on gold. Next, 1% BSA solution was injected. Although BSA was temporally bound on heparinized gold surfaces, it was almost completely removed after the washing step. In contrast, after injecting the bFGF solution, a fast change occurred in frequency (mass increase of ∼340 ng cm⁻²), which soon stabilized, and no further change was observed following the washing step. These results confirmed that the adsorption of bFGF to Hep-SH-immobilized gold occurred via specific heparin affinity.

Fig. 1 (A) Sequential modification of Hep-SH and bFGF on gold and change in contact angle at each surface (B) mass () and frequency () change by the sequential addition of Hep-SH, BSA, and bFGF on gold. There were washing steps with PBS between each addition.

For the electrochemical release of bFGF, various types of biphasic electrochemical stimulation (ES) (Fig. S2, ESI†) were applied and characterized within the range of non-toxic damage and corrosion of the electrode.⁷ The cumulative amounts and release profiles of bFGF for 3 h were compared and analyzed by enzyme-linked immunosorbent assay (ELISA) (Fig. 2). The released amount of bFGF increased significantly upon controlled current at 20/−300 μA, 20/−160 μA, and 20/−100 μA, or controlled voltage at 0.2/−1.2 V and 0.2/−0.8 V compared with the control group with no ES. Current control enabled more efficient electrochemical bFGF release than voltage control did because of the charge-accumulating efficacy at the working electrode (Fig. S2 and S3, ESI†). Note that Jefferys et al. also reported that the ES efficacy of voltage control was lower than that of current control because of the additional voltage drop.⁷

Fig. 2 Comparison of cumulative amounts and release profiles of bFGF released from heparinized gold electrode by various electrochemical stimulations.

Especially, upon ES at 20/−300 μA, over 60% of the loaded bFGF was released initially, followed by a slow release. However, with the application of 20/−160 μA or 20/−100 μA, a significant and continuous release of bFGF was observed after the initial bursts of 13% and 9%, respectively. Therefore, it was possible to modulate the release of bFGF (burst release and sustained release) loaded on heparin-immobilized gold surfaces by controlling the current during ES. To characterize the release mechanism of bFGF, we selected two types of ES (20/−300 μA and 20/−100 μA) and carried out the detailed electrochemical analysis.

We hypothesized that the ES at 20/−300 μA caused a strong reductive potential to induce the desorption of chemisorbed Hep-SH from the gold surface and the subsequent release of large amounts bFGF and Hep-SH complexes with a large initial burst. In contrast, the ES at 20/−100 μA was not large enough to induce the desorption of Hep-SH from the gold surface. Therefore, the main mechanism of continuous and stimulated release of bFGF by ES at 20/−100 μA cannot be attributed to the desorption of Hep-SH. Alternatively, this could be caused by the effect of counter-ion migration, driven by the accumulation of negative charges on gold surfaces, which could induce an electrical double layer with high salt concentrations (e.g., K⁺, Na⁺) near the adsorbed heparin on the gold surface. Then, it could provide a driving force for the dissociation of bound growth factors, meaning the disruption of specific interactions between Hep-SH and bFGF (Fig. 3A). To prove our hypothesis, the changes in the mass and frequency of heparinized gold after loading bFGF or heparinized gold alone were monitored upon ES using an electrochemical QCM (EQCM).

Fig. 3 (A) Schematic comparison of bFGF release by two ES at 20/−300 μA and 20/−100 μA. Mass and frequency change from both bFGF and Hep-SH-modified electrode () and electrode modified by Hep-SH only () after application of 20/−300 μA (B) and 20/−100 μA (C), respectively. Error bars: standard deviation with n = 3.

As shown in Fig. 1B, the amount of Hep-SH absorbed on the gold surface was ∼530 ng cm⁻², and bFGF was further immobilized on the heparinized gold surface at ∼870 ng cm⁻². First, in the case of bFGF loading by ES at 20/−300 μA, the mass was changed by a large amount (∼750 ng cm⁻² for 3 h) with over 50% initial burst of loaded bFGF and Hep-SH complexes. Then, by ES at 20/−300 μA to the heparinized gold surface, most of immobilized Hep-SH was released with over 50% initial burst (∼450 ng cm⁻² for 3 h), indicating the desorption of chemisorbed Hep-SH from the gold surface (Fig. 3B). In contrast, by ES of loaded bFGF and Hep-SH complexes at 20/−100 μA, a significant and continuous mass change (∼150 ng cm⁻² for 3 h) was observed with a small initial burst. However, by ES at 20/−100 μA to the heparinized gold surface, only a small amount of Hep-SH (≪50 ng cm⁻² for 3 h) was released (Fig. 3C).

These results confirmed that the release of bFGF bound to heparin-immobilized gold was remarkably increased in response to the applied electrochemical stimulation via different types of mechanisms (desorption of thiols from gold and modulation of specific interaction between heparin and bFGF).

The amount of electrochemically released bFGF at 20/−100 μA was also compared using EQCM-D and ELISA (Fig. S4, ESI†). The amount of bFGF released was estimated using EQCM-D by subtracting the small amount of Hep-SH released from the surface with Hep-SH alone from the total amount of bFGF and Hep-SH released from the bFGF-loaded heparinized surface (Fig. 3C). The release profile of bFGF estimated by EQCM-D was similar to that evaluated by ELISA.

Long-term (16 h) release profiles were also characterized (Fig. 4A). Upon ES at 20/−300 μA, a large initial burst (∼60%) of bFGF was observed within 10 min, followed by a slow release (up to 90% for 16 h). Upon ES at 20/−100 μA, the bFGF release profile was characterized by a small initial burst of ∼9%, followed by continuous and sustained release (40% for 16 h). By turning-on and turning-off of ES at 20/−100 μA, the abrupt control of release rate was also achieved (Fig. 4B). ES at 20/−100 μA was applied for 200 min with a 100 min cessation. The release rate was dramatically reduced during the off state, but increased again by turning the ES on. The cytotoxic effect of ES was characterized by live/dead double staining and comparing cell proliferation rate. After applying ES at 20/−100 μA for 1 h, both cases (w/o and w/ES) showed over 95% viability. Cell proliferation rate was also same in both cases, confirming no cytotoxicity of the ES used for bFGF release (Fig. 4C). Shi et al. also reported that directly applied current was safe up to 250 μA for fibroblast cells on gold.²⁰ The bioactivity of electrochemically released bFGF was also analysed (Fig. 4D) by bFGF-dependent cell proliferation assay.²¹ Balb/c 3T3 fibroblast cells were cultured in a media supplemented with the same concentration (10 ng mL⁻¹) of pristine bFGF or bFGF released by ES. The degree of enhanced cell proliferation was similar in both cases, showing no deteriorating effect of ES-released bFGF on proliferation of cells, at least in the tested range.

Fig. 4 (A) Long-term release profile of bFGF loaded on heparinized gold via the application of 20/−300 μA (), 20/−100 μA (), and without any stimulation () for 16 h. (B) Controlled release profiles of bFGF from the same electrode by turning-on and -off of electrochemical stimulation (ES) at 20/−100 μA (). (C) Comparison of cell viability and cell proliferation between w/o (control) and w/ES. Scale bar: 50 μm. (D) Bioactivity of bFGF released by ES. Balb/c 3T3 cells were treated with pristine bFGF (10 ng mL⁻¹) or electrochemically released bFGF (10 ng mL⁻¹) and incubated for 2 days. Error bars: standard deviation with n = 3, NS: not significant, *: p < 0.05.

In summary, we developed a versatile platform for electrochemically-controlled release of bFGF using heparin-immobilized gold where heparin was chemisorbed on gold surfaces after thiolation. Spontaneous loading of growth factors onto heparin-tethered gold electrodes was also achieved via heparin's affinity to bFGF. The on/off switchable and long-term release of bFGF was enabled with the application of proper electrochemical stimulation. The ES at 20/−100 μA did not cause any cytotoxic effect on cells, and the released growth factor from heparin-immobilized gold upon the ES maintained its bioactivity. We envision that our electrochemically-modulated release of growth factors from Hep-SH-chemisorbed gold surfaces may find promising applications in various biomedical fields including drug delivery, biosensors, and cell culture in conjunction with micro-patterning techniques.

Acknowledgements

Partial financial support was provided by the WCU program at GIST by NRF, MEST, Korea (R31-2008-000-10026-0), and by the “GIST-Caltech Research Collaboration Project” through a grant provided by GIST, Korea. Also, this work was supported by the GIST Research Institute (GRI) in 2016.

Notes and references

1. G. Hui, M. Yanan, L. Xueyou, L. Yongri, C. Baoquan and M. Jianbiao, Eur. Polym. J., 2011, 47, 1232–1239.
2. B. J. Kim, K. H. Min, G. H. Hwang, H. J. Lee, S. Y. Jeong, E. C. Kim and S. C. Lee, Macromol. Res., 2015, 23, 111–117.
3. Y. Huang, R. Dong, X. Zhu and D. Yan, Soft Matter, 2014, 10, 6121–6138.
4. M. Caldorera-Moore and N. A. Peppas, Adv. Drug Delivery Rev., 2009, 61, 1391–1401.
5. D. S. T. Hsieh, R. Langer and J. Folkman, Proc. Natl. Acad. Sci. U. S. A., 1981, 78, 1863–1867.
6. J. Kost and R. Langer, Adv. Drug Delivery Rev., 2001, 46, 125–148.
7. D. R. Merrill, M. Bikson and J. G. R. Jefferys, J. Neurosci. Methods, 2005, 141, 171–198.
8. B. C. Thompson, R. T. Richardson, S. E. Moulton, A. J. Evans, S. O'Leary, G. M. Clark and G. G. Wallace, J. Controlled Release, 2010, 141, 161–167.
9. N. S. Pesika, K. J. Stebe and P. C. Searson, Langmuir, 2006, 22, 3474–3476.
10. P. Mali, N. Bhattacharjee and P. C. Searson, Nano Lett., 2006, 6, 1250–1253.
11. M. B. Buchmann, T. M. Fyles and T. Sutherland, Bioorg. Med. Chem., 2004, 12, 1315–1324.
12. D. Zheng, X. Li and J. Ye, Bioelectrochemistry, 2009, 74, 240–245.
13. J. Fukuda, Y. Kameoka and H. Suzuki, Biomaterials, 2011, 32, 6663–6669.
14. D. J. Schmidt, J. S. Moskowitz and P. T. Hammond, Chem. Mater., 2010, 22, 6416–6425.
15. N. Graf, A. Tanno, A. Dochter, N. Rothfuchs, J. Vörös and T. Zambelli, Soft Matter, 2012, 8, 3641–3648.
16. K. Gwon, M. Kim and G. Tae, Integr. Biol., 2014, 6, 596–602.
17. A. Servant, V. Leon, D. Jasim, L. Methven, P. Limousin, E. V. Fernandez-Pacheco, M. Prato and K. Kostarelos, Adv. Healthcare Mater., 2014, 3, 1334–1343.
18. G. Tae, Y. J. Kim, W. I. Choi, M. Kim, P. S. Stayton and A. S. Hoffman, Biomacromolecules, 2007, 8, 1979–1986.
19. H. Zhu, K. Ramnarayan, P. Menzel, Y. Miao, J. Zheng and S. Mong, Protein Eng., 1998, 11, 937–940.
20. G. Shi, M. Rouabhia, S. Meng and Z. Zhang, J. Biomed. Mater. Res., Part A, 2008, 84, 1026–1037.
21. M. S. Kim, Y. M. Shin, J. Lee, S. I. Kim, Y. S. Nam, C. S. Shin and H. Shin, Macromol. Biosci., 2011, 11, 122–130.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra18908c

‡ Equally contributing authors.

---