Barstar:barnase — a versatile platform for colloidal diamond bioconjugation

Abstract

We report on a bioconjugation platform based on a high-affinity protein pair, barstar:barnase (Bs:Bn), that provides a modular design toolkit capable of strong (covalent) and versatile assembly of bio/nanocomplexes. Luminescent nanodiamonds (140 nm) serve as the docking station to synthesize Bs:Bn bioconjugates which are characterized and utilized in several applications, including cell transfection.

---

Nanodiamonds (NDs) are tiny particles with a stable, crystalline core and chemically active surface that can dock a variety of biocompatible moieties.¹Nitrogen-vacancy (NV) color centers render nanodiamonds luminescent (LND) under ambient conditions and their strong, spin-sensitive emission provides an exceptional basis for quantum optics and nanoscale magnetometry.^2,3 In addition, low cytotoxicity and photostability of LNDs make them an attractive tool for biolabelling and cell targeted delivery purposes.^4–6 NDs can be produced via two methods: (1) detonation of explosives in an inert atmosphere followed by disintegration yields remarkably monodisperse 5 nm NDs,⁷ or (2) high-temperature high-pressure (HTHP) synthetic growth followed by ball-milling produces NDs of high crystal quality sized 4 nm and larger. Subsequent acid treatment and/or annealing in air removes amorphous carbon from nanocrystallite surface, replacing it with a variety of oxygen-containing groups, such as carboxyl groups.⁸ These groups impart surface charge sufficient to stabilize NDs in distilled water, in virtue of the nanoparticle electrostatic repulsion [Fig. 1(a), LND(DDW)].^2,9

Fig. 1 (a) Dot and Bar plots of the ζ-potential and mean diameter of colloidal LND, an un-reacted suspension of LNDs and Bs (LND+Bs), LND covalently bound to Bs (LND–Bs), LND–Bs linked to Bn (LND–Bs:Bn), and Bn–EGFP (LND–Bs:Bn–EGFP), where EGFP stands for enhanced green fluorescent protein. The samples are designated by solid fraction and, in bracket, solvent. For example, LND(DDW) designates luminescent nanodiamonds in distilled de-ionized water; LND(PB20), LND in 20 mM NaCl phosphate buffer, (b) sample transmission electron microscope (TEM) image of the as-received HTHP LNDs and (c) size histogram of LNDs obtained from TEM image.

LND deployment in biomedical applications critically depends on LND-bioconjugate stability in physiological solutions, currently an issue, as salt-induced surface charge screening leads to flocculation.¹⁰ Existing LND-functionalization methods to counter flocculation and to facilitate bioconjugation can be classified as covalent functionalization, e.g. amidation (utilizing surface carboxyl groups), silanization, adsorption, and lipid/polymer capping.^6,7,11 While the reported covalent functionalization protocols showed stability of the resultant complexes, their realization was often complex and case-dependent, whereas most of the other methods compromised functional stability. Here, we report on a versatile, salt-solution stable LND-bioconjugation platform that is based on a high-affinity protein pair termed barstar:barnase (Bs:Bn‡).¹²

Barnase, a bacterial ribonuclease, and its inhibitor, barstar, are characterized by an extremely small dissociation constant of 10⁻¹⁴ M, only ten-fold larger than that of its ubiquitous functional analogue, streptavidin:biotin.¹² Bs and Bn are water-soluble, comparatively small (M_w = 10.2, 12.4 kDa, respectively), temperature-stable (50 °C, barnase; 70 °C, barstar), and undergo fully reversible unfolding under extreme alkaline (up to pH 12) and acidic (up to pH 2) conditions. Their terminal groups are separated from the binding regions providing convenient docking terminals. Also, because of their bacterial origin, Bs:Bn are preferable over biotin:streptavidin drug/gene delivery modules due to suppressed non-specific binding in eukaryotic systems. Therefore, Bs:Bn represents an attractive molecular construct for interfacing LNDs with the biomolecule world.

The LND biofunctionalization strategy based on Bs:Bn is as follows: a covalently bound sub-unit LND–Bs (or LND–Bn) (referred to as Reaction 1), strongly locks to a pre-fabricated counter sub-unit, i.e. Bn–X (or Bs–X), by simply mixing the two colloidal substrates (referred to as Reaction 2), where X is a terminal molecule/nanoparticle. The terminal molecule can be biologically significant with a potential for targeted drug delivery and/or biolabelling applications.

140 nm HTHP LND in aqueous colloidal suspension, obtained from Academia Sinica, Taiwan,² was chosen as a substrate for the demonstration of the Bs:Bn-based nanoparticle bioconjugation platform. During fabrication, ND was irradiated with a 40 keV ion beam to create vacancies in the nanocrystals that subsequently combined with the substitutional nitrogen defects to form color centers, rendering the ND luminescent. Subsequent acid-treatment of the LNDs formed oxygen-containing surface groups, with carboxyl groups prevalent and convenient for docking biomolecules [see ESI for details†]. Size characterization of LND colloidal solution in water using dynamic light scattering (DLS) (Zetasizer, Malvern, UK) yielded average hydrodynamic diameter of 140 nm, which was 10% greater than that obtained from the average geometric diameter, as measured by transmission electron microscopy (TEM) (deviation from a spherical shape was noticeable) (Fig. 1). Although the LND aqueous colloid was stable, suspension in 20 mM NaCl phosphate buffer (PB20) resulted in formation of aggregates with sizes up to 250 nm. This showed the obvious propensity for aggregation of LNDs in physiological solutions, limiting their application scope.

In order to counter this aggregation issue and prepare the LND surface for bioconjugation, we designed and carried out a reaction to attach barstar to the LND surface covalently. The choice of barstar was guided by the following consideration: our commercially procured (as received) LND aqueous colloid was characterized by a negative zeta-potential (ζ-potential) of −43 mV, as measured by DLS [Fig. 1(a)], which indicated a negatively charged surface of LND. Hence, in order to minimize electrostatic adsorption, negatively charged Bs (isoelectric potential, pI 4.6) was preferred over the positively charged Bn (pI 8.9). A modified EDC/sNHS reaction relying on the covalent linking of LND COOH-group and Bs NH₂-group was carried out (referred to as Reaction 1, detailed in ESI†). An immediate observation of the reaction product (LND–Bs) property was its colloidal stability in PB20, with the average hydrodynamic diameter of LND remaining 140 nm, as in the as-received LND aqueous colloid [Fig. 1(a)].

Measurement of the ζ-potentials, by DLS yielded −43 mV and −30 mV for the as-received- and Bs-conjugated-LNDs, respectively [Fig. 1(a)]. This change in ζ-potential occurred due to the modification of LND surface moieties by the layer of Bs-molecules, strongly suggesting the covalent nature of the LND–Bs bonding, since the control reaction (Reaction 1 carried out without the activation step—see ESI†) yielded Bs-adsorbed LND complexes (LND+Bs), whose ζ-potential remained close to that of the as-received-LNDs [Fig. 1(a), LND+Bs(PB20)]. The LNDs coated with adsorbed proteins, LND+Bs was stable in PB20, as measured by Zetasizer and presented in Fig. 1(a). When similar reactions and characterizations were carried out for HTHP LND–Bs average-sized 35 nm, the results were essentially similar to those presented in Fig. 1, except for the sizes scaled down proportionally (see ESI for details†).

The strong covalent attachment of Bs in the case of LND–Bs in comparison with the unstable adsorption of Bs in LND+Bs was analytically confirmed by acid/alkali washing followed by X-ray Photoelectron Spectroscopy (XPS) analysis (see ESI†). The XPS data were analyzed for a nitrogen (N) spectral signature (at 400.3 eV) that was abundant in proteins, including Bs, and negligible in pristine LND. The adsorbed Bs-molecules, in the case of LND+Bs, were expected to detach during sonication/centrifugation at pH-values corresponding to pI 4.6, the characteristic isoelectric point of Bs. Indeed, the XPS spectrum showed no N-peak in the case of the LND+Bs sample [Fig. 2(b)], whereas the presence of N, and hence barstar surface molecules, in the case of the LND–Bs sample was obvious [Fig. 2(a)], thus confirming the covalent nature of the LND–Bs bonding. The noisy XPS signals were explained by small amount of the samples available for the analysis.

Fig. 2 XPS analysis of (a) LND–Bs and (b) LND+Bs, washed 7 times with PB20 with pH 7, 4, 7, 10.8, 7 and 7 with a dilution factor of 1:15 per wash. The nitrogen peak (N) manifested the presence of Bs–protein on the LND, providing strong evidence in favor of covalent nature of the LND–Bs bioconjugate.

In order to evaluate the biochemical functionality of the LND–Bs conjugate, a high concentration of barnase was added to the LND–Bs solution. Due to a positive charge of a pristine Bn (pI 8.9) (see ESI† for production of Bn), ζ-potential of LND–Bs:Bn colloid should increase, if the strong affinity of the bound Bs to Bn remained unperturbed. The measured ζ-potential of −21 mV confirmed this expectation [Fig. 1(a)].

The powerful versatility of the developed LND–Bs:Bn–X platform was demonstrated by producing (A) biomolecular and (B) nanoparticular complexes. Reaction 2(A): LND–Bs:Bn–EGFP complex was produced by mixing LND–Bs with an excess of Bn–[enhanced green fluorescent protein] (Bn–EGFP)¹³ (see ESI†). The product was washed thrice by centrifugation yielding the final product LND–Bs:Bn–EGFP and supernatants SN₁, SN₂ and SN₃, respectively (see ESI†). Samples were characterized using DLS, a fluorimeter (Fluorolog Tau3 system, JY Horiba) and a fluorescence confocal microscope (Leica TCS SL) (FCM) [Fig. 1(a), 3(a) and (b), respectively].

Fig. 3 (a) Fluorescence spectra of the LND–Bs:Bn–EGFP ([dash dash, graph caption]) conjugate and 2^nd, 3-supernatants (SN₂, SN₃, respectively) ([thick line, graph caption]). The characteristic spectral signatures of EGFP (left panel) and LND (right panel) under respective 480 nm and 532 nm excitations revealed their presence. (b) Spectra of LND–Bs:Bn–EGFP obtained using fluorescence confocal microscope (◇) at the bright image spot site (inset) superimposed on spectra of free EGFP ([dash dash, graph caption]) and LND ([thick line, graph caption]). Inset scale bar, 2 µm.

ζ-Potential of LND–Bs:Bn–EGFP, as measured by DLS, was −17 mV as compared to −19 mV of LND–Bs:Bn [Fig. 1(a)], confirming the strongly linked Bs:Bn-based assembly. The fluorescence spectrum of LND–Bs:Bn–EGFP showed pronounced spectral signatures of the LND⁵ and EGFP upon excitation with a 532 nm laser, and 480 nm spectrally filtered xenon-lamp, respectively. At the same time, the SN₂ and SN₃ samples displayed negligible fluorescence signals, indicating negligible concentrations of the LND and EGFP. This further corroborated the strong bonding between LND and EGFPvia the barstar:barnase link. Comparison of the fluorescence signals yielded a reasonable estimate of 1300 EGFP molecules attached to each LND, as compared with the calculated value of 7000 Bs/LND (see ESI†).

Imaging of LND–Bs:Bn–EGFP sample spun-coated on a glass cover slip by the fluorescence confocal microscopy (FCM) displayed several bright diffraction-limited spots. Emission spectra from these bright spots were acquired by operating the microscope in the λ-stack mode [Fig. 3(a)]. A clear co-localization of both LND and EGFP spectral signatures, shown in Fig. 3(b) inset, provided a strong evidence of the integrity of the LND–Bs:Bn–EGFP complex, which was further supported by the negligible EGFP background.

Reaction 2(B): LND–Bs (both, sized 140 nm and 35 nm) were mixed with an excess of pre-conjugated barnase–nanogold sub-unit (Bn–nAu) (see ESI†) to produce LND–Bs:Bn–nAu. These complexes were readily observable under a transmission electron microscope (TEM) with a depth of focus (front-on section thickness at the object plane) of ∼70 nm. The graphical abstract inset shows the nanogold diadem overlaying the diamond nanocrystal (140 nm). The nanogold particles were pseudocolored to emphasize the material contrast (see ESI† for the full-scale TEM image).

Feasibility of the reported complexes for biomedical applications was demonstrated by carrying out a non-specific cell internalization experiment. Chinese Hamster Ovary (CHO-K1) cells were incubated with LND–Bs:Bn–EGFP for 40 min at 37 °C, followed by FCM-imaging (see ESI†). An example overlay of the FCM and optical transmission phase-contrast microscopy image [Fig. 4(a)], shows a bright fluorescent spot localized in the cellular region. The fluorescence signal from the spot was analyzed to reveal the co-localization of LND and EGFP spectral signatures, as shown in Fig. 4(b). It is likely that the LND–Bs:Bn–EGFP conjugate is sequestered in an early endosome in virtue of the short incubation time, preventing enzymatic degradation of EGFP.

In summary, a new, versatile nanodiamond bioconjugation platform based on the high-affinity protein pair, barstar:barnase, has been demonstrated. The reported reaction was simple, resulting in covalently bound LND–Bs bioconjugates that are stable in buffer solutions for periods greater than several months. Considerable practical utility of this platform was demonstrated by the design, production and characterization of various nanodiamond-biomolecule/nanoparticle complexes followed by the demonstration of biological application. The demonstrated Bs:Bn platform will pave the way for versatile, easy-to-use bioconjugation of nanomaterials.

Fig. 4 (a) Overlay of the FCM and transmission phase-contrast microscopy images. 488 nm laser illumination was used. (b) Spectrum of the bright spot [in (a), demarcated by the arrow endpoint] showed co-localization of LND and EGFP. FCM-induced spectral distortion of the LND spectral shape was minimized by post-processing of the original signal.

Acknowledgements

This work was partially supported by Macquarie University Research Innovation Fund # 1136900; Russian Academy of Sciences Programs Molecular & Cellular Biology and Nanotechnologies & Nanomaterials, Russian Foundation of Basic Research Grants # 09-04-01201, Russian Federal Agency for Science and Innovation.

Notes and references

1. J. Martin, R. Wannemacher, J. Teichert, L. Bischoff and B. Kohler, Appl. Phys. Lett., 1999, 75, 3096–3098.
2. C. C. Fu, H. Y. Lee, K. Chen, T. S. Lim, H. Y. Wu, P. K. Lin, P. K. Wei, P. H. Tsao, H. C. Chang and W. Fann, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 727–732.
3. G. Balasubramanian, I. Y. Chan, R. Kolesov, M. Al-Hmoud, J. Tisler, C. Shin, C. Kim, A. Wojcik, P. R. Hemmer, A. Krueger, T. Hanke, A. Leitenstorfer, R. Bratschitsch, F. Jelezko and J. Wrachtrup, Nature, 2008, 455, 648–U646.
4. A. M. Schrand, S. A. C. Hens and O. A. Shenderova, Crit. Rev. Solid State Mater. Sci., 2009, 34, 18–74.
5. S. J. Yu, M. W. Kang, H. C. Chang, K. M. Chen and Y. C. Yu, J. Am. Chem. Soc., 2005, 127, 17604–17605.
6. R. A. Shimkunas, E. Robinson, R. Lam, S. Lu, X. Xu, X.-Q. Zhang, H. Huang, E. Osawa and D. Ho, Biomaterials, 2009, 30, 5720–5728.
7. A. Krueger, M. Ozawa, G. Jarre, Y. Liang, J. Stegk and L. Lu, Phys. Status Solidi A, 2007, 204, 2881–2887.
8. S. Osswald, G. Yushin, V. Mochalin, S. O. Kucheyev and Y. Gogotsi, J. Am. Chem. Soc., 2006, 128, 11635–11642.
9. K. Ushizawa, Y. Sato, T. Mitsumori, T. Machinami, T. Ueda and T. Ando, Chem. Phys. Lett., 2002, 351, 105–108; J. Y. Park, E. S. Choi, M. J. Baek and G. H. Lee, Mater. Lett., 2009, 63, 379–381.
10. F. Neugart, A. Zappe, F. Jelezko, C. Tietz, J. P. Boudou, A. Krueger and J. Wrachtrup, Nano Lett., 2007, 7, 3588–3591; S. A. Dahoumane, M. N. Nguyen, A. Thorel, J. P. Boudou, M. M. Chehimi and C. Mangeney, Langmuir, 2009, 25, 9633–9638.
11. C. Y. Cheng, E. Perevedentseva, J. S. Tu, P. H. Chung, C. L. Cheng, K. K. Liu, J. I. Chao, P. H. Chen and C. C. Chang, Appl. Phys. Lett., 2007, 90, 163901–163903; A. Kruger, Y. J. Liang, G. Jarre and J. Stegk, J. Mater. Chem., 2006, 16, 2322–2328.
12. S. M. Deyev, R. Waibel, E. N. Lebedenko, A. P. Schubiger and A. Pluckthun, Nat. Biotechnol., 2003, 21, 1486–1492.
13. E. N. Lebedenko, T. G. Balandin, E. F. Edelweiss, O. Georgiev, E. S. Moiseeva, R. V. Petrov and S. M. Deyev, Dokl. Biochem. Biophys., 2007, 414, 120–123.

---

Footnotes

† Electronic supplementary information (ESI) available: Detailed experimental procedures and analysis. See DOI: 10.1039/c0jm02819c

‡ Specific designations for covalent, high-affinity non-covalent binding and adsorption are “−”, “:”, and “+”, respectively.

---