Luminescent Nanomaterials for Molecular-Specific Cellular Imaging

  • Andrei Vasilyevich Zvyagin
  • Zhen Song
  • Annemarie Nadort
  • Varun Kumaraswamy Annayya Sreenivasan
  • Sergey Mikhailovich Deyev
Reference work entry


Imaging of molecular trafficking in cells and biological tissue aided by molecular-specific fluorescent labeling is very attractive, since it affords capturing the key processes in comprehensive biological context. Several shortcomings of the existing organic dye labeling technology, however, call for development of alternative molecular reporters, with improved photostability, reduced cytotoxicity, and an increased number of controllable surface moieties. Such alternative molecular reporters are represented by inorganic luminescent nanoparticles (NP) whose optical, physical, and chemical properties are discussed on the examples of luminescent nanodiamonds (LND) and upconversion nanoparticles (UCNP). The emission origins of these nanomaterials differ markedly. LND emission results from individual nitrogen-vacancy color-centers in a biocompatible nanodiamond host whose properties can be controlled via size and surface groups. Photophysics of UCNP is governed by the collective, nonlinear excitation transfer processes, resulting in conversion of longer-wavelength excitation to the shorter-wavelength emission. The emission/excitation spectral properties of UCNP falling within the biological tissue transparency window open new opportunities of almost complete suppression of the cell/tissue autofluorescence background. The developed surface of these nanoparticles represents a flexible platform populated with biocompatible surface moieties onto which cargo and targeting biomolecules can be firmly docked through a process called bioconjugation. These bioconjugated modules, e.g., nanodiamond-antibody, (quantum dot)-somatostatin, or (upconversion nanoparticle)-(mini-antibody) can gain admission into the cells by initiating the cell-specific, cell-recognized communication protocol. In this chapter, we aim to demonstrate the whole bottom-up bio-nano-optics approach for optical biological imaging capturing luminescent nanoparticle design, surface activation, and bioconjugation and the resultant bioconjugate module deployment in specific internalization in the cell.


Optically Detect Magnetic Resonance Detonation Nanodiamonds Autofluorescence Background Upconversion Nanoparticles Specific Internalization 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

14.1 Introduction

Optical imaging assisted by molecular-specific luminescent labeling is a direct, minimally invasive approach to investigate cellular morphology and processes in living cells and/or tissues, in their comprehensive biological context. Enabled by the recent developments in biophotonics, single-molecule imaging unobscured by the ensemble averaging has become a powerful and common imaging modality suitable for complex biological systems. As an example, single-molecule imaging is instrumental in tracking molecular trafficking events in cell signaling via activation of membrane receptors, and investigating the fates of individual receptors and ligands [1]. Another example represents detection of a single virus tagged with fluorescent molecule that has allowed tracking the virus invasion of the cell and its diffusion toward the nucleus, providing an important insight in the area of virology [2]. These studies place stringent demands on the fluorophore performance, which is frequently compromised by undesirable photophysical properties, such as fluorescence intermittency (blinking) and irreversible light-induced transitions to dark states (photobleaching) that hinder its photostability. The dark state transitions are particularly limiting in the single-molecule studies, which require high illumination intensities. Indeed, if an organic dye fluorophore survives only one million excitation/emission cycles before undergoing an irreversible transition to the dark state (photobleaching), the continuously emitting fluorophore lifetime spans only ≈ 3 ms (assuming the fluorescence lifetime τ = 3 ns under saturation pumping condition). The longer-wavelength fluorophores, such as Cy5 and Cy7, Alexa Fluor 750, and CF dyes, whose excitation/emission fall into the so-called biological tissue transparency window (wavelength range, [The definition of the biological tissue transparency window is not settled varying from the given range to a range of 680–900 nm depending on the biological tissue constituents.] 700–1,300 nm) (Fig. 14.1a), are especially prone to photobleaching. These dyes are also widely used in applications where high signal-to-noise ratios are required, such as fluorescence resonance energy transfer (FRET). Photostability is also critical in applications, including stimulated emission depletion microscopy (STED), and despite progress in the design of organic fluorophores that are less prone to photobleaching, improved alternatives are in high demand. The other shortfalls of the fluorescence dyes include potential cytotoxicity, poor resilience to aggressive chemical and biological environment, and limited ability to carry cargo to cellular sites.
Fig. 14.1

(a) Optical effective attenuation spectrum of living skin tissue (solid green line) dominated by water (H2O, blue -⋅⋅- line), hemoglobin (Hemoglobin, brownline), oxy-hemoglobin (HbO2, red solid line), proteins (not shown), with the scattering effect also taken into account. Biological tissue transparency window ranges from 700 to 1,300 nm. (b) Autofluorescence image of human skin under the excitation at 405 nm. The viable epidermis layer is color-coded purple and marked by an arrow, with cell nuclei visible as dark ovals. Dermis, visualized primarily via collagen and elastin bundles, is situated right next to the epidermis extending to the right

Nanotechnology has provided a powerful impetus to the new generation of molecular probes based on luminescent nanoparticles that are capable of addressing the major shortfalls of the existing molecular probes. In particular, a variety of nanoparticles, including luminescent nanodiamonds (LND) [3] and upconversion NPs (UCNPs) [4], exhibit enviable photostability, i.e., unfading continuous emission. Besides, reduced cytotoxicity of these nanomaterials has been recently reported [5, 6]. The upconversion NPs offer improved signal-to-noise ratio in the time-gated imaging and high-contrast detection employing time-gated schemes that have been reported using lanthanide-based UCNPs [7]. NP surface is amenable to modification to enable grafting a variety of surface moieties suitable for applications, such as molecular targeted drug delivery [8].

We review two types of luminescent nanoparticles in the context of optical cellular/tissue imaging: luminescent nanodiamond and upconversion nanoparticle. The choice of the NPs is underpinned by the diversity of their luminescence properties that capture important classes of nano-emitters. LND comprises several isolated color-centers in a nanodiamond host; UCNP luminescence is governed by nonlinear excitation transfer processes between strongly coupled ionic emitters. The underlying photophysics of these luminescent nanomaterials makes them useful for bioimaging, in particular, render NPs imaging contrast on the crowded cell morphology and autofluorescence background of the cells/tissues. The NP core chemical inertness is essential to preserve NP integrity in often aggressive biological environments and minimize its potential toxicological consequences (or cytotoxicity). Surface moieties are important for interfacing NPs with the biomolecular world. To this end, the basic principles of linking nanoparticles with biomolecules of programmable functionality, the process termed bioconjugation, is covered and several examples are presented. The bioconjugated NP-biomolecule complex is capable of performing programmable functions by, for example, activating a communication protocol with the cell that can result in gaining admission into the cell. We will present several examples of this specific cellular internalization visualized by means of optical luminescence imaging.

This book chapter is organized, as follows: The two types of luminescence nanoparticles are addressed in the first two sections, with particular focus on their photophysical properties in relation to optical biological imaging. Diverse surface properties of these nanomaterials enable adequate coverage of the existing surface functionalization strategies and lead to the following section (Sect. 14.4), where the key principles of the nanoparticles bioconjugation to biological molecules will be addressed and several examples given. Section 14.5 is devoted to reviewing specific interaction of the programmable nanoparticle-biomolecule conjugates with the cell. In Conclusions, the progress in the nanotechnology as applied to luminescent molecular probes is summarized.

14.2 Luminescent Nanodiamonds (LND)

14.2.1 Nanodiamond

Luminescent nanodiamond is usually referred to a diamond nanocrystal (ND) that hosts luminescent color centers. The diamond crystalline matrix is the most rigid material on the Earth, made of carbon atoms bonded together with the energy of ≈ 7.4 eV per atom. The diamond core is chemically inert, so that the centers are well secluded from environmental perturbations in such a crystal host. Even high-energy cosmic radiation inflicts little damage to this crystalline material. At the same time, the diamond surface is chemically reactive, especially when the surface is well developed, as in the case of surface-functionalized nanodiamonds. For example, the most popular acid-treatment [9, 10, 11] or high-temperature annealing in air [8, 12, 13, 14, 48] results in the formation of oxygen-containing surface moieties on the diamond surface that facilitate interfacing with photonic devices and macromolecules. NDs can be produced via two methods: (1) High-temperature high-pressure (HTHP) synthetic growth followed by ball-milling produces NDs of high crystal quality sized 4 nm and larger [15] (19). (2) Detonation of explosives in an inert atmosphere followed by disintegration yields remarkably monodisperse 5-nm NDs [16]. Subsequent acid treatment and/or annealing in air removes the surface layer of amorphous carbon, replacing it with a variety of oxygen-containing groups, such as carboxyl groups [8].

Very high refractive index of the nanodiamond core renders NDs conspicuous on the crowded environment of the cell. Figure 14.2 shows NDs internalized in the cells via a micelle transport vehicle [17]. The nanodiamond clusters encapsulated in endosomes on passing into the cytoplasm became dominant scatterers that overpowered intrinsic scattering signals coming from, e.g., mitochondria, and rendered the cytoplasm bright against the dim nuclei (circled and designated by “N” in Fig. 14.2). However, this scattering contrast is by far insufficient in achieving detection sensitivity at the single nanoparticle level due to the hefty cell scattering overhead that is indistinguishable from the scattering signals of the nanodiamond-based molecular probes. Spectral separation of the scattering background from the molecular probe signals of non-elastic-scattering nature is an efficient strategy to improve the imaging contrast and, hence, detection sensitivity. This is realized by utilizing the luminescence properties of nanodiamonds.
Fig. 14.2

Differential interference contrast image of nanodiamond particle uptake into 3T3 cells. The nanodiamonds exhibit high contrast against the cell background due to strong elastic scattering and are observable as bright rims around the dim cell nuclei which are impermeable for the NDs (a nucleus is indicated with a circle) (Adapted from Ref. [17])

14.2.2 Nitrogen-Vacancy Center

The wide bandgap of diamond (5.5 eV) provides ample energy space to as many as ≈ 500 reported color-centers, where nitrogen-vacancy (NV), silicon-vacancy [18], and nickel-related [19] and chromium-related [20] centers are among the brightest, with the emission bands falling into the biological tissue transparency window (cf. Fig. 14.1a), making them promising candidates for optical biomedical applications. The NV center, however, exhibits several exceptional properties that attract considerable interest of the nano-optics community [21]. Therefore, only NV center in ND will be addressed hereafter, and referred to as “luminescent nanodiamond.” The NV center in diamond is formed by replacing one carbon atom with a nitrogen atom (termed “substitutional nitrogen,” N s ) and a vacancy at a location adjacent to the nitrogen atom, as schematically shown in Fig. 14.3, inset. The production process usually involves irradiation of diamond samples with high-energy electron or light-ion beams forming vacancy defects. In order to initiate migration of these vacancy defects in a diamond (nano)crystal, the sample is annealed at a high temperature ( ≈ 800 \(^{\circ }\)C) in a high vacuum or oxygen-free atmosphere. This causes vacancy trapping at the substitution nitrogen sites that are abundant in the majority of synthetic diamond samples. The NV center can remain neutral (NV0) or acquire an additional electron from the neighboring donor defects, e.g., N s , becoming negatively charged NV center (NV). Both centers are luminescent, with the spectra shown in Fig. 14.3. The acclaimed merits of the NV center are commonly associated with the negative-charge state of the NV center, and can be interpreted in terms of its physical, optical, and magneto-optical properties.
Fig. 14.3

Luminescence spectra normalized to their respective maximum value of single NV (red curve) and NV0 (blue curve) color centers in diamonds. The zero-phonon lines (demarcated  ⋆ ) of NV, NV0 emissions are located at wavelengths 637 and 575 nm, respectively. The inset shows the atomic structure of the NV defect, consisting of a substitutional nitrogen atom (N s , demarcated N) associated to a vacancy (V) in an adjacent lattice site of the diamond crystalline matrix (Reproduced from Ref. [22])

The NV center action cross-section is large, \(\eta \sigma _{a} = 3 \times 1{0}^{-17}\) cm2 (σ a – absorption cross-section; η – quantum yield) [23], with the quantum yield reaching ≈ 80 % under favorable excitation conditions [24]. The energy diagram of the NV center (shown in Fig. 14.4) elucidates estimation of η, as stemming from the branching ratio of the radiative and non-radiative transitions, as well as the NV optical spin polarization property that is highly acclaimed in quantum science and spin magnetometry, and briefly touched below.
Fig. 14.4

Simplified five-level schematic diagram of the electronic levels of the NV color center appropriate for room temperature. The m S = 0 levels are indicated in red, the m S = ± 1 in blue, and the singlets in green. The double-headed arrows indicate the radiative transitions. The solid blue and red arrows give the spin conserving transitions in the visible. The green wavy arrows indicate non-radiative decay that involves E-vibrations

The ground state of NV is an orbital singlet and spin triplet. In trigonal symmetry the spin m S = + 1, and − 1 states are exactly degenerate and, when modeling the optical cycle of the NV center, the m S = ± 1 can be described by one level (assuming no magnetic fields). The excited state is also a spin triplet, which is treated as an orbital singlet in the phenomenological model. The model, therefore, includes two spin levels, 0 and 1, in the ground electronic state and two electronic states in the excited state, 0 e and 1 e (Fig. 14.4). The optical transition (at a rate α) between the ground and excited state is independent of the spin projection giving rise to two transitions of identical strength between levels 0 and 0 e and between 1 and 1 e , where the spin projection is unchanged, despite the vibration coupling. Should these be the only energy levels and only transitions, the center would cycle between the ground and excited states, with constant spin projection. However, the situation is entirely changed by the presence of an intermediate state (denoted “S” in Fig. 14.4). The intermediate state is attributed to a spin singlet (or states) S = 0, with no spin projection. Inter-system crossing takes place from the excited state to the singlet and from the singlet to the ground state. These crossings are spin-dependent and give rise to spin polarization and optical spin readout. The approximate rates of the crossing from the excited state spin levels to the singlet are \(\gamma _{1} = 0.55\alpha\) and \(\gamma _{0} = 0.15\alpha\), and from the singlet to the ground state spin levels are \(\beta _{0} = 0.035\alpha\) and \(\beta _{1} = 0.022\alpha\) (Fig. 14.4). The decay rate from the singlet to the ground state is low, with a combined rate of 0.056α. The decay out of the triplet to the singlet is comparable to the optical transition and the rate associate with m S = ± states (from 1 e ) is more rapid than for m S = 0 states (from 0 e ). These inter-system crossings underpin the NV center unique properties. They result in (a) population being transferred out of m S = ± 1 states, (b) the emission rates from m S = + − 1 states being weaker than from m S = 0 states, and (c) the associate luminescence lifetime (in diamond crystals) from m S = + − 1 states being shorter at 7.8 ns than that from m S = + − 1 states at 12 ns [25]. With continuous excitation, the relative populations in the excited state is given by P ±P 0 by \(\gamma _{0}/\gamma _{1} \times \beta _{1}/\beta _{0} = 0.15/0.55 \times 0.022/0.035 = 0.17\). This implies 14.5 % of the population is in the excited m S = ± spins and 85.5 % in m S = 0. The ground state polarization is less by the ratio of the lifetimes 7.8/12 and, hence, somewhat lower at 74 % in m S = 0.

Depopulation of the spin-specific (m S = ± 1) ground state in favor of the m S = 0 ground state in the course of optical excitation represents a venerable property of the NV center, which is realized by a possibility to read out thus-prepared spin-state by optical means. The coherence time of this spin state at room temperature is as short as several microseconds, but its extension to 1 ms has been demonstrated by the appropriate choice of an ultrapure diamond host in terms of the much reduced defects and impurities, especially N s . This enables preparing and interrogating the entangled spin states. The quantum science applications rely on the preparation of an NV center spin-state (qubit), followed by its coupling to the neighbor spins associated with defects, such as NV center spins, substitutional nitrogen (N s ) or isotopic C13 nuclear spins, creating quantum entanglement of the qubits. Reading out the NV entangled spin state optically provides a measure of the entangled state of the ensemble. The magnetometry applications rely on the high sensitivity of the NV center polarized spin-state to (local) magnetic fields, as shown and briefly discussed in Fig. 14.7. The spin-state perturbation is read out optically by means of the optically detected magnetic resonance (ODMR) and reports on the local value and orientation of the perturbing magnetic field vector. The ultimate sensitivity is estimated at the level of a single spin at the nanoscale proximity [26]. The virtually unlimited photostability of the NV center is another important property manifested by uninterrupted continuous emission under continuous excitation, and it is important in biomedical imaging applications. This represents a radical solution to the problem of poor photostability of the widespread emitters, organic fluorescent dyes that undergo irreversible conversion to a low- or no-emission state (or photobleaching) after, approximately, one million excitation cycles. Luminescence intermittency, or blinking, represents another photostability impairment mechanism, and is reported in virtually all single emitters [27, 28] especially affecting semiconductor quantum dot emission [29, 30, 31]. Although the NV centers have also been reported to exhibit blinking behavior, this can be readily avoided under controllable conditions [32].

14.2.3 Luminescent Nanodiamonds for Cellular Imaging Applications

The first observation of NV-ND imaging in cells was reported by [3], where high-contrast imaging and virtually unlimited photostability of 35-nm LNDs were reported. The authors demonstrated nonspecific internalization of LNDs in the cells (see Fig. 14.5), followed, by the demonstration of several applications of single LND imaging and tracking in the cellular environment [33]. Nonspecific internalization refers to poorly controlled process of the extraneous material uptake by the cell, and is discussed in Sect. 14.4.
Fig. 14.5

Observation of single LNDs in a HeLa cell. (a) Bright-field and epi-luminescence images of a HeLa cell after uptake of 35-nm LNDs. Most of the up-taken LNDs are seen to distribute in the cytoplasm. (b) Epi-luminescence image of a single HeLa cell after the LND uptake. An enlarged view of the fluorescence spots (denoted by “1” and “2”) with diffraction-limited sizes (FWHM ≈ 500 nm) is shown in inset. (c) Intensity profile of the fluorescence image along the line drawn in b inset. (c inset) Integrated fluorescence intensity (after subtraction of the signals from cell autofluorescence and background fluorescence from the microscope slides) as a function of time for particle “1.” The signal integration time was 0.1 s. No sign of photobleaching was detected after continuous excitation of the particle for 20 min (Reproduced from Ref. [3])

Our own observation of the nonspecific internalization of a single luminescent nanodiamond sized ≈ 100 nm in Chinese Hamster Ovary (CHO-K1) cells confirmed the feasibility of single luminescent nanoparticle imaging, with the scattering background largely suppressed [34]. Combined with the reported minute cytotoxicity of the nanodiamonds [35, 36], the single-particle imaging demonstration sparked considerable interest in the applications of LND, as a molecular probe in live systems. These high expectations were underpinned by the nanodiamond surface properties that made this material amenable to bioconjugation leading to targeted delivery in cells and specific tissue sites. And, indeed, a number of interesting demonstrations of the specific internalization in cells of the LND bioconjugated with targeting biomolecules have been demonstrated (see, e.g., comprehensive review by Schrand et al. [37]). It is worthwhile to note that the LND bioconjugates can serve not only as the molecular probes, but also as biomolecule delivery vehicles [38]. This aspect of luminescent nanotechnology will be detailed in Sect. 14.5 of this chapter.

The optical spin-selective properties of NV-ND have also found new applications in cellular imaging applications. In particular, monitoring the decoherence rates in response to changes in the local environment may provide new information about intracellular processes [39]. The reported experiments demonstrate the viability of controlled single spin probes for nano-magnetometry in biological systems, opening up a host of new possibilities for quantum-based imaging in the life sciences. Figure 14.7 shows optically detected magnetic resonance (ODMR) signals from the single NV centers in different ND crystals spatially localized in the cell cytoplasm. The ODMR signals appeared to be capable of reporting on the intracellular environment.

14.2.4 Ultrasmall LND

Considering imaging and sensing applications, the single-spin sensitive magnetometry [26] and Förster resonance energy transfer [40] require the nanodiamond layer to be thinned to exploit the strong dependence of the signal on the interaction radius r (1/r6 and 1/r3, respectively). The single-spin-sensitive imaging at the nanoscale led to an estimation that the NV center had to be as close as 10 nm to an interrogated single spin for reliable detection [26]. In optical biological imaging applications, ultrasmall diamonds are crucial to minimize disruption to the molecular trafficking under observation [33]. It is desirable to reduce the size of NDs to match the size of an average protein, i.e., 5 nm, which is achievable due to the progress in the production of ultrasmall nanodiamonds, also referred to as ultra-dispersed nanodiamonds [41]. It has also been reported that NV-NDs of the size range less than 10 nm retain their luminescent properties characteristic of the NV centers.

At the same time, the NV center stability sets a size limit on the NV nanodiamond (NV-ND) matrix. A 2.5-nm ultra-small nanodiamond may be incapable of hosting an NV center in the core, as predicted theoretically [42]. The environmental susceptibility of the NV-ND emission in small crystals, or NV centers in the nanoscale proximal surface layer, has been reported on several occasions [22, 32, 43]. The question of the NV center stability versus its distance to the diamond surface became essential two decades ago due to the progress toward production of the ultrasmall NDs of the detonation origin [9]. The problem of the existence of NV centers in such small nanocrystals became a hot research topic in view of the significant appeal of applications of low-cost detonation NDs. On the other hand, the experimental observations provided little support in favor of the existence of the NV-ND. For example, the theoretical calculations favored the location of nitrogen on the surface rather than in the core, which seemed to explain the limited observation of NV centers in chemical vapor deposition and high-pressure high-temperature (HPHT) grains less than 40 nm in size [44, 45], and favored the prediction that NDs smaller than 10 nm in size were deprived of NV centers [18, 42]. In particular, the spectroscopic studies revealed broadband luminescence [46] that was wavelength blue-shifted from the characteristic NV spectra, shown in Fig. 14.3. The origin of this luminescence was attributed to a graphitic layer that formed in the detonation process. The photoluminescence excitation spectra analysis of HPHT and CVD diamonds also corroborated the notion of NV center instability in ultrasmall NDs by showing broadband component attributed to the amorphous carbon phase [47].

The obscuring signal overhead of the graphite layer to the overall luminescence signal from NV-NDs was suppressed by employing a time-gated detection scheme making use of the difference in the luminescence lifetimes of graphite and NV (sub-ns versus  ≈ 10–20 ns, respectively). As a result, the characteristic NV center spectral signature was observed, thus providing an unambiguous evidence of the existence of NV centers in the detonation nanodiamonds [43]. A direct obviation of the overshadowing luminescence signal background from the graphite layer was reported by [32]. Thoroughly acid-cleaned discrete detonation nanodiamonds were sparsely dispersed on a glass slide and interrogated individually, thus minimizing the graphite luminescent signal background, and so a signal originated from only one ND became observable. As a result, the characteristic NV spectral signatures from single centers were acquired.

An alternative top-down approach of production of the ultrasmall NDs has been reported by [15]. This approach was based on milling a millimeter-sized diamond crystal, following the NV center implantation using the conventional high-energy electron irradiation and high-temperature annealing procedure. The authors reported stable luminescence and high-contrast ODMR signal from the single NV centers in NDs sized less than 8 nm. Oxidative etching in air/oxygen at high temperature is another top-down approach of ultrasmall ND production. This method relies on the removal of carbon atoms from the ND surface at a controllable rate via oxidation reactions. Since an outer ND layer in the graphitized (sp2) form is removed at a higher rate [48, 49, 50] than that of the diamond sp3 phase, this method was originally employed to remove sp2 carbon from ND, as a surface-cleaning procedure [8]. This method was later extended for in situ studies of ND size evolution, as the oxidative etching in air provides a least disruptive method of graphite cleaning and ND core size reduction [51]. NV emission in the ND crystals sized less than 10 nm has been reported [52].

14.2.5 Surface Effects on the NV Emission

Diamond surface effects appear to play a considerable role in the emission behavior of the NV centers, if these are situated at the nanometer proximity. Three principal cases of the surfaces state are as follows: (1) Emission, absorption and quenching by the graphite sp2-layer. (2) Surface functional groups affect the NV center emission state – O- or H-containing groups promote or inhibit the NV emission. (3) NV centers luminescence intermittency (blinking) takes place, if the center center is proximal to a substrate, with abundant electron acceptors.

It has been demonstrated that a graphite (sp2) shell surrounding the detonation nanodiamonds causes suppression of the NV luminescence [43], whereas oxygen-containing surface groups have little effect on the NV emission and, moreover, are capable of healing the surface defects that results in the NV emission activation [53]. The sp2 shell is a low-density layer of graphite impregnated with noncarbon impurities that is loosely attached to the ND surface. This lax graphite layer is removable by acid cleaning or oxidative etching. As a result, the NV emission is converted from the latent non-emission state to active emission state. In addition, this sp2 black carbon layer contributes to attenuation, of the excitation/emission light. Removal of this layer reduces the attenuation dramatically improving the NV excitation/emission optical properties. Besides, the sp2 layer functions as an electron acceptor causing discharge of the NV center accompanied by its conversion to NV0 or to “dark” non-emission state [22, 54].

The hydrogen termination of the ND surface resulted in a number of active NV emitters reduced almost tenfold in comparison with that of the original oxygenated NV-ND sample (Fig. 14.8). A subsequent oxidative treatment redressed the LND surface with oxygen groups and resulted in the recovery of the NV emission, as shown in Fig. 14.8c, where a number of the NV luminescent centers reappeared at their original sites (circled). Also, note that a number of blinking centers recovered their original blinking behavior after being in a latent dark state in the H-terminated LND (marked by a white downward arrowhead, Fig. 14.8).

The effect of the hydrogen surface group on the NV emission in ND on the charge state of the nitrogen-vacancy electron configuration has received much attention lately [53, 55]. In brief, the H-moieties cause an up-shift of the NV energy structure in the ND due to the modulation of the electron affinity. The valence band energy edge rises above the electrochemical potential of water, driving electrons from the nanometer-thick ND subsurface layer to water until the two-dimensional gas of holes in this ND subsurface layer compensates the gradient forces. It causes charge depletion of the NV center that enters either NV0 charge state [53] or the dark state [53].

The surface effect on the NV emission state can be summarized, as follows: In general, the LND luminescence is immune to the surface effects, as long as the surface is deprived of e-acceptors, and the diamond crystal host is not distorted. This condition is realized in case of the surface-oxygenated ND, even if the centers are situated as close as 1 nm to the surface [22, 56].

14.2.6 Limitations of LND for Optical Biomedical Imaging

There are two major problems that limit the application scope of LNDs for cellular nano-photonics. Firstly, production of NV-ND remains expensive and cumbersome, since it relies on the high-energy electron or light-ion irradiation. These irradiation sources are costly, not readily available, and the production process is not easily scalable, despite some progress in this direction [33]. So, production of LND is limited to a relatively small amount of the nanomaterial. Annealing at relatively high temperatures in vacuum represents another complication in the LND production process that requires specialized equipment. Secondly, the LND excitation takes place in the visible spectral region, ranging from 488 nm (Ar-ion laser) to 545 nm (helium-neon laser), where an excitation at 532 nm (second harmonic, Nd:YAG laser) is often preferred. Biological systems respond to this excitation by emitting fluorescence, termed “autofluorescence,” which is spectrally broad extending to the near-infrared spectral range (tailing off at about 800 nm). This autofluorescence signal is spectrally overlapped with the broadband luminescence of LND (Fig. 14.2), thus making spectral filtering of the LND signal inefficient. Indeed, one can see non-negligible background in Figs. 14.5 and 14.6 due to the cell autofluorescence, which ultimately limits the detection sensitivity of LND precluding the use of ultrasmall LND for imaging in live cells. More efficient mechanisms of suppression of the excitation light and biological system autofluorescence are required to realize the optical biomedical imaging at the single nanoparticle sensitivity. These mechanisms and their realizations are discussed in the following section.
Fig. 14.6

Observation of a single LND bioconjugate in the CHO-K1 cell. (a) Laser scanning confocal microscopy of a ≈ 100-nm luminescent nanodiamond (red spot). An arrow points from the LND to (b) Fluorescence spectra of LND and green fluorescence proteins (EGFP) conjugated to the nanodiamond surface (Reproduced from Ref. [34])

Fig. 14.7

(ac) Confocal image of HeLa-1, with luminescence gated around the NV emission (650–800 nm). The nucleus and cell membrane are indicated with dashed lines for clarity. The optically detected magnetic resonance spectra of NV-1a (a) and NV-1b (b) show the different strain splitting between the two centers (Reproduced from Ref. [39])

Fig. 14.8

The effect of the surface hydrogenation on the luminescent NDs. (a) NV luminescence after thermal oxidation. The black and white arrowheads point to stable and blinking NV emitters. (b) Hydrogenated NV centers exhibit diminished emission. (c) The sample redressing with oxygen groups led to the recovery of NV emission and blinking. Noticeable background fluorescence in (c) was activated by the H- and O-treatment (Reproduced from Ref. [52])

14.3 Upconversion Nanoparticle (UCNP)

The unique photophysical properties of UCNP allow almost complete suppression of the biological tissue background that leads to the ultimate single nanoparticle photodetection sensitivity.

14.3.1 Photophysics of Upconversion Nanomaterials

Photon upconversion is a nonlinear process manifested by the conversion of the long-wavelength excitation to short-wavelength emission. The mechanism of upconversion is based on sequential absorption of two or more photons by the metastable long-lived energy states, as shown in Fig. 14.9a and briefly discussed later, which can be induced by relatively low-intensity (1 ×103 W cm− 2) continuous-wave excitation conveniently provided by an inexpensive 980-nm laser [4]. This process differs from a simultaneous two-photon absorption that occurs via a virtual intermediate level, stimulated by the excitation intensity as high as 1 ×105 W cm− 2 usually realized by means of a complex and expensive femtosecond laser [57]. The majority of the upconversion processes involve absorption and non-radiative energy transfer steps. Theoretically, luminescent materials featuring 4f and 5d shell ions that have more than one metastable level are able to generate upconversion luminescence. Referring to Fig. 14.9a, a two-level Yb3 + ion serves as the donor (also known as the sensitizer) whose role is to absorb an IR excitation photon via 2F\(_{7/2} {\rightarrow \ }^{2}\)F5 ∕ 2 transition path, and non-radiatively transfer the excitation energy to a neighboring Yb-ion, or Er3 + (Tm3 +) ions, also known as activators. As a result, an ensemble of the excited dopant Yb-ions forms a non-localized exciton spread across the entire nanocrystal matrix maintained by the non-radiative energy exchange process. This excitation energy is continuously transferred to the network of the activators, Er3 + or Tm3 +, with the process rate order evaluated as 1,000 s− 1 for the representative crystal matrix of Cs3Lu2Br9 [58]. The activated ion, for example, Er3 + in the 4I11 ∕ 2 metastable state, can coalesce with the metastable Yb3 + (2F5 ∕ 2) via a collective pair-wise process of the energy transfer upconversion (Fig. 14.9b). As a result, the Er3 + (4I11 ∕ 2) makes a transition to the next-level excited state Er3 + (4F7 ∕ 2), followed by a rapid at the expense of the Yb3 + (2F5 ∕ 2) → Yb3 + (2F7 ∕ 2) non-radiative transition to Er3 + (4S3 ∕ 2) (refer to Fig. 14.9a, b). The Er3 + (4S3 ∕ 2) radiates in a multiplet green spectral band, and also in a multiplet red spectral band, following the non-radiative phonon-assisted transition Er3 + (4S3 ∕ 2) → Er3 + (4F9 ∕ 2), as one can see in Fig. 14.9c. The Er3 + (4F9 ∕ 2) population is also built up via an alternative excitation path: non-radiative transition Er3 + (4I11 ∕ 2) → Er3 + (4I13 ∕ 2), followed by the energy transfer upconversion [Er3 + (4I13 ∕ 2) → Er3 + (4F9 ∕ 2)] and [Yb3 + (2F5 ∕ 2) → Yb3 + (2F7 ∕ 2)].
Fig. 14.9

Detailed and simplified schematic energy level diagram of Ytterbium Yb3 + ions with participating Erbium (Er\(^{3+}\)) or Thulium (Tm3 +) ions, respectively. Yb3 + and Er3 + (Tm3 +) serve as the sensitizer and activator, respectively. Luminescent spectra of (c) NaYF4:Yb:Er; and (d) NaYF4:Yb:Tm nanomaterials, with their corresponding transmission electron microscopy images

This process is described by a set of kinetic equations of the Er3 + or Tm3 + energy level populations. Levels 1 (ground), 2, and 3 refer, respectively, to the energy states 4I15 ∕ 2, 4I11 ∕ 2, and 4F9 ∕ 2 (or 4S3 ∕ 2) – for Erbium ions, and 3H6, 3H5, and 3H4 – for Thulium ions, as shown in Fig. 14.9a. A simplified energy level diagram is presented in Fig. 14.9b.
$$\displaystyle\begin{array}{rcl} dN_{3}/dt& =& W_{t}N_{2}^{2} - (W_{ 31} + W_{32})N_{3}\end{array}$$
$$\displaystyle\begin{array}{rcl} dN_{2}/dt& =& -2W_{t}N_{2}^{2} - W_{ 2}N_{2} + W_{32}N_{3}\end{array}$$
$$\displaystyle\begin{array}{rcl} dN_{1}/dt& =& W_{t}N_{2}^{2} + W_{ 2}N_{2} + W_{31}N_{3}\end{array}$$
where N 1, N 2, N 3 are populations of the levels 1, 2, 3, respectively; W 31, W 32 transition probabilities 3 → 1 and 3 → 2, respectively; W t the upconversion probability, resulting in simultaneous transition of one ion to the level 3, and the other to the ground level 1. By analyzing the equation, one can conclude that at the low excitation intensity (I), the signal luminescence is proportional to I2 in virtue of the N 2 [59]. Thus, it becomes clear that the power quantum yield (PQY) of the UCNP is a linear function of the pumping intensity, and its measurement versus the intensity is the key to optimal design of the synthesis of UCNP. It is also important to note that the level 3 decay rate is governed by the dependence of exp( − W 31 t) in case of the pulsed excitation, with t = 0 set to the excitation pulse falling edge. W 31 contains both radiative ( ≈ 5,000 s− 1 [58]) and non-radiative components that compete for the level 3 population decay. When the nanocrystal size becomes small ( < 30 nm), the non-radiative component ( ≈ 20,000 s− 1) prevails causing a dramatic reduction of PQY [60]. This poses a challenge on the production and surface passivation of UCNP in enhancing the upconversion energy transfer W t , preserving the crystal quality, and minimizing bulk and surface defects – to suppress the non-radiative component of W 31. The former is addressed by the breakthrough synthesis, and is described below.
The most popular UCNP represents an inorganic nanocrystal matrix (NaYF4) co-doped with sensitizer ytterbium (Yb3 +) and activator erbium (Er3 +) or thulium (Tm3 +) lanthanide ions. In a particular case of NaYF4:Yb:Tm, an ensemble of Yb-ions sensitizes infrared radiation at a wavelength of 980 nm and transfers it to Tm-ions characterized by multiple excited states with exceptionally long (sub-ms) lifetimes. As a result of the energy transfer upconversion, Tm3 + radiates at 474 nm (three sequential photons) and 800 nm (two sequential photons) spectral bands (Fig. 14.9d). Since the UCNP excitation/emission process is nonlinear, power quantum yield (η), i.e., the emission/excitation (Watts/Watts) power ratio, grows linearly versus the excitation intensity reaching saturation at a high intensity value of I\(_{\mathrm{sat}} \approx \) 10–1,000 W cm− 2 [59], as shown in Fig. 14.10. Until recently, the dramatic reduction of η versus the crystal host size at the nanometer scale precluded practical applications of UCNPs. A breakthrough in the UCNP synthesis [61, 62] has resulted in much brighter nanoparticles, where “brightness” is measured in terms of the power quantum yield.
Fig. 14.10

Plots of the power quantum yield, η of (a) α- and (b) β-phases UCNPs. Particle mean diameters, (a) 10 nm and (b) 60 nm. The saturation intensity is (a) > 300 W cm− 2, (b) 70 W cm− 2

14.3.2 Production and Characterization of UCNP

An oxygen-free two-step protocol is used for the synthesis of the UCNPs [63]. The typical sample of NaYF4:Yb:Er has molar concentrations of Yb and Er as 18 and 2 %, respectively. In step 1, aqueous solution of LnCl3 (1.0 mM, Ln = Y, Yb, Er) is mixed with oleic acid and octadecane in a three-neck, round-bottom flask and heated first at 150 C and at 110 C on the addition of NH4F and NaOH. This procedure yields small-size ( < 10 nm) NaYF4:Yb:Er of cubic crystal phase (α-phase) characterized by low η. Step 2 is performed to convert these nano-crystallites in to the hexagonal phase (β-phase). The reaction mixture is sealed, quickly heated to 320 C, while purged with argon, and nanocrystals are precipitated with acetone. As a result, one obtains nanocrystal powder of sizes ranging from < 10 to 300 nm. This synthesis has been quickly adopted and modified to produce a great variety of UCNPs of different sizes (Fig. 14.9 insets) and spectral and temporal upconversion properties [64].

The power conversion coefficient has been greatly improved as a result of this new synthesis approach. Our measurements carried out using an integrating sphere showed a dramatic improvement of η of the α-UCNP from 0.03% (particle diameter, 10 nm) to 2 % of the β-UCNP (particle size, 60 nm) with a saturation intensity of ≈ 70 W cm− 2.

14.3.3 Merits of the UCNP-Assisted Optical Biomedical Imaging

There are several key advantages of UCNP in the context of biomedical optical imaging. Firstly, the NaYF4:Yb:Tm luminescence band (800 nm, see Fig. 14.9d) falls into the biological tissue transparency window. The excitation and emission radiation in biological tissue experiences reduced absorption and scattering in comparison with that in the ultraviolet (UV) and infrared (IR) spectral ranges. The biotissue absorption is dominated by oxygenated and deoxygenated hemoglobin constituents of blood, proteins (e.g., melanin) in the UV visible range of the spectrum, and by water in the IR spectral range. The biotissue scattering extinction represents a monotonic function slowly decreasing from the UV toward IR. The combined effect of the biotissue scattering/absorption is calculated and plotted in Fig. 14.1a using the following light transport equation, in the diffusion approximation, to describe the optical response of the tissue: \(\mu _{\mathrm{eff}} = {(3\mu _{a}[\mu _{a} +\mu _{s}(1 - g)])}^{1/2}\), where absorption μ a and scattering μ s coefficients and anisotropy factor g for human skin epidermis and dermis layers (see Fig. 14.1b) were chosen for this calculation [65]. As one can see, absorption of the excitation light at 980 nm by water constituent and hemoglobin of the biological tissue, such as skin (Fig. 14.1a), is compensated by the reduced scattering of biotissue at this wavelength, and the UCNP-pertinent excitation/emission properties are advantageous for biomedical optical imaging.

Secondly, UCNP excitation at 980 nm elicits very little autofluorescence from biological tissue. Presuming a small autofluorescence signal, it has to be Stokes-shifted to the longer-wavelength, whereas the detection takes place in the shorter wavelength band ranging from UV to near-IR depending on the UCNP dopant structure. As we have demonstrated, UCNP topically applied both on freshly excised human skin and micro-needle-delivered in the dermis layer exhibited ultimately high imaging contrast, with the biotissue-induced laser light scattering and autofluorescence background totally suppressed (the background level was detected at the electronic noise level using a high-end cooled CCD camera). The acquired epi-luminescence images of skin autofluorescence and UCNP distribution excited, respectively, at 365 and 980 nm are shown in Fig. 14.11.
Fig. 14.11

NaYF4:Yb:Tm nanoparticle distribution in human skin in case of (top two rows) the topical application, and (bottom row) micro-needle treatment. (a), (d), (g), UV (365 nm)-excited autofluorescence images of skin; (b), (e), (h), images of UCNP excited by 980-nm laser; (c), (f), (i), pseudo-color overlaid images of (a), (d), (g) showing UCNPs (purple color) in skin folds and dermis (green color), respectively

It turns out that the excitation light scattering bled through the interference filters contributes more to the imaging background than the autofluorescence signal, which is an unusual situation in optical biomedical imaging. In order to counter the excitation light bleeding into the detection path, the detection band can be spectrally offset by as much as 330 nm from the excitation band, by picking the NaYF4:Yb:Er upconversion emission multiplet centered at 550 nm (Fig. 14.9c, green shaded band) [66]. The other major advantage of UCNP lies in a million-fold difference in the emission lifetimes of UCNP and endogenous fluorophores that make up the autofluorescence background (\(\tau _{uc} \approx 1\) ms versus \(\tau _{af} \approx 3\) ns) that allows complete suppression of the autofluorescence and excitation background by an optical time-gated approach alone, thus avoiding the use of expensive interference filters. This scheme is realized by setting a time delay ( ≈ 10 ms) between the excitation laser pulse and photodetection of the sample response. The long τ uc of UCNP also lead to the enhanced efficiency of Förster resonance energy transfer (FRET) processes, used to investigate protein interactions [67].

14.3.4 Optical Imaging of UCNP in Biological Samples

The promise of upconversion nanoparticles has been recently demonstrated through the imaging of UCNP bioconjugates in cell cultures [4, 68] and whole animal models [6, 69], with the autofluorescence background suppressed. Figure 14.12 shows live-cell imaging of UCNPs in cells by using laser-scanning luminescence confocal microscopy [4]. The confocal imaging modality allows high-sensitivity detection since the high intensity at the sub-femtoliter focal excitation volume, such as 10 ×105 W cm− 2 (used to obtain images in Fig. 14.11), is easily achievable. Evading the signal cross talks caused by the multiple scattering of the luminescence photons in the biological samples is important for single nanoparticle imaging, which would otherwise be obscured by intense signals from UCNP clusters. At the same time, the confocal imaging modality suffers from long image acquisition times arising from the necessity to dwell for several τ uc -s on each acquisition pixel. As an example, consider \(\tau _{uc} \approx 1\) ms that requires, at least, 5 ms per pixel and c.a. 1/2 h to capture a 512 ×512 pixel image – this is unaffordable for most practical applications.
Fig. 14.12

Live-cell imaging of UCNPs in NIH 3T3 murine fibroblasts using laser-scanning luminescence confocal microscopy. (a) Bright field image of a cell with endocytosed UCNPs, (b) upconverted luminescence following 980-nm excitation, and (c) overlay. Scale bar, 10 μm

Intensive research activity has been streaming toward optical tomography of UCNP-tagged tissue sites, where upconversion nanoparticles are deployed as the molecular probes. In particular, the whole animal imaging has been reported on a number of occasions, where the UCNP-assisted target delivery and optical tomography imaging of cancer tumor sites demonstrated the imaging depth of up to 2 cm in transmission mode [70].

This raises the question on the ultimate sensitivity limit achievable for the UCNP imaging in biological tissue. In order to answer this question, we performed the single UCNP imaging through a layer of the hemolyzed blood using full-field optical microscopy configuration. It was realized that as long as the scattering of the biological sample was not overwhelming, such as in the case of the hemolyzed blood, sample absorption presented little problem for imaging, and no degradation of the imaging of a single (UCNP) nanoparticle was observed (see Fig. 14.13).
Fig. 14.13

Optical epi-luminescence microscopy of a single UCNPs deposited on a glass slide using the “blood immersion” objective lens shown in (a). (b) and (c) Images of the constellation of UCNPs acquired using correspondingly water- and blood-immersion objective lenses. The images were matched to those acquired by transmission electron microscopy. A single UCNP particle is encircled. The epi-luminescence image acquisition time is 1 s

14.3.5 Shortcomings of the UCNP Technology

These studies have revealed shortcomings of the UCNP technology. These include limited penetration depth of the UCNP-assisted optical imaging in whole animals because of the requirement of high excitation intensity that is usually realized by focusing, and is not possible in turbid biological tissue [66]. This point is illustrated in Fig. 14.14, where a comparison between the optical tomography imaging deploying organic fluorescence dye and UCNP in an animal model is presented [66]. Although the UCNP-assisted optical tomography provided much more accurate localization of the luminescence probe (due to the background-free imaging and nonlinear dependency of the luminescence signal versus the excitation intensity), the sensitivity of the UCNP detection is inferior to that of the conventional infrared Cy5 fluorescence dye. It is important to note that, unlike in the case of Cy5, the UCNP signal intensity is quickly degraded versus the imaging depth. This sets an inherent limit on the applicability of the UCNPs for deep tissue imaging, especially considering that the maximum intensity allowed at 980 nm in humans is limited to 726 mW cm− 2. Besides, the spectral filtering efficiency of the UCNP luminescence is compromised by the broad angular distribution of unwanted background photons emerging from biological tissue (10 ×106-fold versus 10 ×103-fold [71]), although this can be mitigated by the time-gated imaging mode. These findings call for redefinition of the target applications, which will capitilize on the key UCNP merits, while minimizing their limitations. One such application is ultrasensitive imaging of UCNPs in thin tissue slices, such as skin, aiming at demonstration of background-free imaging of nanoparticles paving a way toward ultrasensitive in vivo imaging of nanoparticle penetration in skin.
Fig. 14.14

Left panel (top): Image of a mouse with a thin capillary tube inserted into the esophagus. The tube is filled with Cy5.5 dye (excitation at 680 nm and emission at 712 nm). Left panel (bottom): The same mouse with UCNP instead of the dye (excitation at 980 nm and emission at 670 nm). The absence of autofluorescence in the bottom image is obvious. However, a much stronger excitation power is required in the case of UCNP probe. This is illustrated on the right panel where the decay of the signal is shown, as a function of the tissue imaging depth. The signal decays much faster for the UCNP, because it is proportional to the intensity in the power of 2.3 (Reproduced from Ref. [66])

14.4 Luminescent Nanoparticle Bioconjugation

In most cases, living systems treat extraneous nanomaterial, as an unwelcome intruder, and engage disposal protocols. In order to usher in a more meaningful communication with cells, a nanoparticle needs to be decorated with biomolecule(s) whose role is to establish a communication interface between the nanoparticle and the cell. A luminescent nanoparticle participating is such communication events furnished by biomolecular cascades can report on the molecular trafficking and serve diagnostic and therapeutic purposes. Therefore, grafting biomolecules onto the nanoparticle surface represents the key procedure underlying the convergence of Nanotechnology with Life Sciences.

14.4.1 Colloidal Stability in Water and Buffer Solutions

Deployment of luminescent nanoparticle in biomedical applications critically depends on NP bioconjugate stability in physiological solutions, currently an issue, as oftentimes synthesized NPs, such as upconversion NPs, are hydrophobic. Their dispersion in water results in rapid formation of aggregates that precipitate in the form of an insoluble residue. However, hydrophilicity of some NPs, such as the surface-oxygenated nanodiamonds, does not guarantee LND aqueous colloidal stability; unless the functional groups the NP surface acquire sufficient charge in water. This surface charge is measured in terms of the zeta-potential units, mV. Surface-oxygenated nanodiamonds are stable in water at the zeta-potential absolute value greater than 30 mV. However, when the nanodiamond aqueous colloid is transferred to buffer, salt-induced surface charge screening leads to flocculation [72]. Existing NP surface 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. 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. In the following two subsections, we address two surface functionalization and bioconjugation strategies using the examples of LND and UCNP to illustrate the main aspects of this side of bio-nanotechnology.

14.4.2 A Universal Bioconjugation Platform Based on a High-Affinity Molecular Pair Barnase:Barstar

Most of the nanodiamond surface cleaning procedures promotes formation of oxygen-containing surface functional groups, including carboxyls, hydroxyls, ketons, and aldehydes. Many of these groups are biologically compatible, thus forming a lucrative possibility for the direct covalent attachment of biomolecules using, e.g., a carboxyl group (COOH). COOH is universally employed in living systems to assemble proteins by fusing together a COOH terminal of one protein with an amino-terminal (NH2) of the other protein, forming a strong covalent amide bond. Since COOH-abundance of the acid-cleaned nanodiamond surface is c.a. 7 %, it is conceivable to make use of this terminal to attach a biomolecule containing a NH2-terminal. The implementation of this strategy is not straightforward due to the notorious colloidal instability of nanodiamonds in saline solutions, which are essential to maintain the structural conformation of biomolecules. Also, amide-bonding reactions can alter the functionality of biomolecules and, hence, need adjustments specific to each biomolecule. This is cumbersome, and calls for a universal straightforward bioconjugation approach. A novel nanoparticle-bioconjugation platform has been recently reported [73]. This approach is based on a high-affinity protein pair, barstar:barnase (Bs:Bn), a functional analogue of streptavidin:biotin, that provides a modular design toolkit by way of locking organic and inorganic modules together in a straightforward and robust manner. The LND biofunctionalization strategy based on Bs:Bn is as follows: A covalently assembled subunit LND-Bs (or LND-Bn) strongly attaches to a prefabricated counter subunit Bn-X (or Bs-X) by simply mixing the two colloidal substrates, where X is a terminal molecule/nanoparticle. The terminal molecule can be biologically significant, with a potential for targeted drug delivery and/or biolabeling applications (see Fig. 14.15).
Fig. 14.15

Diagram of a universal platform for bioconjugation of macromolecules and nanoparticles to the nano-vehicle surface, in this example, nanodiamond. Covalently grafted to the diamond surface molecules, barstar (green caps) are able to selectively bind to molecules of barnase (brown cones), which, in turn, are pre-conjugated with such macromolecules as green fluorescent protein (fragment), antibodies (fragment at the top), or nanoparticles, e.g., nanogold (fragment). Transmission electron microscopy image shows the assembly of nanodiamond-nanogold, called “nano-diadem” (inset) [34]

Non-cytotoxic, biocompatible, and bright luminescent nanodiamonds (LNDs) represented a particularly attractive “docking station” due to their appeal for molecular probe and targeted delivery applications. The modified EDC/sNHS reaction yielded a covalently bonded 140- and 35-nm LND-Bs conjugate, which showed remarkable colloidal stability in buffer solutions. Using this platform, LND-Bs:Bn-(green fluorescent protein) and LND-Bs:Bn-nanogold were synthesized and tested in saline solutions and cellular environments. Assaying and optical/TEM imaging of these synthesized organic and inorganic complexes revealed their functional stability and structural integrity [34]. Most recently, the authors (AZ, SD, et al.) have demonstrated implementation of the same protocol to obtain antibody bioconjugates of upconversion nanoparticles UCNP-Bs:Bn-4D5, where 4D5 represents a recombinant mini-antibody to the HER2/neu receptor, a type of epithelial growth factor receptor.

14.4.3 Bioconjugates of Upconversion Nanoparticle Coated with Amphiphilic polymer

The bioconjugation strategy described for nanodiamonds represents an exceptional case of nanoparticles featuring biocompatible surface moieties, and often stable in aqueous colloids. In general, inorganic nanoparticles are non-mixable with water, and their surface provides no anchoring points for attachment of biomolecules. Considering a number of interesting luminescent nanoparticles emerging as a result of the rapid developments in Nanotechnology, a universal bioconjugation approach is highly demanded. Upconversion nanoparticles based on the popular nanocrystalline host NaYF4 represent one of the most challenging cases. Firstly, as-synthesized UCNPs are surface-coordinated with oleic groups that render this nanomaterial hydrophobic, i.e., not mixable with water and buffer solutions. Secondly, NaYF4 does not seem to provide amiable surface anchoring points for firm docking of biomolecules or auxiliary moieties. Thirdly, the UCNP power quantum yield appeared to be highly susceptible to the surface functionalization, where high-energy vibronic modes of some functional groups, such as COOH, provided non-radiative relaxation path to the upper excited states of Er or Tm in NaYF4:Yb:Er(Tm) [64]. Early attempts to displace the weakly bonded oleic groups with more affine moieties, such as citric [4] and mercaptopropionic [74] groups, have demonstrated only marginal colloidal stability of UCNPs at the neutral pH (pH 7) values, precluding biomedical applications, where pH varies widely. This implies that the functional group attachment to the UCNP surface was governed by electrostatic and/or Van der Waals forces, i.e., attachment by adsorption. The absorption-based biofunctionalization protocol has been employed by Niedbala and coworkers to attach NeutrAvidin to the UCNP surface by a facile pot-mix reaction [75]. This approach is subject to desorption of the adsorbed molecules in biological solutions, and also depends on the pH of the medium. A common solution is to cover the surface of nanoparticles by polyethylene-glycol (PEG). These groups prevent nanoparticle merging and forming of aggregates, which adversely affect the colloidal stability, via the mechanism of steric hindrance [76]. Also, PEGylation represents the core of the so-called stealth technology preventing PEGylated nanoparticles intravenously injected into the blood stream from rapid immune-mediated removal. However, PEG groups have their disadvantages associated with poorly controlled polymer chains, and, most importantly, PEG chemistry is expensive.

A radical solution to avoid the dependency of bioconjugation reactions on the UCNP surface anchoring lies in coating nanoparticles with an additional layer that is amiable for subsequent bioconjugation. Coating the UCNP nanoparticle with a silica shell [77, 78] represents an attractive approach due to the shell stability in the range of pH, and maturity of the silica surface coating technology [77]. Another approach makes use of amphiphilic polymers that represent molecules with hydrophobic and hydrophilic terminals. This approach has been successfully demonstrated for surface activation of quantum dots [79], and partly accounted for the commercial success of quantum dots-based procedures. An amphiphilic polymer was wrapped around the particles, exploiting the nonspecific hydrophobic interactions between the alkyl chains of poly(maleic anhydride alt-1-tetradecene) and the nanocrystal surfactant molecules (oleic acids). This resulted in transferring hydrophobically capped nanocrystals from organic to aqueous solution. Addition of bis(6-aminohexyl)amine resulted in the cross-linking of polymer chains around each nanoparticle [79]. Our initial experiments using this protocol have resulted in UCNPs being stable in water and buffer solutions for a month. Nanocrystals coated with COOH-amphiphilic polymer [79] are readily processed using a universal bioconjugation protocol, i.e., EDC/sNHS reaction, which we have demonstrated by attaching streptavidin to UCNP, as presented in Fig. 14.16. A facile pot-mix reaction will lead to a strong attachment of a targeting biomolecule of choice, if this molecule is biotinylated. A range of commercially available biotinylated ligands and antibodies, including antibodies to epitope tags, permits versatile design of various bioconjugates that are linked together via a strong bond between streptavidin and biotin. An alternative approach, demonstrated by us, is based on a high-affinity protein pair barnase:barstar that makes use of the advantageous recombinant protein technology [34].
Fig. 14.16

Fluorescent microscopy images of UCNP bioconjugated with streptavidin-dylight (dylight – organic dye) obtained under excitations of 470 and 980 nm, and overlaid false-color image. Colocalization of several nanoparticles (circled) is clearly observed, demonstrating successful bioconjugation of streptavidin to UCNP

14.5 Communication of Nanoparticles and Cells

To sustain the lifecycle of living cells, extraneous material has to be absorbed from the extracellular medium. The absorption mechanisms fall into two categories: specific internalization, which requires the cell to actively recruit molecules into the cytoplasm, and nonspecific internalization, which is essentially a random process in which the cell exercises no active control; it can vary depending on the cell type and has poor material selectivity. Receptor-mediated endocytosis is an important example of the specific internalization mechanism that facilitates import of selected extracellular molecules and mediates intercellular signaling. This process is regulated by plasma membrane receptors that are predominantly activated by receptor-specific ligands. As a result, only biomolecular complexes grafted with these ligands can activate these receptors and gain access into the cell, and these represent a focus of this book chapter. In the process of endocytosis, the plasma membrane is invaginated inward from specialized membrane micro-domains forming either clathrin- or caveolin-coated pits. This specific cellular uptake mechanism has been a subject of intense research driven by the demand to elucidate cellular molecular trafficking and potential applications in targeted drug delivery.

Aiming at achievement of the highest possible target delivery specificity, we need to understand nonspecific (in contrast with the specific internalization via receptor-mediated endocytosis) internalization of nanoparticles that depends on many parameters of nanoparticles, such as size, surface charge, and surface functional groups, as we have presented in a separate publication on the example of quantum dots [80], and shown in Fig. 14.17. Charge affects the level of nanoparticle uptake in the cells considerably, and should be taken into account, especially when conducting experiments with cell cultures. Despite that, the mechanism of the cellular uptake is defined as nonspecific, it is mediated by the receptors, and realized either through direct interaction between charged particles and the receptor or through a preliminary attachment of free protein to the NP surface. Surface modification of nanoparticles with hydrophilic polymers, such as PEG, inhibits the nanoparticle uptake, making it ideally suitable for the negative controls to various cellular models, where the terminal amino group suppressed internalization significantly, as shown in Fig. 14.17 (columns, N-pQD e , C-pQD e ). It is expected that amino or carboxyl surface functionalization of NP, deprived of PEG, will facilitate intensive uptake of particles in the cell, despite the positive and negative terminal charges. However, the positively charged nanoparticles are absorbed less effectively, and represent a good choice for the experiments on specific labeling and activation of the cells.
Fig. 14.17

Images showing the level of the quantum dot (QD) uptake in three immortalized cell cultures AR42J, AtT20, and GH4C1 depending on the surface of functional groups. Direct attachment of amino groups (far left column, N-CC-QDi) shows the highest level of internalization, while a PEG-linker (amino-terminal, N-pQD e and carboxyl-terminal, C-pQD e ) heavily suppresses internalization of nanoparticles (Reproduced from Ref. [80])

In order to investigate molecular trafficking using NP bioconjugates, the biomolecular part of the NP module must activate the biological process. In case of cell communication, it is often a ligand attached to the NP that activates the cell receptor, and thereby launches a cascade of biomolecular events that may be regarded as a process of communication of the cell and the nanoparticle. As a result, the particle is translocated to the cell cytoplasm, where it participates in the other sequence of the post-endocytosis events. As an example of the ligand that activates the communication protocol with the cell, we consider somatostatin (SST). Among the multitude of physiological functions that SST participates in, some are of high clinical relevance. The biological functions of SST are initiated upon its binding to trans-membrane G-protein-coupled receptors of six subtypes, including sst2A , which are differentially distributed in organs such as brain, spinal cord, and pancreas, and over-expressed in neuroendocrine tumor cells. The physiological effects of SST, evoked after ligand binding and receptor activation, are mediated through intracellular signaling mechanisms including the adenylyl cyclase inhibition, K+ channel opening, and, in some cases, the activation of phospholipase C leading to the opening of intracellular Ca2 + stores [81].

Somatostatin analogs, bearing a contrast agent, are useful for a variety of applications ranging from basic research into elucidating the post-endocytotic molecular pathways in cells to clinical scenarios, such as the targeting/diagnosis of neuroendocrine tumors. Prior to demonstration of the NP bioconjugate complex of NP-SST designed for specific target internalization in cells, the somatostatin activity and potency was tested using cell cultures that were activated by SST, together with the demonstration of the specific molecular processes that accompany receptor-mediated endocytosis.

For example, it is known that the specific receptor sst2A activation triggers a cascade of processes, such as the opening of the calcium repositories in the cytoplasm, in genetically modified Chinese hamster ovary cells (CHO-K1). The design, production, and pharmacological characterization of a recombinant fluorescent SST analogue fused with a monomeric red fluorescent protein (SST-mRFP) has been reported [82], with the main result shown in Fig. 14.18. The SST moiety in the SST-mRFP complex was demonstrated to be active and potent for triggering signaling in both wild-type and transfected cells containing somatostatin receptors. The fluorescent complex of SST was internalized, as a result of the processes launched by the receptor activation, leading to the ligand translocation to the cellular cytoplasm in the process of endocytosis, and eventual accumulation in the perinuclear region (Fig. 14.18, bottom-right panel). It is interesting to note that SST and sst2A remain colocalized even at the later stage of the endocytosis, being encapsulated either in the late endosomes or lysosomes, and visualized in a yellow-orange color in Fig. 14.18, bottom-right panel. This shows potential of fluorescent ligand complexes to elucidate the molecular fate of post-endocytotic events, which in this particular case demonstrate the receptor-ligand integrity.
Fig. 14.18

(Top) A diagram of specific internalization of peptide somatostatin fused with a red fluorescent protein (RFP) (center) by binding and activating receptors sst2A (green, upper left panel) in Chinese hamster ovary cells (CHO-K1). As a result of the endocytosis, SST and sst2A are taken inside the cells (right top panel). (Bottom) Fluorescent confocal images of the CHO-K1 cells: (right) showing the colocalization of peptide somatostatin/red fluorescent protein (SST-mRFP) with the sst2A receptor antibodies, labeled with fluorescent dye FITC, rendered in orange. The image on the left shows a negative control, obtained by the inhibition of SST-RFP (Reproduced from Ref. [82])

The development of a somatostatin analogue that bears a luminescent nanoparticle contrast agent represented by a quantum dot has been recently reported. In virtue of exceptional luminescence properties and commercial availability of various surface functionalized types, QD was selected as the model nanoparticle. A biotinylated analogue of SST (SST-2B) was conjugated to a streptavidin-coated QD (Sav-QD), forming an SST-2B:Sav-QD complex. However, this preformed SST-2B:Sav-QD complex was incapable to bind/activate the somatostatin receptors. In order to mitigate this problem, an in situ two-step conjugation strategy was introduced to overcome the Sav-induced inhibitory effect, an approach that proved to be successful for specific targeting of the somatostatin receptors. Using this strategy, the receptor-mediated endocytosis of the nanoparticle-labeled SST was optically imaged. Despite the original inactivity, SST-2B and similar biotinylated SST-ligands are envisaged to be made functional again using a two-step approach. Their deployment for the targeted delivery into cells that express somatostatin receptors, e.g., in brain regions responsible for the regulation of blood pressure will be valuable (see Fig. 14.19).
Fig. 14.19

Fluorescence confocal images of cells treated with (a) 20-μM SST-2B, also known as SRIF-2B, and (b) blank negative control, followed by 2-nM Sav-QD at low temperature. The cells, when restored to 37 C, internalized the pre-formed receptor:ligand complexes. The QD-luminescence is color-coded red and the cell-morphological contrast, originated from laser back-scattering, is color-coded green. The data is representative of at least three independent experiments. Scale-bar, 10 μm

14.6 Conclusion

We described a bottom-up strategy of luminescent nanoparticle-assisted imaging of cellular processes, such as the receptor-mediated endocytosis and associated molecular events. This strategy has three main building blocks: bright photostable luminescent nanomaterials conspicuous on the background of cell autofluorescence, surface modification to enable interfacing with the biological world, and modular engineering of the biomolecular complexes equipped with a ligand for programmable functions in the cell. Luminescent nanodiamonds (LND) and upconversion nanoparticles (UCNP) were selected as representatives of different classes of luminescent emitters, and their photophysics was discussed in the context of optical biological imaging. The rapidly growing area of nanotechnology offers a number of lucrative possibilities of nanomaterial deployment in bio-nano-optics applications. The nanomaterial choice, however, needs to be carefully scrutinized based on the material’s physical, chemical, and optical properties, where the core chemical inertness, amiable surface chemistry, and minimal cytotoxicity are essential. Optical properties remain the key factors determining the utility of these nanomaterials, where the nanoparticle brightness defined as the action cross section is a necessary, but not sufficient, prerequisite. It appears that the imaging contrast of luminescent nanoparticle against the crowded morphological and autofluorescence background of the cell needs to be taken into account. Despite the lower action cross-section of the UCNP, they offer excellent means for discrimination of the excitation and autofluorescence background paving a way toward ultimate detection sensitivity at the single-molecule level.

Interfacing of nanoparticles with biomolecules using bioconjugation procedures poses a challenge due to the poor dispersability of (often hydrophobic) nanoparticles in aqueous/buffer colloids exacerbated by the surface inability to anchor biomolecules or intermediate linkers. Luminescent nanodiamond represents a fortunate exception, whose surface is decorated with biologically amiable carbonyl groups that enable direct realization of a modular assembly of a LND bioconjugate in a facile pot-mix reaction. Hydrophobic UCNPs require several extra-procedures to render these particles mixable with buffer solutions and graft COOH-groups on their surface. This is achieved by complete surface reshaping with a shell comprising, e.g., amphiphilic polymer prefabricated with COOH-groups that renders UCNP surface amenable to bioconjugation.

Biological properties of the nanoparticle bioconjugates are determined by a ligand firmly attached to the nanoparticle surface, and the choice of ligands and participating biological systems is substantial. We chose a peptide somatostatin, as the ligand, to demonstrate the potential of the luminescent nanoparticle tagging to elucidate its molecular pathway during the endocytotic and post-endocytotic events.

The development of luminescent nanoparticle bioconjugates capable of programmable interactions with the targeted cells opens new opportunities in several areas of life sciences, including new imaging modalities, enabling long time-lapse investigation of molecular trafficking in cells, cellular imaging at the unprecedented single-molecule level, in vivo diagnosis based on ultrahigh-sensitivity optical tomography imaging, and, targeted drug delivery. This work sets first grounds for these exciting future prospects.


  1. 1.
    W.E. Moerner: Proc. Natl. Acad. Sci. 104, 12596 (2007)ADSCrossRefGoogle Scholar
  2. 2.
    G. Seisenberger, M.U. Ried, T. Endre, H. Bning, M. Hallek, C. Bruchle: Science 294, 1929 (2001)ADSCrossRefGoogle Scholar
  3. 3.
    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, W. Fann: Proc. Natl. Acad. Sci. 104, 727 (2007)ADSCrossRefGoogle Scholar
  4. 4.
    S.W. Wu, G. Han, D.J. Milliron, S. Aloni, V. Altoe, D.V. Talapin, B.E. Cohen, P.J. Schuck: Proc. Natl. Acad. Sci. U.S.A. 106, 10917 (2009)ADSCrossRefGoogle Scholar
  5. 5.
    A.M. Schrand, H. Huang, C. Carlson, J.J. Schlager, E. sawa, S.M. Hussain, L. Dai: J. Phys. Chem. B 111, 2 (2006)Google Scholar
  6. 6.
    T.Y. Cao, Y. Yang, Y.A. Gao, J. Zhou, Z.Q. Li, F.Y. Li: Biomaterials 32, 2959 (2011)CrossRefGoogle Scholar
  7. 7.
    J.L. Yuan, G.L. Wang: Trac-Trends in Analytical Chemistry, 25, 490 (2006)CrossRefGoogle Scholar
  8. 8.
    S. Osswald, G. Yushin, V. Mochalin, S.O. Kucheyev, Y. Gogotsi: J. Am. Chem. Soc. 128, 11635 (2006)CrossRefGoogle Scholar
  9. 9.
    V. Dolmatov, M. Veretennikova, V. Marchukov, V. Sushchev: Phys. Solid State 46, 611 (2004)ADSCrossRefGoogle Scholar
  10. 10.
    E. Osawa: Pure Appl. Chem. 80, 1365 (2008)CrossRefGoogle Scholar
  11. 11.
    S. Turner, O.I. Lebedev, O. Shenderova, I.I. Vlasov, J. Verbeeck, G. Van Tendeloo: Adv. Funct. Mater. 19, 2116 (2009)CrossRefGoogle Scholar
  12. 12.
    F.K. de Theije, O. Roy, N.J. van der Laag, W.J.P. van Enckevort: Diam. Relat. Mater. 9, 929 (2000)ADSCrossRefGoogle Scholar
  13. 13.
    F.K. de Theije, N.J. van der Laag, M. Plomp, W.J.P. van Enckevort: PhilosophicalMagazine A 80, 725 (2000)ADSGoogle Scholar
  14. 14.
    R.R. Nimmagadda, A. Joshi, W.L. Hsu: J. Mater. Res. 5, 2445 (1990)ADSCrossRefGoogle Scholar
  15. 15.
    J. Tisler, G. Balasubramanian, B. Naydenov, R. Kolesov, B. Grotz, R. Reuter, J.-P. Boudou, P.A. Curmi, M. Sennour, A. Thorel, M. Borsch, K. Aulenbacher, R. Erdmann, P.R. Hemmer, F. Jelezko, J. Wrachtrup: ACS Nano 3, 1959 (2009)CrossRefGoogle Scholar
  16. 16.
    A. Krueger, M. Ozawa, G. Jarre, Y. Liang, J. Stegk, L. Lu: Phys. Status Solidi A 204, 2881 (2007)ADSCrossRefGoogle Scholar
  17. 17.
    B.R. Smith, M. Niebert, T. Plakhotnik, A.V. Zvyagin: J. Lumin. 127, 260 (2007)CrossRefGoogle Scholar
  18. 18.
    I.I. Vlasov, A.S. Barnard, V.G. Ralchenko, O.I. Lebedev, M.V. Kanzyuba, A.V. Saveliev, V.I. Konov, E. Goovaerts: Adv. Mater. 21, 808 (2009)CrossRefGoogle Scholar
  19. 19.
    A. Smith, A. Mainwood, M. Watkins: Diam. Relat. Mater. 11, 312 (2002)ADSCrossRefGoogle Scholar
  20. 20.
    I. Aharonovich, S. Castelletto, D.A. Simpson, A.D. Greentree, S. Prawer: Phys. Rev. A 81, 043813 (2010)ADSCrossRefGoogle Scholar
  21. 21.
    A. Gruber, A. Drabenstedt, C. Tietz, L. Fleury, J. Wrachtrup, C. von Borczyskowski: Science 276, 2012 (1997)CrossRefGoogle Scholar
  22. 22.
    L. Rondin, G. Dantelle, A. Slablab, F. Grosshans, F. Treussart, P. Bergonzo, S. Perruchas, T. Gacoin, M. Chaigneau, H.C. Chang, V. Jacques, J.F. Roch: Phys. Rev. B 82, 115449 (2010)ADSCrossRefGoogle Scholar
  23. 23.
    T.-L. Wee, Y.-K. Tzeng, C.-C. Han, H.-C. Chang, W. Fann, J.-H. Hsu, K.-M. Chen, Y.-C. Yu: J. Phys. Chem. A 111, 9379 (2007)CrossRefGoogle Scholar
  24. 24.
    A.V. Zvyagin, N.B. Manson: Adv. Nanodiam. Sci. Technol. in press (2012)Google Scholar
  25. 25.
    A. Batalov, C. Zierl, T. Gaebel, P. Neumann, I.Y. Chan, G. Balasubramanian, P.R. Hemmer, F. Jelezko, J. Wrachtrup: Phys. Rev. Lett. 100, 077401 (2008)ADSCrossRefGoogle Scholar
  26. 26.
    J.R. Maze, P.L. Stanwix, J.S. Hodges, S. Hong, J.M. Taylor, P. Cappellaro, L. Jiang, M.V.G. Dutt, E. Togan, A.S. Zibrov, A. Yacoby, R.L. Walsworth, M.D. Lukin: Nature 455, 644 (2008)ADSCrossRefGoogle Scholar
  27. 27.
    F. Cichos, C. von Borczyskowski, M. Orrit: Science 12, 272 (2007)Google Scholar
  28. 28.
    R.M. Dickson, A.B. Cubitt, R.Y. Tsien, W.E. Moerner: Nature 388, 355 (1997)ADSCrossRefGoogle Scholar
  29. 29.
    M. Kuno, D.P. Fromm, H.F. Hamann, A. Gallagher, D.J. Nesbitt: J. Chem. Phys. 112, 3117 (2000)ADSCrossRefGoogle Scholar
  30. 30.
    P.A. Frantsuzov, R.A. Marcus: Phys. Rev. B 72, 155321 (2005)ADSCrossRefGoogle Scholar
  31. 31.
    T. Jau, R.A. Marcus: J. Chem. Phys. 123, 054704 (2005)ADSCrossRefGoogle Scholar
  32. 32.
    C. Bradac, T. Gaebel, N. Naidoo, M.J. Sellars, Twamley J., L.J. Brown, A.S. Barnard, T. Plakhotnik, A.V. Zvyagin, J.R. Rabeau: Nat. Nano 5, 345 (2010)Google Scholar
  33. 33.
    Y.-R. Chang, H.-Y. Lee, K. Chen, C.-C. Chang, D.-S. Tsai, C.-C. Fu, T.-S. Lim, Y.-K. Tzeng, C.-Y. Fang, C.-C. Han, H.-C. Chang, W. Fann: Nat. Nano 3, 284 (2008)CrossRefGoogle Scholar
  34. 34.
    V.K.A. Sreenivasan, E.A. Ivukina, W. Deng, T.A. Kelf, T.A. Zdobnova, S.V. Lukash, B.V. Veryugin, O.A. Stremovskiy, A.V. Zvyagin, S.M. Deyev: J. Mater. Chem. 21, 65 (2011)CrossRefGoogle Scholar
  35. 35.
    S.J. Yu, M.W. Kang, H.C. Chang, K.M. Chen, Y.C. Yu: J. Am. Chem. Soc. 127, 17604 (2005)CrossRefGoogle Scholar
  36. 36.
    A.M. Schrand, L. Dai, J.J. Schlager, S.M. Hussain, E. Osawa: Diam. Relat. Mater. 16, 2118 (2007)ADSCrossRefGoogle Scholar
  37. 37.
    A.M. Schrand, S.A. Ciftan Hens, O.A. Shenderova: Crit. Rev. Solid State Mater. Sci. 34, 18 (2009)CrossRefGoogle Scholar
  38. 38.
    D. Ho: Nanodiamonds: Applications in Biology and Nanoscale Medicine. Technology and Engineering (Springer, New York, 2009)Google Scholar
  39. 39.
    L.P. McGuinness, Y. Yan, A. Stacey, D.A. Simpson, L.T. Hall, D. Maclaurin, S. Prawer, P. Mulvaney, J. Wrachtrup, F. Caruso, R.E. Scholten, L.C.L. Hollenberg: Nat. Nano 6, 358 (2011)CrossRefGoogle Scholar
  40. 40.
    Y.-Y. Chen, H. Shu, Y. Kuo, Y.-K. Tzeng, H.-C. Chang: Diam. Relat. Mater. 20, 803 (2011)ADSCrossRefGoogle Scholar
  41. 41.
    A. Krger, F. Kataoka, M. Ozawa, T. Fujino, Y. Suzuki, A.E. Aleksenskii, A. Ya Vul, E. Osawa: Carbon 43, 1722 (2005)CrossRefGoogle Scholar
  42. 42.
    A.S. Barnard, M. Sternberg: J. Phys. Chem. B 109, 17107 (2005)CrossRefGoogle Scholar
  43. 43.
    B.R. Smith, D.W. Inglis, B. Sandnes, J.R. Rabeau, A.V. Zvyagin, D. Gruber, C.J. Noble, R. Vogel, E. Osawa, T. Plakhotnik: Small 5, 1649 (2009)CrossRefGoogle Scholar
  44. 44.
    C. Bradac, T. Gaebel, N. Naidoo, J.R. Rabeau, A.S. Barnard: Nano Lett. 9, 3555 (2009)ADSCrossRefGoogle Scholar
  45. 45.
    J.R. Rabeau, A. Stacey, A. Rabeau, S. Prawer, F. Jelezko, I. Mirza, J. Wrachtrup: Nano Lett. 7, 3433 (2007)ADSCrossRefGoogle Scholar
  46. 46.
    P.H. Chung, E. Perevedentseva, C.L. Cheng: Surf. Sci. 601, 3866 (2007)ADSCrossRefGoogle Scholar
  47. 47.
    K. Iakoubovskii, G.J. Adriaenssens: Philos. Mag. Lett. 80, 441 (2000)CrossRefGoogle Scholar
  48. 48.
    R.R. Nimmagadda, A. Joshi, W.L. Hsu: J. Mater. Res. 5, 2445 (1990)ADSCrossRefGoogle Scholar
  49. 49.
    Z. Du, A.F. Sarom, J.P. Longwell, C.A. Mims: Energy Fuels 5, 214 (1991)CrossRefGoogle Scholar
  50. 50.
    T. Ando, M. Ishii, M. Kamo, Y. Sato: J. Chem. Soc. Faraday Trans. 89, 1783 (1993)CrossRefGoogle Scholar
  51. 51.
    T. Gaebel, C. Bradac, J. Chen, J.M. Say, L. Brown, P. Hemmer, J.R. Rabeau: Diam. Relat. Mater. 21, 28 (2011)ADSCrossRefGoogle Scholar
  52. 52.
    C. Bradac, T. Gaebel, C.I. Pakes, J.M. Say, A.V. Zvyagin, J.R. Rabeau: Effect of the Nanodiamond Host on a Nitrogen-Vacancy Color-Centre Emission State. Small, 9:132–139 (2013)CrossRefGoogle Scholar
  53. 53.
    M.V. Hauf, B. Grotz, B. Naydenov, M. Dankerl, S. Pezzagna, J. Meijer, F. Jelezko, J. Wrachtrup, M. Stutzmann, F. Reinhard, J.A. Garrido: Phys. Rev. B 83, 081304 (2011)ADSCrossRefGoogle Scholar
  54. 54.
    C. Santori, P.E. Barclay, K.-M.C. Fu, R.G. Beausoleil: Phys. Rev. B 79, 125313 (2009)ADSCrossRefGoogle Scholar
  55. 55.
    M.T. Edmonds, C.I. Pakes, S. Mammadov, W. Zhang, A. Tadich, J. Ristein, L. Ley: Appl. Phys. Lett. 98, 102101 (2011)ADSCrossRefGoogle Scholar
  56. 56.
    K.M.C. Fu, C. Santori, P.E. Barclay, R.G. Beausoleil: Appl. Phys. Lett. 96, 121907 (2010)ADSCrossRefGoogle Scholar
  57. 57.
    M. Yu, F. Li, Z. Chen, H. Hu, C. Zhan, H. Yang, C. Huang: Anal. Chem. 81, 930 (2009)CrossRefGoogle Scholar
  58. 58.
    M.P. Hehlen, G. Frei, H.U. Gudel: Phys. Rev. B 50, 16264 (1994)ADSCrossRefGoogle Scholar
  59. 59.
    R.H. Page, K.I. Schaffers, P.A. Waide, J.B. Tassano, S.A. Payne, W.F. Krupke, W.K. Bischel: J. Opt. Soc. Am. B 15, 996 (1998)ADSCrossRefGoogle Scholar
  60. 60.
    J.C. Boyer, F. van Veggel: Absolute quantum yield measurements of colloidal NaYF4: Er3+, Yb3+ upconverting nanoparticles. Nanoscale, 2:1417–1419 (2010)ADSCrossRefGoogle Scholar
  61. 61.
    F. Wang, R. Deng, J. Wang, Q. Wang, Y. Han, H. Zhu, X. Chen, X. Liu: Nat. Mater. 10, 968 (2011)ADSCrossRefGoogle Scholar
  62. 62.
    H.-X. Mai, Y.-W. Zhang, R. Si, Z.-G. Yan, L.-d. Sun, L.-P. You, C.-H. Yan: J. Am. Chem. Soc. 128, 6426 (2006)Google Scholar
  63. 63.
    H.-X. Mai, Y.-W. Zhang, L.-D. Sun, C.-H. Yan: J. Phys. Chem. C 111, 13721 (2007)CrossRefGoogle Scholar
  64. 64.
    F. Wang, J.A. Wang, X.G. Liu: Angew. Chem. Int. Ed. 49, 7456 (2010)CrossRefGoogle Scholar
  65. 65.
    V.V. Tuchin: Tissue Optics: Light Scattering Methods and Instruments for Medical Diagnosis (SPIE, Bellingham, 2007)CrossRefGoogle Scholar
  66. 66.
    C. Vinegoni, D. Razansky, S.A. Hilderbrand, F.W. Shao, V. Ntziachristos, R. Weissleder: Opt. Lett. 34, 2566 (2009)ADSCrossRefGoogle Scholar
  67. 67.
    P.R. Selvin: Ann. Rev. Biophys. Biomol. Struct. 31, 275 (2002)CrossRefGoogle Scholar
  68. 68.
    Q. Liu, Y. Sun, T.S. Yang, W. Feng, C.G. Li, F.Y. Li: J. Am. Chem. Soc. 133, 17122Google Scholar
  69. 69.
    Q. Zhan, J. Qian, H. Liang, G. Somesfalean, D. Wang, S. He, Z. Zhang, S. Andersson-Engels: ACS Nano 5, 3744 (2011)CrossRefGoogle Scholar
  70. 70.
    A.D. Ostrowski, E.M. Chan, D.J. Gargas, E.M. Katz, G.Han, P. James Schuck, D.J. Milliron, B.E. Cohen: ACS Nano 6, 2686 (2012)CrossRefGoogle Scholar
  71. 71.
    F. Leblond, S.C. Davis, P.A. Valds, B.W. Pogue: J. Photochem. Photobiol. B 98, 77 (2010)CrossRefGoogle Scholar
  72. 72.
    F. Neugart, A. Zappe, F. Jelezko, C. Tietz, J.P. Boudou, A. Krueger, J. Wrachtrup: Nano Lett. 7, 3588 (2007)ADSCrossRefGoogle Scholar
  73. 73.
    S.M. Deyev, R. Waibel, E.N. Lebedenko, A.P. Schubiger, A. Pluckthun: Nat. Biotechnol. 21, 1486 (2003)CrossRefGoogle Scholar
  74. 74.
    D. Li, B.A. Dong, X. Bai, Y. Wang, H.W. Song: J. Phys. Chem. C 114, 8219 (2010)CrossRefGoogle Scholar
  75. 75.
    H. Zijlmans, J. Bonnet, J. Burton, K. Kardos, T. Vail, R.S. Niedbala, H.J. Tanke: Anal. Biochem. 267, 30 (1999)CrossRefGoogle Scholar
  76. 76.
    J.C. Boyer, M.P. Manseau, J.I. Murray, F. van Veggel: Langmuir 26, 1157 (2010)CrossRefGoogle Scholar
  77. 77.
    S.A. Osseni, S. Lechevallier, M. Verelst, C. Dujardin, J. Dexpert-Ghys, D. Neumeyer, M. Leclercq, H. Baaziz, D. Cussac, V. Santran, R. Mauricot: J. Mater. Chem. 21, 18365 (2011)CrossRefGoogle Scholar
  78. 78.
    R.S. Niedbala, H. Feindt, K. Kardos, T. Vail, J. Burton, B. Bielska, S. Li, D. Milunic, P. Bourdelle, R. Vallejo: Anal. Biochem. 293, 22 (2001)CrossRefGoogle Scholar
  79. 79.
    T. Pellegrino, L. Manna, S. Kudera, T. Liedl, D. Koktysh, A.L. Rogach, S. Keller, J. Radler, G. Natile, W.J. Parak: Nano Lett. 4, 703 (2004)ADSCrossRefGoogle Scholar
  80. 80.
    T.A. Kelf, V.K.A. Sreenivasan, J. Sun, E.J. Kim, E.M. Goldys, A.V. Zvyagin: Nanotechnology 21, 285105 (2010)CrossRefGoogle Scholar
  81. 81.
    M.Z. Strowski, A.D. Blake: Mol. Cell. Endocrinol. 286, 169 (2008)CrossRefGoogle Scholar
  82. 82.
    V.K.A. Sreenivasan, O.A. Stremovskiy, T.A. Kelf, M. Heblinski, A.K. Goodchild, M. Connor, S.M. Deyev, A.V. Zvyagin: Bioconjug. Chem. 22, 1768 (2011)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Andrei Vasilyevich Zvyagin
    • 1
  • Zhen Song
    • 2
  • Annemarie Nadort
    • 3
  • Varun Kumaraswamy Annayya Sreenivasan
    • 4
  • Sergey Mikhailovich Deyev
    • 5
  1. 1.MQ Biofocus Research Centre and MQ Photonics Research CentreMacquarie UniversitySydneyAustralia
  2. 2.MQ Biofocus Research Centre and MQ Photonics Research CentreMacquarie UniversitySydneyAustralia
  3. 3.Department of Biomedical Engineering and Physics, Academic Medical CenterUniversity of Amsterdam1100 DE AmsterdamThe Netherlands
  4. 4.MQ Biofocus Research Centre and MQ Photonics Research CentreMacquarie UniversitySydneyAustralia
  5. 5.Laboratory of Molecular ImmunologyShemyakin & Ovchinnikov Institute of Bioorganic Chemistry of the Russian Academy of SciencesMoscowRussia

Personalised recommendations