Golden Opportunities: Plasmonic Gold Nanostructures for Biomedical Applications based on the Second Near‐Infrared Window

Abstract

For biomedical applications, the NIR-II window provides several advantages over the conventional NIR-I window, including deeper penetration depth, low autofluorescence, and higher value of maximum permissible exposure to laser power. An overview of recently reported NIR-II-window-responsive plasmonic gold nanostructures is presented, and the opportunities, challenges, and future directions for these nanostructures in biomedical research fields are discussed. Colloidal plasmonic nanoparticles (PNPs) have attracted enormous attention because of their unique optical, chemical, electronic, and catalytic properties. They have been widely applied in various fields including plasmonics, sensing, catalysis, and biomedical imaging/diagnostics/therapeutics.1-14 Among the many PNPs, gold nanoparticles (AuNPs) offer a suitable platform, especially for biomedical applications, owing to their chemical/biological inertness and low cytotoxicity in a variety of cell and animal models. They also exhibit versatile and straightforward surface functionalization with a wide range of biological ligands such as oligonucleotides, proteins, and antibodies by forming stable bonding pairs such as Au–thiol bonds for specific/selective binding to biological targets or organs.15-18 Since AuNPs show different pharmacokinetic properties (absorption, distribution, metabolism, and excretion) according to their size, morphology, and surface moieties, they are highly promising materials for in vivo studies and customized practical purposes.19-21 One of the most prominent and useful features of AuNPs for biomedical applications is their unique and tunable optical properties, in particular, the localized surface plasmon resonance (LSPR) modulated by different sizes, shapes, and couplings of the nanoparticles. Upon irradiation with light of an appropriate wavelength, the incident light excites the free electrons in the AuNPs and simultaneously induces collective oscillation of the conduction-band electrons, which is known as LSPR. These surface plasmons reflect the underlying strong absorption and scattering of light, and these unique features make AuNPs fascinating candidates for biological sensing and imaging applications such as naked-eye-based in vitro target-detection assay22-24 or surface-enhanced Raman scattering (SERS)-based ultrasensitive imaging systems.25, 26 Furthermore, the plasmonic property of the AuNPs makes them potential therapeutic agents. Once surface plasmons are generated, they transfer their absorbed energy to the metallic lattice and the surrounding medium in the form of thermal energy during the plasmon relaxation process, which is called the photothermal effect.27 During the treatment of cancer, specifically accumulated AuNPs near the cancerous cell generate localized heating under external light irradiation, resulting in cell death.28, 29 For the practical use of the external-light-mediated plasmonic properties of AuNPs, they should be highly responsive to external light with an appropriate wavelength. However, visible-light-responsive AuNPs have significant limitations for in vivo applications because the visible light is almost fully absorbed or scattered by biological tissues such as blood, water, melanin, and fat (Figure 1a).30 As a result, external light hardly penetrates or propagates to the deep tissue region, resulting in insufficient energy transfer to the deeply localized AuNPs in the tissue. To circumvent this problem, researchers have explored a longer wavelength light source in the wavelength range of 750 to 1400 nm, known as the near-infrared (NIR) window, where light has its maximum depth of penetration in tissues (Figure 1a).31 Especially, in comparison with the traditional first NIR window (NIR-I) (750–900 nm), the imaging resolution (signal-to-noise ratio) and the penetration depth of external light are greatly improved in the second NIR window (NIR-II) (1000–1400 nm) due to the reductions in the intensity of autofluorescence and the absorption/scattering of photons.32 Thus, only ultralow doses of external light intensity, which are far below the skin-tolerance threshold set by the American National Standards Institute (ANSI), are required for making the NIR-II window potentially suitable for clinically practical applications.33 Furthermore, the NIR-II window has a higher value of maximum permissible exposure (MPE) to laser light compared to NIR-I (the MPEs of skin to laser are 1 and 0.33 W cm−2 for the NIR-II and NIR-I windows, respectively). Thus the NIR-II window is a relatively safe spectral region for practical purposes.34 For this reason, researchers have developed small-molecule fluorescent dyes,32, 35-37 quantum dots,38-43 carbon nanotubes,44-50 etc. as imaging agents within the NIR-II region. However, the NIR-II-window-related properties of plasmonic AuNPs for biomedical applications have rarely been studied. Interestingly, the SPR wavelength of AuNPs can be diversely tuned over a wide range of wavelengths from the visible to the NIR region depending on the size and the shape of the AuNPs.51, 52 In this regard, appropriate design (in terms of size, shape, and plasmonic coupling) of the plasmonic AuNPs, which are highly responsive over a wide range of the NIR-II window, is a key requirement for biomedical applications utilizing plasmonic properties. Several studies based on plasmonic Au nanostructures, which have their LSPR peak in the NIR-II region, have been carried out. Specially designed nanostructures, such as Au nanorings, Au double nanopillars with SiO2 spacers, nanocross cavities, and Au nanocrescent antennas, exhibit optical signals in the NIR-II region.53-57 Au nanorings are prepared by the evaporation of Au films on polystyrene colloidal particles on a substrate.53 Free-standing Au nanorings are obtained after the removal of the polystyrene particles, and the optical property of this nanostructure depends on the structural parameter, such as the diameter. Au nanocrescents represent another nanostructure that can be tailored to have different widths and lengths to exhibit short-axis dipole resonance positioned around 1300–1500 nm.55 However, all these nanostructures are fabricated on a substrate, preventing or limiting their practical feasibility for cell and in vivo applications. As a nanoprobe, the Au nanorod is one of the most widely used nanostructures for tuning the LSPR into the visible or the NIR regions. While the transverse LSPR mode appears at a fixed wavelength of around 520 nm, the longitudinal LSPR mode appears at longer wavelengths and highly depends on the aspect ratio of the Au nanorods. Therefore, the LSPR wavelength can be easily adjusted by controlling the length of the Au nanorods, for example, by using different amounts of additives such as silver nitrate and hydrochloric acid during synthesis.58 It has been reported that the longitudinal LSPR of Au nanorods is positioned around 1100 nm at an aspect ratio of 8. There are methods to make Au nanorod derivatives. Liz-Marzan and his group investigated the influence of iodide ions during the growth of Au nanorods, and reported that the longitudinal LSPR band of the as-grown nanorods can exceed 1000 nm (Figure 1b).59 Similarly, Song et al. demonstrated that programmable DNA sequences can control the overgrowth of Au nanorods, resulting in the formation of nanodumbbells with their LSPR at 1011 nm (Figure 1c).60 An Au shell deposited on a silica core also generates an LSPR peak in the NIR-II region.61 The optical signals of these nanoparticles can be increased above 1000 nm by altering the relative dimensions of the core and the shell. Recently, there have been some efforts to utilize Au nanomaterials for in vivo bioapplications in the second biological window region. In 2010, Xia and co-workers fabricated Au/Ag alloyed nanocages and applied an NIR-II laser for three-photon-luminescence (3PL) tissue imaging.62 The Au/Ag alloyed nanocages were excited using a femtosecond laser at 1290 nm, and 3PL was observed in the visible region, which was one order of magnitude stronger than that from pure Au or Ag nanoparticles. By using a low power (≈4.0 mW) NIR-II laser, of which the wavelength was beyond the plasmon resonance peak of the nanocages, very little autofluorescence background and undetectable photothermal toxicity were obtained. Apart from the above, few studies on the photoinduced therapeutic utilization of second-biological-window-active Au nanomaterials have been reported. Yeh and co-workers developed rod-in-shell nanostructures smaller than 100 nm and finely tuned their plasmonic properties by tailoring the gap distance between the AuNR core and the AuAg nanoshell, in order to evaluate their therapeutic effects in the NIR-I and the NIR-II regions (Figure 1d).63 They presented the first demonstration of in vitro and in vivo photothermal therapy in the NIR-II window region. However, the applied laser power of 3 W cm−2 is far higher than the skin tolerance threshold value of 420 mW cm−2, which limits their use for the clinical treatment of tumors. Recently, two studies using cell-endurable values of laser power have been published. Hwang and co-workers used a unique Au nanoechinus (AuNEs) structure with extraordinarily high extinction coefficients of ≈1012 M−1 cm−1 in the NIR-II region for dual-modal photodynamic and photothermal therapy in the second-biological-window region.64 They used laser powers of ≈130 mW cm−2, which is ≈2–3 times lower than the standards set by ANSI for skin burning. However, the size of the AuNEs was much larger than 100 nm, making them unsuitable for in vivo applications. Jiang and co-workers fabricated a plasmonic hybrid nanostructure of Au–Cu9S5 with well-controlled interfaces (Figure 1e), and it exhibited an increase of 50% in the absorbance at 1064 nm compared with that of the pure Cu9S5 nanoparticles.34 The overall lateral dimension of the nanostructure was about 20 nm, and they used a laser power below 0.6 W cm−2 at 1064 nm to induce photothermal therapeutic activity. Furthermore, they confirmed the computed-tomography (CT) imaging potential of Au–Cu9S5 nanostructures with excellent X-ray attenuation ability. Even though this nanostructure exhibited potential for multimodal, NIR-II-window-active, therapeutic and deep-tissue imaging applications, the value of the used laser power is higher than the skin-tolerance threshold, and the exposed Cu9S5 is recognised as a more toxic material than Au. It should be noted that the equipment required for the NIR-II window, such as the laser and the detector, is more expensive than that for the NIR-I window, which limits the practical utilisation of the NIR-II window. The development of a more-cost-friendly NIR-II laser system is an important challenge for practical applications. Besides, the plasmonic nanoparticles need to be ensured for biocompatibility and cytotoxicity if they are to be widely applied for in vivo applications. Up to now, various types of plasmonic nanoparticles that are highly responsive over a wide range of the NIR-II window have been developed by altering their structure and metal compositions.34, 62-64 However, almost all NIR-II-window-responsive nanomaterials are bimetallic, which means that they contain not only gold, which is chemically/biologically inert and biocompatible with a low cytotoxicity, but also Ag or Cu, which might induce toxicity during in vivo applications. In spite of the potential usability of plasmonic Au nanomaterials for in vivo applications, NIR-II-window-responsive monometallic AuNPs have rarely been studied to date, in terms of synthesis, optical properties, and biomedical suitability. In this regard, the facile and straightforward synthesis, the finely tunable and highly enhanced optical properties over a wide range of the NIR-II window, and the higher degree of biocompatibility including pharmacokinetic studies of the AuNPs are essential challenges for the practical use of these plasmonic nanoparticles with the NIR-II window in the area of biomedicine. NIR-II-window-responsive plasmonic Au nanomaterials have been mainly used for photoinduced therapeutic applications with laser power ranging from ≈130 mW cm−2 to 3 W cm−2.34, 63, 64 Even though the NIR-II region has a relatively higher value of MPE to laser power, the development of photoinduced therapeutic strategies that can operate at a much lower laser-power range (<100 mW cm−2) will boost its potential clinical usage by preventing latent skin damage during the therapeutic treatment. In order to reduce the working range of the laser power, appropriate and in-depth design and analysis of the photothermal transduction efficiency at the NIR-II-window region are required with respect to conclusive parameters such as size, shape, absorption and scattering tendency, shell coating, and assembly condition.65-67 Before designing an NIR-II-window-active photoinduced therapeutic platform, the photothermal transduction efficiencies and the photothermal transduction cross sections of the targeted plasmonic Au nanomaterials need to be theoretically predicted, and those values must be verified experimentally after preparing the NIR-II-window-active nanomaterials. Ultimately, the working range of the laser power must be scanned thoroughly during the designated time range to induce a realistic therapeutic effect. Furthermore, highly delicate temperature control is an important task and challenge for real applications of photoinduced therapy at clinical level. Under local laser irradiation, healthy cells adjacent to the lesions can be damaged due to significant heat transfer. In order to overcome this side effect, precise temperature control interworked with real-time thermal monitoring of subcutaneous cells and use of highly tunable nanostructures must be accomplished. Even though some researchers have attempted to solve this issue in the NIR-I-window region,68-71 to date, there has been no noticeable advancement in this aspect in the NIR-II-window region. Regarding the functionality as a nanoprobe, there are a few studies that demonstrate multimodality in the NIR-II region, including the examples described in the section above. Multimodality can be explored in the NIR-II region with various types of imaging techniques such as SERS, metal-enhanced fluorescence (MEF), magnetic resonance imaging (MRI), CT, photoacoustic tomography (PAT) and positron emission tomography (PET), and therapy systems including photodynamic therapy (PDT), photothermal therapy (PTT), and chemotherapy similar to that in the NIR-I region (Figure 2). The use of dual imaging or therapy can compensate the drawbacks of a single technique, improving the imaging sensitivity or resolution or allowing better therapeutic performance with mild laser exposure conditions.71-73 In addition, theranostic nanoprobes can be developed for simultaneous imaging and therapy. The techniques such as SERS, MEF, and PTT can be demonstrated by positioning the LSPR peak of the plasmonic nanomaterials in the NIR-II region. Other drawbacks can be overcome using hybrid systems made by designing core–shell nanostructures or modifying the surface of the nanostructure with corresponding molecules. Apart from bimodal nanoprobes for imaging with MEF combined with PDT,74 there are several studies in pursuit of three or even more modalities. Some of the representative examples are SERS–MR–PA trimodal imaging with PTT, CT–PAT dual imaging with PTT–chemotherapy dual therapy, and upconversion luminescence (UCL)–CT–PA–MRI with PTT.75-77 Despite the pronounced advantages for in vivo biomedical applications, plasmonic Au nanomaterials utilizing the NIR-II-window region have not been well studied because it is difficult to design and synthesize these materials with a high response in the NIR-II-window region while considering all the practical and clinical aspects of these materials. Although noticeable progress in the area of NIR-II-responsive plasmonic nanomaterials has been reported recently, the achievements in this field do not match up with those realized in the NIR-I-window region. For long-term safety and better controllability, NIR-II-sensitive pure Au nanostructures with appropriate size and shape need to be developed without any additional residual metal, and the efficiency of conversion of the photoenergy to other energies must be improved in order to use lower laser-power densities for photoinduced therapeutic clinical applications. Furthermore, high-level multimodality has not successfully materialized for simultaneous in vivo imaging and therapy due to several issues, including the toxicity of materials, tissue-penetration depth, and skin tolerance. Addressing these issues and the aforementioned other challenges will open up new prospects in the field of in vivo bioapplications of plasmonic nanomaterials including AuNPs. J.-E.P., M.K., and J.-H.H. contributed equally to this work. J.-M.N. was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science, ICT & Future Planning (MSIP)) (No. 2016R1A2A1A05005430) and BioNano Health-Guard Research Center funded by the MSIP of Korea as a Global Frontier Project (H-GUARD_2013M3A6B2078947). This research was also supported by the Pioneer Research Center Program through the National Research Foundation of Korea funded by the MSIP (NRF-2012-0009586).