Raman Spectroscopy in Nanomedicine

Surface-Enhanced Raman Spectroscopy

The phenomenon of SERS was described as early as 1974,^[49,50] and is understood to arise from a localized increase in the coupling between the electromagnetic field of the incident radiation and the polarization of the analyte in the presence of optically induced surface plasmons on a metal surface. Increases in Raman intensities as high as 10¹⁰ have been reported,^[51] although the spatial range of enhancement is only in the order of tens of nanometers. The enhancement process can be achieved using a number of substrates, including roughened metallic surfaces, structured metal arrays and specially imprinted surfaces.

Notably, the SERS effect can be induced through the use of metallic nanoparticles and nanocolloid aggregates. SERS is a direct enhancement of the Raman signal and, in the case of nanoparticles, this occurs in the immediately surrounding local vicinity. The true principle that governs SERS enhancement is not fully understood, although the effect has largely been attributed to an electronic enhancement caused by local fields generated by surface plasmon resonances at the metal surface. Alternatively, the enhancement has been attributed to a charge transfer process between the analyte and the surface, although it is probable that the processes act in tandem.^[52] The technique of SERS in a biomedical context is reviewed in greater detail elsewhere.^[53–55]

Nanoparticles and aggregates that are used for SERS enhancement typically consist of a metallic nanoparticle, most commonly gold or silver. Frequently, these particles are subsequently modified via surface functionalization, which can include targeting moieties designed for specific applications, especially those used as nanosensors. The particle may also be labeled with a Raman reporter moiety, which allows for identification of the particle in the biological milieu. Using these particles, it has been possible to apply SERS to a number of biological scenarios, including diagnostic studies in vitro, ex vivo^[56,57] and in vivo,^[58,59] novel bioassays^[60–62] and cellular studies.

SERS has been proposed as a method for understanding how nanomaterials behave in a cellular environment, which is important in the study of the fundamental interactions of nanoparticles in the context of toxicology, drug delivery or contrast agents for diagnostics. In 2002, Kneipp et al. proposed that, by using SERS, it would be possible to probe the chemical nature of the subcellular environment and the intracellular distribution of biomolecules.^[63] This work was extended by incorporation of Raman reporters, which allowed for localization of the SERS probe within the cell, leading to chemical probing of subcellular nanostructures.^[64–67] For example, in 2010, the group showed how a SERS nanosensor was capable of investigating pH changes in a cell during the different stages of the endocytic pathway of the nanoparticle probe. The study was based on changes in the pH of the local environment in different cellular organelles, which can be monitored via changes in the pH sensitive nanoprobe over time.^[68]

Other cellular studies have also investigated the possible use of SERS in the investigation of cell surface receptors associated with cancer. In one such study, Kong et al. used organometallic SERS-active nanoparticles, which were targeted to live cells expressing EGFR. The SERS nanoparticles were shown to be capable of specific targeting to the cell surface and offered increased sensitivity compared with other imaging modalities.^[69] Figure 1A–E shows oral squamous cell carcinoma cells expressing EGFR, Figure 1C & E shows the SERS image generated by CO at 2030 cm⁻¹ and protein at 1600 cm⁻¹, respectively. The targeting was verified in a non-EGFR-expressing cell line SKOV3 (ovarian carcinoma; Figure 1F–J) and by blockage of the EGFR using an EGFR antibody (Figure 1K–O).

(Enlarge Image)

Figure 1.

Bright-field, dark-field and surface-enhanced Raman spectroscopy imaging of carcinoma cells.
(A–E) Oral squamous cell carcinoma cells, (F–J) SKOV3 cells not expressing EGFR and (K–O) oral squamous cell carcinoma cells treated with anti-EGFR. (A, F & K) The bright-field and (B, G & L) dark-field images of the nanoparticles; (C, H & M) the SERS image of CO at 2030 cm⁻¹; (D, I & N) merged SERS and brightfield; and (E, J & O) the SERS image generated using the protein band at 1600 cm⁻¹.
SERS: Surface-enhanced Raman spectroscopy.
Reproduced with permission from [69].

Another study demonstrated the use of SERS in the analysis of human serum. Lin et al. demonstrated the power of SERS coupled with multivariate analysis to distinguish between patients previously diagnosed with colorectal cancer and control patients in a noninvasive way with 100% diagnostic sensitivity and specificity.^[70]

In vivo SERS is also possible and has been demonstrated as a potential labeling method for a number of applications. SERS has been used in vivo to investigate how enhancement of the Raman signal can be used as a method for tumor detection. Qian et al. showed how EGFR targeting PEGylated gold nanoparticles labeled with a Raman reporter were capable of >200-times greater signal generation in the IR compared with that of near-IR fluorescent quantum dots, which allowed for the possible identification of small tumors at penetration depths of approximately 1–2 cm.^[58] Other in vivo applications of SERS have also been explored, including an in vivo study of inflammation in mice,^[71] demonstrating improvements over fluorescence-based methods. SERS has also been shown to be capable of single molecule detection in vitro, a sensitivity that sets it apart from spontaneous Raman spectroscopy.^[72]

More complex Raman-based investigations have also taken advantage of the surface enhancement process. Techniques such as deep-penetrating spatially offset Raman (SORS) have been combined with nanoparticle-based SERS in surface-enhanced SORS (SESORS).^[73,74] In brief, in the SORS technique, introduced in a paper by Matousek et al., the Raman spectra are collected at positions spatially offset from the point of incidence of the probe laser beam.^[75] Rather than using microscopic objectives for delivery and collection, fiber probes are used. By moving the collection point away from the probe launch site, contributions from the surface Raman photons are diminished and those of Raman photons from deeper within the sample are increased. Using multivariate statistical methods, it is possible to reconstruct spectra from the different layers with a much greater depth of penetration than a traditional confocal microscopy setup.^[75] Depth sensitivities of up to several millimeters are now achievable and examples of emerging applications include noninvasive diagnosis of bone disease, cancer and monitoring of glucose levels.^[76] SESORS uses this same principle, taking advantage of the surface enhancement of the Raman signal from metallic nanoparticles embedded within the sample. In a recent publication by Xie et al., SESORS was used to identify bisphosphonate-functionalized nanotags on bone through 20 mm of porcine tissue.^[77] This study highlights the increasing potential for in vivo applications that SORS and SESORS may have in the field of nanomedicine.

SERS has enjoyed increasing popularity over the past decade, particularly since the emergence of an increasing range of nanoprobes. However, the uptake rates and mechanisms, as well as the subsequent trafficking, may be specific to the nanoparticle type, size and surface chemistry. Most SERS probes are specifically designed for a targeted application and are, therefore, labels themselves for the SERS signal. Furthermore, the molecular specificity of the surface enhancement process is not well understood. Therefore, a truly label-free method for generic monitoring and characterization of the cellular uptake and subcellular localization of nanoparticles in general is still required.

TERS is another method for generating enhancement of the Raman signal. Like nanoparticle-based SERS, this method is also based on probing the inherent nanoscale environment of the sample in close proximity to a nanoprobe and, therefore, will be discussed.