1. Introduction

Terahertz (THz) technology has been advancing rapidly because of its tremendous potential applications [1–9]. Among them, THz cancer diagnosis is drawing much attention as THz waves can detect the variation of cells caused by cancer, thereby rendering a new modality of medical imaging [10–13]. The sensitivity of THz electromagnetic (EM) waves to water molecules allows the utilization of the THz technique in diagnosing cancers because in cancerous tumors, diseased tissues contain more interstitial water than healthy tissues. The higher water content, combined with structural changes such as increased cell and protein density, leads to a larger THz absorption and refractive index for tissues with tumors [12 and 13]. This system requires further development, however, so that it can be actually adopted in clinical medicine. The problem with THz cancer diagnosis is the difficulty in identifying the tumor in tissues unless the cancer is already well developed. The antibody-conjugated contrast agents for THz EM waves, similar to the technique adopted in MRI, may solve the problem [14 and 15]. This study thus examines the interactions between THz EM waves and metal nanoparticles of gold nano-rods (GNRs) in cancer cells for use in the nanoparticle-contrast-agent-enabled terahertz imaging (CATHI) technique, which realizes a high-sensitivity THz imaging for cancer diagnosis by targeting the nanoparticles to the tumors.

The interaction between nanoparticles and THz waves is expected to be small as the size of nanoparticles is from three to four orders of magnitude smaller than the THz wavelength. Nevertheless, the cell with nanoparticles can be heated up with surface plasma polaritons by illuminating it with an infrared (IR) laser beam [15 and 16]. The THz absorption and refractive indices are sensitive to the temperature change in water or in cells that contain large amounts of water [17], and this property enables the modulation of the THz signal with an IR laser irradiation. The employment of IR laser beam for imaging and signal detection also realizes a practical THz endoscopy for diagnosing cancers of internal organs, such as the large and small intestines, which cannot be diagnosed in vivo with the MRI technique [8].

2. Experiments

Various GNRs were fabricated with a surface plasmon resonance of around 800 nm of wavelength in water, as shown Fig. 1. The GNRs were synthesized using a seed-mediated growth technique according to a previously published protocol, with some modifications [18]. Their morphology was confirmed with a transmittance electron microscope (TEM) image, as shown in Figs. 1(a), (b), and (c), and their aspect ratios were 3.2, 4.0, and 4.2, respectively. The optical properties of these GNRs in water were measured with a UV-visible spectrometer. Figure 1(d) shows that the wavelength of peak absorbance shifted from 750 to 850 nm as the aspect ratio increased. The absorbance of the GNRs arose from the surface plasmon resonance effect through the electron oscillation along the longitudinal and transverse axes of the GNRs and the boundary between the water and the GNRs. This resonance induced the increase in the temperature of the water around the GNRs, and the temperature variation affected the reflectivity of the THz waves from the water because the refractive index and the power absorption increased as the temperature rose [13].

To demonstrate the principle behind the CATHI technique in vitro, epidermoid carcinoma A431 cell lines were cultured and the cells (1 × 10⁶ cells/ml) were seeded in a 96-well plate. The A431 cells were incubated for 1 hour with synthesized GNRs with a concentration of 60 μg/ml, after which the treated cells were washed three times.

 

Fig. 1. TEM images of gold nano-rods (GNRs) with aspect ratios of (a) 3.2, (b) 4.0, and (c) 4.2; (d) UV-visible absorption spectra of the GNRs of (a), (b), and (c).

Download Full Size | PDF

THz responses from the samples were measured using a reflection-mode THz time–domain spectroscopic (THz–TDS) system, as shown in Fig. 2, which can also be used as a medical imaging setup. A mode-locked Ti:sapphire laser, which provided 80-fs pulses at a wavelength of 800 nm, was divided into two beams. One of the beams was illuminated on a p-InAs wafer to generate THz pulses, and the other was focused on a photoconductive antenna fabricated on a low-temperature-grown GaAs to detect the THz pulses. The THz waveforms were measured by sampling the cross-correlated signal of the optical gating pulse and the THz pulse at the detector. The generated THz pulses were focused on a sample with two parabolic mirrors, and the reflected THz pulses from the sample were guided to the detector with a high-resistivity silicon beam splitter and a third parabolic mirror. The THz beam propagation area was sealed with an airtight box and purged with dry air to eliminate the absorption caused by water vapor. The IR laser, which induced the surface plasmon resonance, was a continuous wave (CW) Ti:sapphire laser that provided a beam with a center wavelength of 800 nm and was irradiated on the same surface of the sample with THz pulses. The sample was located at the focus of the THz pulses and the image was obtained by two-dimensional scanning of the sample.

 

Fig. 2. Schematic of a reflection-mode THz imaging setup with an infrared (IR) laser for the induction of surface plasma polaritons.

Download Full Size | PDF

3. Results

The THz reflection signals from the two samples of water with and without GNRs were almost identical without CW IR laser irradiation, which implied that there was no interaction between THz waves and GNRs. The amplitude of the reflected THz time–domain waveform increased, however, as shown in Fig. 3, when the IR laser was illuminated on the sample with GNRs, although the waveforms were almost identical with and without irradiation with the IR laser. The concentration of the GNRs in the water was 3.2 mg/ml, the diameters of the terahertz and the IR beams were 5 mm and 2 mm, respectively, and the IR intensity was 10 W/cm².

 

Fig. 3. Reflected THz time-domain waveforms from the water with GNRs before (black line) and 5 minutes after their irradiation with an IR laser (red line).

Download Full Size | PDF

Figure 4 shows that the peak reflection amplitude increased upon the irradiation of the IR laser beam on the water or the cancer cells with GNRs, whereas there was almost no change in the water or the cancer cells without GNRs even under IR irradiation. This was due to the higher refractive index and absorption of water at a higher temperature. The water temperature rose due to the hyperthermia effect induced by the surface plasmon resonance. The IR intensity and GNR density dependencies of the reflected THz amplitude from the water with GNRs are shown in Figs. 4(a) and (b), respectively. Figure 4(a) shows that the peak reflection amplitudes increased as the IR laser intensity increased, the experimental condition was the same as that for the acquisition of the data shown in Fig. 3, except for the IR intensity variation. The reflection amplitude change was more than 20% from the water with GNRs after 120 seconds of IR laser irradiation. Furthermore, the reflection change occurred instantly within 2 seconds as soon as the IR laser was illuminated, and then gradually increased. This rapid rise makes it possible to construct images pixel by pixel using the raster scan technique. Figure 4(b) shows that the higher GNR concentration resulted in a higher reflection change and, even with a low concentration of 0.1 mg/ml, a clear reflection change was observed under an IR intensity of 10 W/cm².

The THz response from epidermoid carcinoma A431 cells with and without GNRs upon IR laser irradiation was also measured. The concentration of GNRs in the A431 cells was 60 μg/ml, and the diameters of the terahertz and IR beams were 5 and 2 mm, respectively. As shown in Fig. 4(c), the peak reflection amplitude increased under IR laser irradiation, although there was almost no change in the A431 cell without GNRs. The reflection change in the A431 cell with GNRs at an IR intensity of 20 W/cm² after 120 seconds was 20%. An abrupt change in the THz reflection amplitude was also observed in the cancer cells bound with GNRs. Figure 4(d) shows the enlarged data from Fig. 4(c) in time. It clearly shows that the change occurred in 2 seconds regardless of the IR intensity, similar to the case of the water samples. The IR laser did not injure the cells without GNRs, although the energy of around 800 nm was larger than the amount of molar energy required for adenosine triphosphate (ATP) hydrolysis [11, 19], because the absorption coefficient of the tissue in the NIR region was below 0.1 cm^-1 [20]. Therefore, the IR illumination hardly affected the tissue. These results offer the foundation for sensitive THz diagnosis of cancerous tumors targeted with nanoparticle contrast agents.

 

Fig. 4. (a) IR laser intensity dependence and (b) GNR density dependence of the reflected THz signals from the water with GNRs. (c) Peak reflection changes of the THz signals from live cancer cells with and without GNRs at IR intensities of 10 and 20 W/cm² and (d) their enlarged data in time. Arrows indicate the onset of NIR excitation.

Download Full Size | PDF

Images were acquired to demonstrate the principle behind the CATHI technique. The concentration of the GNRs was 60 μg/ml, and the diameters of the terahertz and IR beams were 5 mm and 2 mm, respectively, similar to the conditions for the data shown in Figs. 4(c) and (d). The scan step and the time delay per pixel were 500 μm and 500 ms, respectively. Figure 5 shows the two-dimensional visible and THz images of the A431 cells with and without GNRs. The two wells were prepared as shown in Fig. 5(a). One was filled with the A431 cells with GNRs, and the other, without GNRs. The THz reflection images from the two samples with and without GNRs were almost identical, as shown in Fig. 5(b). The image of the cell with GNRs, however, became brighter upon IR laser beam irradiation than the THz–only image for the same sample, whereas there was almost no change in the cells without GNRs, as shown in Fig. 5(c). Figure 5(d) shows the reflection amplitudes along the lines as shown in Figs. 5(b) and (c), wherein the black line in Fig. 5(d) is for Fig. 5(b) and the red line is for Fig. 5(c). The enhancement was approximately 10% under the IR irradiation. A surprising result emerged when the difference between the images in Figs. 5(b) and (c) was determined, as shown in Fig. 5(e). The differential image without GNRs was almost indecipherable, but that with GNRs showed a clear image. The amplitude along the line in Fig. 5(e) is shown in Fig. 5(f), and the ratio of the amplitude with GNRs to that without GNRs was approximately 30. Therefore, cancer cells with GNRs can be distinguished by the IR laser irradiation with a very high contrast. These results show that highly sensitive THz imaging for cancer diagnosis can be accomplished with nanoparticle contrast agents. In addition, cancerous tumors can be identified by measuring a signal at a point without imaging an area, as the differential signal comes out only from where nanoparticles are targeted.

 

Fig. 5. Cancer cell images with and without GNRs. (a) Visible image; (b) THz image without IR irradiation; (c) THz image with IR irradiation; (d) amplitudes along the lines in (b) (black) and (c) (red); (e) differential image between (b) and (c); and (e) amplitude along the line in (e).

Download Full Size | PDF

4. Conclusion

In conclusion, this study demonstrated the principle behind the nanoparticle-contrast-agent-enabled terahertz imaging (CATHI) technique. With the CATHI technique, significant sensitivity enhancement is achieved in THz cancer imaging with nanoparticle contrast agents that can be targeted to cancerous tumors.

The THz reflection amplitude from the cancer cells with gold nano-rods (GNRs) increased by 20% upon IR laser irradiation compared to cancer cells without GNRs. In the differential mode, the difference between the two cases was more evident because the THz signal from the cancer cells with GNRs was 30 times higher than that from cancer cells without GNRs. This means that the CATHI technique can be utilized in duplex modes for cancer diagnosis either via imaging or merely by monitoring the signal at a point.

As the THz modulation occurs by irradiation of an IR laser beam and the differential technique offers extremely high sensitivity, THz cancer imaging can be realized with a micron resolution, which would facilitate the diagnosis and study of cancers at a very early stage.