Ultrabright contrast agents with synergistic Raman enhancements for precise intraoperative imaging and photothermal ablation of orthotopic tumor models | Journal of Nanobiotechnology

janeiro 20, 2025

2

Synthesis and optical properties of the au NR-based SERRS contrast agents

The protocols for synthesizing the SERRS agents were developed by integrating several established methods. Gold nanorods (Au NRs) were synthesized using the classical seed-mediated growth method with cetyltrimethylammonium bromide (CTAB) as the surfactant [34,35,36]. Since the localized surface plasmon resonance (LSPR) enhancement for SERS depends on the aspect ratio of the Au NRs, we synthesized two types of Au NRs with different aspect ratios, designated as Au NR 800 and Au NR 600. These aspect ratios were selected to match the excitation wavelengths of lasers used in Raman spectroscopy.

To address the cytotoxicity of CTAB, it was replaced with heparin (Hep) through a modified ligand exchange method [37, 38]. This exchange was confirmed by the disappearance of the characteristic Raman bands of CTAB in the Raman spectrum of the Au NR-Hep (AH) solutions (Fig. S1). Finally, the PPy-PDA hybrid was coated onto the Au NR-Hep nanoparticles using a polysaccharide-assisted copolymerization of pyrrole and dopamine monomers. In this study, heparin was employed as a stabilizing agent, not only for its compatibility but also for its reported antitumor effects [26].

The morphological properties of the as-synthesized SERRS contrast agents were characterized using transmission electron microscopy (TEM) and dynamic light scattering (DLS). TEM images revealed that the Au NRs were clearly visible, while the PPy-PDA hybrid formed a faint contrast around the surface of the Au NRs (Fig. 1a). Elemental mapping via energy-dispersive X-ray spectroscopy (EDX) confirmed the presence of nitrogen, indicating the successful coating of the PPy-PDA hybrid on the Au NR surfaces (Fig. 1b-c). Additionally, sulfur signals detected in the EDX analysis indicated residual heparin on the surface of the SERRS contrast agents, correlating with their observed negative zeta potential in phosphate-buffered saline (PBS) (Fig. S2).

Dynamic light scattering analysis showed a bimodal hydrodynamic size distribution for Au NR800-Hep@PPy-PDA, with transversal dimensions of 30.4 ± 5.2 nm and longitudinal dimensions of 255.0 ± 33.1. In contrast, Au NR600-Hep@PPy-PDA exhibited a unimodal size distribution. This difference may result from aggregation during the preparation process, leading to overlapping transversal and longitudinal sizes in Au NR600-Hep@PPy-PDA. The mean hydrodynamic sizes derived from the unimodal distribution were approximately 190.1 ± 42.2 nm (Fig. 1d).

The optical properties of the as-synthesized SERRS contrast agents, including their absorption and Raman spectra, were systematically investigated. The absorption spectra of Au NRs showed two distinct peaks: a transversal LSPR peak at approximately 500 nm and a longitudinal LSPR peak that can be tuned from the visible to the mid-infrared region by increasing the aspect ratio of the Au NRs [34]. This tunability of LSPR is a key advantage of Au NRs over other gold nanostructures.

The absorption spectra of Au NR-Hep and Au NR-Hep@PPy-PDA nanoparticles revealed only minor differences (Fig. 1e). The characteristic absorbance of the PPy-PDA hybrid around 700 nm was negligible, likely due to the low content of the PPy-PDA hybrid in the as-synthesized SERRS contrast agents [26]. Consequently, the optical properties of the Au NRs dominate the imaging performance of the Au NR-Hep@PPy-PDA nanoparticles.

The Raman properties of the as-synthesized contrast agents were further evaluated, focusing on their enhancement mechanisms and factors influencing Raman signal amplification. The Raman signal enhancement primarily depends on the properties of the SERS substrate and the interactions between the SERS substrate and the surface reporter, the PPy-PDA hybrid. Together, these components are the key contributors to Raman enhancement. To quantify the Raman enhancement, the peak area of the C = C stretching band around 1600 cm^− 1 was selected as the reference [39]. The SERS effect provided by the Au NRs was assessed using the PPy-PDA hybrid as a control group. By comparing the pristine PPy-PDA hybrid with the PPy-PDA in the SERRS contrast agents at equivalent mass concentrations, it was observed that the Raman intensity of Au NR-Hep@PPy-PDA nanoparticles was approximately 10³-10⁴ times higher than that of the PPy-PDA hybrid alone (Fig. 1f). The RRS contribution from the PPy-PDA hybrid was estimated using Au NR-Hep@PPy nanoparticles as a control group. In our previous research, the RRS enhancement in the PPy-PDA hybrid was about 3 times greater than that of PPy alone. However, the actual RRS enhancement observed in the SERRS contrast agents ranged from 3 to 6 times. Raman signals obtained at 638 nm (Fig. S3a) increased approximately 4.4 and 3.1 times in the Au NR600 SERRS and Au NR800 SERRS contrast agents, respectively. Similarly, Raman signals at 785 nm (Fig. S3b) increased by 5.0 and 6.3 times for the Au NR600 SERRS and Au NR800 SERRS agents, respectively. These results confirm that the integration of Au NRs and the PPy-PDA hybrid generates additional Raman scattering enhancement, likely due to a charge transfer process between the two components. This finding validates the effectiveness of our strategy for synergistically amplifying Raman signals.

Remarkably, even at a concentration as low as 1 ppm, the Raman signals from the SERRS contrast agents were still clearly detectable (Fig. 2a). The sensitivity of our SERRS contrast agents was in the magnitude of 10^− 14-10^− 13 M, which is comparable to other recently reported Raman contrast agents [5, 14]. While a much lower laser intensity and reduced acquisition time was used in our research, in other words, our Raman contrast agents demonstrate improved sensitivity compared with existing ones.

Since SERS enhancement is the predominant mode in our SERRS contrast agents, we further assessed the contributions of aspect ratio-related and aggregation-related LSPR to the overall SERS enhancement by monitoring changes in peak area upon dilution. The aspect ratio-related LSPR is determined by the morphology of individual particles, while the aggregation-related LSPR is concentration-dependent. In more dilute solutions, nanoparticles have limited interactions with each other, reducing the formation of hotspots for Raman enhancement. For both types of SERRS contrast agents, signal intensities at 100 ppm show a moderate reduction compared to the original, undiluted samples (Fig. S4 and Table S1). When the concentration is reduced from 100 ppm to 10 ppm, Au NR600 SERRS agents exhibit approximately a 30–50% reduction in Raman intensity, while Au NR800 SERRS agents show a 40–50% reduction. However, the reduction in Raman intensity becomes more pronounced at 1 ppm, indicating an accelerated decrease in signal. The moderate reduction in Raman signals at 10 ppm suggests that aggregation-related LSPR plays a dominant role in the SERS enhancement observed in the concentrated solutions of our SERRS contrast agents.

In vitro Raman imaging of the au NR-based SERRS contrast agents

Before evaluating the in vivo imaging performance of our SERRS contrast agents, we assessed their biocompatibility at both cellular and tissue levels. At the cellular level, Au NR600-Hep@PPy-PDA nanoparticles demonstrated negligible cytotoxicity toward normal cell lines, such as L929 mouse fibroblasts, as well as tumor cell lines, including HeLa and MNNG/HOS cells (Fig. 2b). However, at high concentrations, Au NR800-Hep@PPy-PDA nanoparticles induced significant tumor cell death (Fig. 2c). The observed cytoto xicity of the SERRS contrast agents in cancer cell lines is likely due to two primary factors: (1) The inherent antitumor properties of heparin, which has been reported to inhibit tumor growth and metastasis [40, 41]. (2) The Au NR800-Hep@PPy-PDA nanoparticles are speculated to have higher cellular internalization compared to the Au NR600 ones, which has been reported in other researches [42]. To further assess the biotoxicity of the two SERRS contrast agents, each contrast agent was administered intravenously in mice. HE staining images of the control group and treated groups show negligible differences in tissue morphology, indicating that both SERRS contrast agents cause no detectable irritation in major organs, including the heart, liver, spleen, lungs, and kidneys, over a two-week period (Fig. S5).

With their satisfactory biocompatibility established, we evaluated the feasibility of using the SERRS contrast agents for cellular imaging with MNNG/HOS osteosarcoma cells. The interactions between the cells and the SERRS contrast agents were analyzed based on variations in their Raman spectra. Cells incubated with both the Au NR600 and Au NR800 SERRS agents exhibited elevated fluorescence backgrounds (Fig. 2d-e), indicative of dissociation between the Au NR substrates and PPy-PDA reporters. This dissociation is likely due to the degradation of heparin. To minimize this effect, cellular imaging was performed after a 1-hour incubation period. Both Au NR600 and Au NR800 SERRS contrast agents enabled accurate cellular imaging, with Raman bands at 900 cm^− 1 (green) and 1600 cm^− 1 (red) clearly mapped (Fig. 2f). The imaging performance of the SERRS contrast agents was comparable to fluorescence imaging achieved using Cy7-labeled Au NR-based nanoparticles (Fig. S6). These results confirm the accuracy of Raman imaging with our SERRS contrast agents.

In addition to imaging quality, differences between the two SERRS contrast agents were observed in the Raman images acquired from different Raman bands. Cells incubated with Au NR800 SERRS agents displayed well-merged spectral images combining the 1600 cm^− 1 and 900 cm^− 1 bands. However, cells incubated with Au NR600 SERRS agents showed weaker intensity in images derived from the 900 cm^− 1 band compared to those from the 1600 cm^− 1 band. This reduced Raman scattering is likely due to disrupted interactions between the Au NRs and PPy-PDA hybrids, attributed to heparin degradation. Heparin degradation has been reported to occur via heparanase secreted by highly invasive cancer cells [43, 44]. This degradation effect was further validated by incubating the SERRS contrast agents with heparanase in phosphate-buffered saline, which similarly reduced Raman scattering (Fig. S7). The differences in imaging properties between the two SERRS agents are likely influenced by their cellular accumulation. Au NR800-based agents exhibited higher accumulation in MNNG/HOS cells, which promoted nanoparticle aggregation and the formation of “hot spots.” These “hot spots” refer to localized areas of intense electromagnetic fields that arise when nanoparticles aggregate. The overlapping plasmon resonance of closely packed nanoparticles amplifies the Raman scattering signal, significantly enhancing sensitivity. In contrast, Au NR600-based agents showed reduced SERS enhancement due to disrupted interactions caused by heparin degradation, which inhibited the formation of “hot spots.” These cellular imaging results highlight not only the excellent imaging capabilities of our SERRS contrast agents but also the critical role of cellular accumulation and hot spot formation in enhancing imaging properties. For in vivo imaging, where enrichment and clearance processes are more complex, optimizing conditions for intraoperative Raman imaging requires further investigation in animal models.

In vivo Raman imaging of orthotopic tumor models

The feasibility of Au NR SERRS contrast agents for in vivo imaging was further assessed in orthotopic osteosarcoma models. Unlike conventional subcutaneous tumor models, orthotopic osteosarcoma models feature highly heterogeneous vasculature and a complex tumor microenvironment, closely mimicking clinical scenarios. These characteristics are evident in the PA images, where bright signals are produced by blood flow (Fig. 3a). However, this vascular signal interferes with the accumulation and identification of contrast agents. To overcome this limitation, Raman imaging was employed as an alternative method for visualizing malignant tissues.

Raman imaging of the orthotopic osteosarcoma model was conducted after removing the skin of the lap region, simulating the conditions of image-guided surgery (Fig. 3b). Accumulated Au NR800 contrast agents distinctly visualized malignant tissues, with the Raman intensity in the tumor region increasing approximately 5.1-fold compared to surrounding tissues. In contrast, tumor-bearing mice injected with Au NR600 SERRS contrast agents showed reduced imaging quality (Fig. 3c-d). Moderate Raman signals were detected around vascular tissues in mice administered with Au NR600 contrast agents, as opposed to the more uniform and continuous signal distribution observed with Au NR800 contrast agents. The Raman intensity in the tumor region increased by only 3.4-fold in mice injected with Au NR600 contrast agents. Furthermore, Raman images of these mice exhibited a significant rise in fluorescence background and loss of characteristic Raman bands, as shown in the inserted spectra in Fig. 3c and d. These discrepancies in imaging performance between Au NR600 and Au NR800 contrast agents are consistent with the differences observed in cellular imaging experiments. The evaluation of imaging performance in orthotopic osteosarcoma models reinforces the conclusion that SERRS contrast agents with larger aspect ratios, such as Au NR800, are better suited for in vivo imaging. Their superior Raman signal intensity and lower background interference make them ideal candidates for intraoperative imaging in complex tumor microenvironments.

Intraoperative Raman imaging of subcutaneous tumor models

The feasibility of our SERRS contrast agents for imaging-guided surgery was evaluated in subcutaneous tumor models. The optimal accumulation time for the SERRS contrast agents was determined by monitoring changes in PA signals after their administration. The PA spectra of the SERRS contrast agents correspond to the longitudinal LSPR peak of the gold nanorods, with the optimal acquisition wavelengths being 700 nm for Au NR600 agents and 850 nm for Au NR800 agents (Fig. 4a-d).

Following the administration of Au NR600 and Au NR800 SERRS contrast agents, their highest accumulation was observed between 1 and 3 h post-administration (Fig. 4e-f). Additionally, the accumulation of these contrast agents was significantly enhanced by modifying them with targeting moieties, such as cyclic RGD peptides (cRGD). For the Au NR800 contrast agents, the peak value of PA signals showed a 200% increase in the cRGD-modified contrast agents compared to a 71% increase in the unmodified ones (Fig. 4g-h). Similarly, for the Au NR600 contrast agents, the peak value of PA signals demonstrated a 166% increase in the cRGD-modified contrast agents, compared to an 83% increase in the unmodified ones (Fig. 4i-j). These results highlight the efficacy of cRGD modification in enhancing the targeting and accumulation of SERRS contrast agents, thereby improving their imaging capabilities.

Based on the results, a 2-hour duration was selected for intra-operative Raman imaging and imaging-guided surgery following intravenous injection of cRGD-modified SERRS contrast agents. For the excitation wavelength, 785 nm was chosen due to its higher tissue penetration depth and reduced background fluorescence compared to the 638 nm laser. Similar to cellular imaging, two types of Raman images were generated by 900 cm^− 1 and 1600 cm^− 1. The Au NR800-cRGD SERRS contrast agents demonstrated exceptional imaging performance, accurately identifying malignant tissues and clearly visualizing the entire surgical process in Raman images from both the 900 cm^− 1 and 1600 cm^− 1 bands (Fig. 5a). The Raman intensity of the tumor was approximately 4.5-fold higher than that of the surrounding tissues, confirming the superior imaging capability of the Au NR800 SERRS contrast agents. In contrast, the Raman images obtained using Au NR600-cRGD contrast agents showed only discrete signals during surgery (Fig. 5b). This difference aligns with earlier findings in cellular and orthotopic tumor imaging but is even more pronounced in subcutaneous tumor models. A noticeable decrease in the characteristic Raman bands of the PPy-PDA hybrid, accompanied by a significant increase in fluorescence background, was also observed in the Raman spectra of Au NR600 SERRS contrast agents (Fig. 5b). These spectral variations suggest substantial changes in the interactions between Au NRs and the PPy-PDA hybrid. Based on the cellular experiments, we attribute these differences primarily to the aggregation behavior of SERRS contrast agents with different aspect ratios, which significantly impacts their imaging performance.

Photothermal ablation and long-term inhibition of reoccurrence

After surgical resection, some discrete Raman signals from residual tumors (< 1 mm) persisted within the surgical bed, raising concerns about disease recurrence (Fig. 5). To address this, post-surgical treatments, particularly photothermal ablation, were adopted to ensure effective clearance of residual tumors. The treatment process is illustrated in Fig. 6a. Tumor-bearing mice were randomly divided into six groups: saline only (control), routine surgery (S), two groups for imaging-guided surgery (denoted as AHPP600c + S and AHPP800c + S), and two groups for imaging-guided surgery followed by photothermal ablation (denoted as AHPP600c + S + PTT and AHPP800c + S + PTT). Photothermal ablation was implemented using 808 nm laser irradiation (1 W/cm²) for 10 min. Tumor sizes were evaluated via luciferase intensity.

The results indicated significant tumor growth in the routine surgery group, underscoring incomplete tumor removal without imaging guidance. In contrast, imaging-guided surgery groups (AHPP800c + S and AHPP600c + S) demonstrated slower recurrence rates (Fig. 6b-c). Additionally, incorporating photothermal ablation effectively eliminated minor residual tumors, achieving long-term suppression of recurrence. Notably, body weight changes during the entire treatment period were insignificant across groups, confirming the biosafety of our SERRS contrast agents for theranostic applications (Fig. 6d).

Interestingly, recurrence was significantly more pronounced in the AHPP600c + S + PTT group compared to the AHPP800c + S + PTT group. This difference can be attributed to two factors: (1) Superior imaging performance of the Au NR800 contrast agents, enabling more precise removal of malignant tissues during imaging-guided surgery. (2) Higher photothermal conversion efficiency of the Au NR800 contrast agents compared to Au NR600 contrast agents (Fig. 6e-f). The elevated temperature achieved in the AHPP800c + S + PTT group effectively induced necrosis of residual tumors, leading to sustained suppression of recurrence.Ultimately, 100% inhibition of recurrence was observed exclusively in the AHPP800c + S + PTT group, demonstrating its remarkable efficacy (Fig. 6g).

We next analyzed the survival rates across all groups. No mortality was observed in any of the groups, except the control group, within 28 days post-tumor inoculation (Fig. S8). The extracted tumor sizes after the 4-week treatment indicate a higher level of recurrence compared to the in vivo fluorescence signals (Figs. 6c and 7a–c). Notably, only the AHPP 800c + S + PTT group achieved 100% inhibition of recurrence. This discrepancy may be attributed to the limited penetration depth of fluorescence signals, which could result in microsized tumors being undetected during in vivo imaging. Furthermore, histological analysis of major organs revealed negligible side effects in all treatment groups (Fig. 7d), indicating the safety of the applied procedures. Collectively, these results highlight that effective resection of primary tumors is essential for achieving long-term suppression of tumor recurrence.