Investigating the influence of gold nanoparticle clustering on macrophage uptake and photoacoustic signal generation to enable cell tracking

Introduction: : Macrophages play a key role in many biological processes such as immune response, tissue repair, and disease pathogenesis. To leverage macrophages as cellular therapeutics, noninvasive cell tracking techniques are needed to monitor their behavior in vivo. Photoacoustic (PA) imaging, an emerging medical imaging modality commonly paired with ultrasound, is a promising approach due to its high-resolution, deep-tissue imaging capabilities with added optical contrast. However, optimizing cell labeling strategies to maximize nanoparticle (NP) uptake and imaging contrast remains a challenge. Gold nanospheres (AuNSs) are commonly used as PA contrast agents due to their biocompatibility, tunable size, and high photostability. When AuNSs cluster, surface plasmon resonance (SPR) coupling alters their optical absorption properties. In the case of cell labeling, AuNS uptake and localization to endosomes causes a broadening of their spectral signature towards longer wavelengths, ideal for PA imaging. However, this clustering is random, leading to unpredictable spectral shifts and PA signals, creating challenges for longitudinal imaging. To overcome this, gold nanoaggregates—pre-clustered assemblies of AuNSs—offer a reproducible alternative. These constructs maintain a stable and predefined PA spectrum, even after cellular uptake. Hyperbranched gold nanoconstructs (HBGNCs) are an example of a gold nanoaggregate, consisting of a spherical gold core with protruding gold branches [1]. Although both HBGNCs and AuNSs have advantages for PA imaging, their cellular uptake and PA performance have not been directly compared. Here, PA imaging performance of macrophages labeled with different sizes of HBGNCs and AuNSs is compared to identify optimal strategies for in vivo cell tracking.

Materials and

Methods: : AuNSs and HBGNCs were synthesized in-house following established protocols to model random and organized NP clustering during cell labeling, respectively [1, 2]. Nanoparticles were characterized by transmission electron microscopy (TEM; Fig. A & Fig. B) and UV-Vis spectroscopy (Fig. C) to assess particle size/morphology and optical absorption properties, respectively. In an 8 mL total reaction volume, varying the seed volume allowed for size control: 6.4 mL of seed yielded ~10 nm AuNSs, 400 µL yielded ~25 nm AuNSs, and 40 µL yielded ~50 nm AuNSs (Fig. A & Fig. B). HBGNC size was similarly tuned based on AuNS seed size, with 25 nm seeds producing ~65 nm HBGNCs and 50 nm seeds producing ~85 nm constructs (Fig. A & Fig. B). Macrophages (RAW264.7) were cultured and incubated overnight with nanoparticles at an optical density of 4. After incubation, media was aspirated, and cells were washed twice with PBS to remove residual NPs and fixed in 10% neutral buffered formalin at a concentration of 2,000 cells/μL. For photoacoustic (PA) imaging, tissue-mimicking gelatin phantoms were prepared. Dome-shaped inclusions contained fixed, NP-labeled cell samples or unlabeled controls. Ultrasound/photoacoustic (US/PA) imaging datasets were acquired using the Vevo LAZR-X system (15-29 MHz; 680–970 nm) to evaluate changes in the PA signal and spectrum following cellular uptake of the different types of nanoparticles.

Results, Conclusions, and Discussions:: Macrophages were labeled with different types of NPs synthesized in-house (Fig. D). Compared to particles in solution (Fig. C), clustering was observed in all AuNS-labeled cell groups, as indicated by broadening of the absorbance spectrum—consistent with plasmonic coupling of aggregated particles (Fig. E). In contrast, HBGNC particles maintained similar spectra (Fig. C) after cell labeling (Fig. E). UV-Vis data indicated a preference for cell uptake of 10 nm AuNSs and 85 nm HBGNCs, based on the highest absorption values and are therefore expected to produce the strongest PA signals (Fig. E). PA images further demonstrate distinct differences in spectral behaviors of AuNSs vs. HBGNCs (Fig. F). AuNSs predominately absorb light at shorter wavelengths, whereas HBGNCs maintain a broad spectrum across the PA imaging wavelength range, with statistically significant differences between the two groups at 700 nm and 900 nm wavelengths (Fig. G & Fig. H). The resulting PA signal from AuNS-labeled cells showed size-dependence with statistically significant differences between each of the three AuNS groups. The 10 nm AuNSs produced the highest signal at 700 nm wavelength, likely due to uptake preference towards smaller NPs and increased plasmonic coupling. In contrast, PA signals from HBGNCs were consistent across sizes (Fig. G & Fig. H). Overall, both AuNSs and HBGNCs show potential for longitudinal imaging, but the data suggest that each has different advantages and NP selection should be guided by requirements of the imaging application. The 10 nm AuNSs produced the highest PA signal, but the peak was observed at shorter wavelengths, limiting imaging depths. A steep spectral decay was also observed, which may be advantageous for spectral unmixing and isolating PA signals of interest. HBGNCs, while generating lower overall signals, had consistent PA spectra post-uptake, extending to longer wavelengths, making them more suitable for deep tissue imaging with more predictable signals. Towards long-term macrophage tracking, future studies should investigate retention within cells for different types of NPs over extended periods of time.

Acknowledgements and/or References (Optional):: 1) Kim, M., VanderLaan, D., Lee, J., Choe, A., Kubelick, K. P., Kim, J., & Emelianov, S. Y. (2023). Hyper-branched gold nanoconstructs for photoacoustic imaging in the near-infrared optical window. Nano Letters, 23(20), 9257–9265. https://doi.org/10.1021/acs.nanolett.3c02177
2) Gao, C., Vuong, J., Zhang, Q., Liu, Y., & Yin, Y. (2012). One-step seeded growth of au nanoparticle with widely tunable sizes. Nanoscale, 4(9), 2875–2878. https://doi.org/10.1039/c2nr30300k