One of the greatest challenges of translating nanotechnologies to the clinical realm is optimizing in vivo delivery. Maximizing AuNP accumulation at the tumor site has the potential to enhance photothermal cancer therapy, as well as other applications such as optical imaging. In this study, we show that human T cells can be used to transport AuNPs to distant tumor sites following intravenous administration. Following short term incubation with AuNPs, T cells can be efficiently loaded with over 14,000 AuNPs per cell without affecting cell viability, proliferation, and cytokine production. Importantly, T cells loaded with AuNPs retain their ability to migrate in vitro, and demonstrate tumor-specific homing in mice. Using T cells as a vehicle to deliver AuNPs resulted in a four-fold increase in the efficiency of AuNP tumor accumulation, demonstrating that active transport of AuNPs by cellular chaperones is superior to that of passive accumulation through the EPR effect.
Stephan et al.  recently demonstrated that synthetic drug-carrier nanoparticles could be stably conjugated to the surface of immune cells, including T cells, for delivery of therapeutic molecules. In these studies, T cells efficiently carried surface-tethered nanoparticles to tumors in mice, and when loaded with cytokines to support T cell growth, dramatically increased antitumor efficacy. However, our study conclusively demonstrates in vivo that internal loading of AuNPs in T cells can improve tumor localization, and thus may be a useful technology for a variety of nanoparticle based therapies.
In this study, we elect to use AuNPs. AuNPs are known to have low cytotoxicity, and gold has been used in humans for the treatment of arthritis for over 50 years , which makes AuNPs a logical choice in the pursuit of clinical applications. For this study, 40-45 nm gold colloidal nanospheres were selected for internalization by activated human T cells. The internalization of nanoparticles by cells is believed to be accomplished predominantly by receptor-mediated endocytosis, and particle size is an important variable in determining the kinetics of cellular uptake, with maximal uptake in a size range of 40-50 nm [35, 36]. We selected the size of our AuNPs for this proof-of-concept delivery study to optimize nanoparticle cellular uptake. We modulated the degree of nanoparticle internalization by altering the concentration of nanoparticles incubated with the T cells (Figure 1c). We also evaluated nanoparticle uptake using T cells isolated from three different human donors (Figure 1c) and saw only small variation, suggesting that this technique could be extrapolated to the T cells of any patient.
The internalized gold colloid used in this study also had no detrimental impact on the viability or function of activated human T cells in vitro (Figure 2), and the T cells were able to migrate to tumors in vivo while maintaining their AuNP payload (Figure 3). In addition to their ability to carry AuNPs to tumors, T cells can be selected for tumor-specificity for adoptive immunotherapy studies [37–39]. Furthermore, T cells may be genetically engineered to improve their function [40, 41] or enhance their ability to migrate to tumors in vivo[42, 43]. It has been demonstrated that systemically administered AuNPs tend to accumulate mainly in the perivascular regions of the tumor , limiting passive accumulation of nanoparticles by the EPR effect to well-vascularized regions of the tumor. T cells may naturally localize to tumors, and tumor-specific T cell clones have been demonstrated to penetrate into the hypoxic cores of the tumors in vivo. The more extensive infiltration of tumor sites by antigen-specific T cells may permit enhanced penetration of the tumor when compared to freely-injected nanoparticles, potentially augmenting therapeutic efficacy.
The use of T cell vehicles also significantly affects nanoparticle biodistribution (Figure 4). Freely injected nanoparticles (40-45 nm gold colloidal nanospheres coated with 5000 MW PEG) accumulate most significantly in well-vascularized organs such as the liver, spleen, kidney, and gut (Figure 4). Maximal AuNP tumor accumulation for the freely injected PEG-AuNP group is seen at 24 h (Figure 5). After 24 h, increased gold content for the PEG-AuNP group is seen in the spleen, liver, and kidney with a corresponding decrease in gold content within the tumor and other organs, which represents a shift towards AuNP clearance.
AuNP-T cells present a much different biodistribution from the systemically administered nanoparticles that correlates with the expected biodistribution of T cells. After adoptive transfer of AuNP-T cells, a large percentage of the ID is seen within the liver and lungs at 24 h. T cells are known to accumulate within the liver and lungs after administration due to the vascularity and number of adhesion molecules present in these organs . This pattern of T cell migration is consistent with the biodistribution of adoptively transferred T cells seen in previous studies [33, 45]. AuNP-T cells are also seen accumulating in the spleen and bone of the mice; these locations are also normal reservoirs of T cells . The large number of AuNP-T cells seen in the liver likely represents apoptotic T cells. This large accumulation is not observed by bioluminescence imaging in Figure 3, and the liver is a known site where apoptotic T cells are entrapped . Tumor accumulation of AuNP-T cells increases from 24 to 48 h as T cells escape from the lungs and migrate to the tumor (Figure 5). The biodistribution of AuNP-T cells matches the expected biodistribution of normal activated T cells, suggesting that AuNP biodistribution can be modulated based on the selection of the cellular vehicle. In the case of T cells, it is possible that the biodistribution may be altered to to further favor tumor accumulation and persistence by manipulating cell culture conditions  or by genetic modification of T cells .
Using T cells as cellular vehicles for AuNP delivery, we achieve a four-fold increase in tumor delivery efficiency at 48 h when compared to freely injected PEG-coated AuNPs at 24 h (Figure 5). This represents a significant increase in delivery efficiency (P < 0.01, Student's t-test) using T cells. These results demonstrate for the first time that T cells can be used to enhance AuNP delivery to a tumor in vivo. The use of AuNPs and T cells together combines the photothermal therapy and imaging advantages of AuNPs with the immunotherapy and biodistribution advantages of T cells. Future directions will focus on utilizing the AuNP-T cell system for cancer therapy by modifying the T cells to further enrich AuNP tumor accumulation and enhance anti-tumor effects.