Immunomodulation of Curcumin on Adoptive Therapy with T Cell Functional Imaging in Mice

Adoptive T-cell therapy is a promising therapeutic strategy to treat malignancies and has provided impressive results in cancer patients (1, 2). However, this approach remains challenging because of the lack of persistence and trafficking, and the loss of effector function of transferred T cells (3). The immunosuppressive mechanism within the tumor microenvironment has been proposed as playing a critical role in mediating the antitumor effect of adoptive therapy (4–6). TGF-β is one of the most important immunosuppressors that cancer cells use to hamper the T-cell response. TGF-β has been shown to promote tumor growth through angiogenesis and to inhibit the activation and proliferation of CD8⁺ T cells by suppression of cytolytic gene expression, such as perforin, granzyme B, IFN-γ, and Fas ligand (7). Moreover, TGF-β is attributed to the development, maintenance, and recruitment of regulatory T cells (Treg) that actively attenuate tumor-specific activation of T cells (8, 9). Indoleamine 2,3-dioxygenase (IDO), another immunosuppressor with ability to catalyze the essential amino acid tryptophan via kynurenine pathway, is also a major impediment to successful adoptive therapy (7, 10). Because T cells are highly sensitive to tryptophan shortage, IDO has been proposed to inhibit T-cell proliferation and cause T-cell anergy by degrading tryptophan (11). Indeed, the expression of IDO is found in most human tumors, characterized by a lack of accumulation of CD8+ T cells, and serves as a significant predictor of poor prognosis in ovarian and colorectal cancers (12–14). Because IDO can be induced by IFN-γ, IDO-negative cancer cells may be initiated to express IDO when exposed to an inflammatory context in the tumor microenvironment, and thereby hinder CD8⁺ T-cell–mediated tumor regression before it can be effective (4, 5, 12). Recently, strong evidence has indicated that modification of the tumor microenvironment, such as the blockade of the TGF-β signaling pathway (15, 16), inhibition of IDO (17, 18), or depletion of Tregs (19, 20) is able to augment the cytotoxicity of CD8⁺ T cells.

Curcumin, a natural compound derived from the plant Curcuma longa, has previously been shown to possess anticancer properties through mediating multiple molecular targets related to apoptotic and survival pathways in tumors (21). Other studies have also provided important insight into the relationship between curcumin and the immune system. Curcumin can inhibit the production of inflammatory cytokines and enzymes through inhibition of NF-κB–binding activity of innate and adaptive immunity and thus has potential against various autoimmune diseases and inflammations (22, 23). Moreover, curcumin has been found to mediate the growth and function of various immune cells (22, 24). Some studies show that curcumin can attenuate tumor-induced depletion of T cells and restore antitumor activity of CD8⁺ T cells in tumor-bearing animal models (25, 26). Although the connection between curcumin, the immune system, and growth inhibition of cancer needs to be further investigated, the above evidence suggests that the multitargeting benefit and immunomodulation of curcumin may render adoptive T-cell therapy more efficient in combination treatment.

To determine the combined effects of curcumin with adoptive therapy, we first examined the function of CD8⁺ T cells and immunosuppressive factors in vitro and with an animal model. Because CD8⁺ T cells exert cytotoxicity through the perforin/granzyme B pathway, the expression of granzyme B signifies their full differentiation and acquisition of killing potential toward tumors. We have earlier validated an imaging strategy by linking a reporter gene to the granzyme B promoter coupled with CMV enhancer, allowing noninvasive monitoring of T-cell receptor (TCR)-based T-cell activation in living subjects using bioluminescent imaging (BLI; ref. 27). We herein utilized a E.G7/OT1 animal model that involved adoptive transfer of OVA-specific CD8⁺ T cells transduced with a granzyme B promoter–driven firefly luciferase and tomato fluorescent fusion reporter gene (pGBeLT) to evaluate the immunomodulatory effect of curcumin in OVA-expressing E.G7 tumor–bearing mice. Our results support the better understanding of curcumin-mediated immunomodulation and suggest the application of multitargeting drugs as a complementary strategy in adoptive therapy.

Eight-week-old male C57BL/6 mice purchased from the National Laboratory Animal Center, Taiwan, and OT-1 transgenic mice obtained from the Jackson Laboratory were housed in the Laboratory Animal Center of National Yang-Ming University. All animal protocols were approved by the Animal Care and Use Committee at National Yang-Ming University. OVA-expressing E.G7 mouse lymphoma cell line from American Type Culture Collection (ATCC; Jan 2010) was maintained in cRPMI-1640 (RPMI-1640, 10% FBS, 1% penicillin/streptomycin, and 25 mmol/L HEPES) with 400 μg/mL G418. Jurkat human T-cell leukemia cell line from ATCC (Jan 2010) was cultured in cRPMI-1640. Lentiviral-producing HEK-293FT cell line from Invitrogen (Jan 2010) was maintained in cDMEM [Dulbecco's modified Eagle's medium (DMEM), 10% FBS, 1% penicillin/streptomycin, 2 mmol/L l-glutamine, and 0.1 mmol/L MEM nonessential amino acids] with 500 μg/mL G418. All cell lines were authenticated by short tandem repeat profiling prior to their usage and passaged for less than 6 months in our experiments. Cell culture tested Mycoplasma free routinely with MycoAlert Mycoplasma Detection Assay (Lonza).

Curcumin (Sigma-Aldrich) was dissolved in 0.5 N and 0.025 N NaOH at the stock concentrations of 60 mmol/L and 7 mg/mL for cell culture and in vivo studies, respectively, as previously described (28).

Splenocytes from OT-1 transgenic mice were harvested, strained through a 40-μm filter (BD Biosciences), and subjected to erythrocyte lysis using an ACK lysis buffer. After washing, CD8⁺ T cells were isolated using a mouse T-cell enrichment kit (Stemcell Technologies). CD8⁺ T cells were prestimulated in cRPMI-1640 containing 2.5 μg/mL Concanavalin A (Calbiochem), 50 μmol/L β-mercaptoethanol (β-ME; Bio-Rad), 10 ng/mL interleukin (IL)-7 (R&D Systems) for 36 hours, and thereafter maintained in cRPMI in the presence of 50 μmol/L β-ME, 10 ng/mL IL-15 (R&D Systems), 10 U/mL IL-2 (R&D Systems), and 50 mmol/L α-methylmannoside (Calbiochem) for 24 hours.

High-titer lentiviruses were produced using transient cotransfection of HEK-293FT cells with pGBeLT, the packaging vector pCMVΔ8.74, and the envelope-encoding vector pMD2.G as previously described (27). Titration of lentiviruses was carried out using Jurkat cells. Briefly, 1 × 10⁵ Jurkat cells were transduced with serially diluted viruses for 3 to 4 hours and then activated with 1 μg/mL of anti-human CD3 and CD28 antibodies (R&D Systems) to trigger the expression of reporter gene. The titer of lentiviruses was determined using flow cytometry. In some experiments, mouse CD8⁺ T cells were prestimulated for 16 hours prior to lentiviral transduction (MOI = 5). Thereafter transduced CD8⁺ T cells were cultured in cRPMI with 50 μmol/L β-ME, 10 ng/mL IL-15, 10 U/mL IL-2, and 50 mmol/L α-methylmannoside for 24 hours prior to further study.

Cell lysates were run on SDS-PAGE, then transferred to polyvinylidene difluoride membranes. The membranes were blocked with 5% nonfat milk for 1 hour and probed with anti-mouse IDO monoclonal (Millipore), anti-mouse TGF-β monoclonal (Cell signaling), anti-mouse active caspase-3 (Millipore), and anti-mouse β-actin monoclonal (Millipore) antibodies (mAb) at 4°C overnight. Membranes were washed in Tris-Tween buffer saline and incubated with horseradish peroxide–conjugated secondary antibodies (Jackson ImmunoResearch Laboratories) at room temperature for 1 hour. Final detection was achieved by using an ECL chemiluminescent detection system (Millipore). ImageJ (NIH) was used for the quantitative analysis.

Eight-week-old C57BL/6 male recipient mice were injected subcutaneously with 2 × 10⁶ E.G7 cells/100 μL PBS, and tumors were allowed to grow for 6 to 7 days. For drug and T-cell alone treatment, mice received 70 mg/kg/d curcumin intraperitoneally for 5 days and 10 × 10⁶ or 5 × 10⁶ CD8⁺ T cells intravenously, respectively. For combined treatment, one day after curcumin administration, tumor-bearing mice were intravenously injected with 5 × 10⁶ CD8⁺ T cells (see Fig. 3A for details). Tumor volumes were measured with a digital caliper and calculated according to the formula: length × width² × 0.523.

Tumors, spleens, and tumor-draining lymph nodes (TDLN) were harvested from mice, and single-cell suspensions were stained with the following conjugated Abs: anti-Vα2-FITC (BD Pharmingen), anti-Vβ5-PE (BD Pharmingen), anti-CD69-PerCP (Biolegend), and anti-CD8-APC (Biolegend) to determine the activation and percentage of adoptively transferred CD8⁺ T cells during the treatment. For the intracellular cytokine staining assay, cells were first stained for surface markers, fixed and permeabilized, and finally stained with intracellular IFN-γ–PerCP/Cy5.5 Ab (Biolegend). To identify Tregs, single-cell suspensions were stained with anti-FOXP3-Alexa Fluor 488/CD4-APC/CD25-PE Abs using a Mouse Treg Flow Kit (Biolegend) according to manufacturer protocols. Sample acquisition was done with a FACSCalibur flow cytometer (BD Biosciences), and the data were analyzed using FlowJo (Tree Star).

Mice injected with pGBeLT-transduced CD8⁺ T cells received 150 mg/kg d-luciferin intraperitoneally and were anesthetized using 1% to 3% isoflurane 15 minutes before imaging. The photons emitted from tumors were detected by the IVIS50 Imaging System (Xenogen). All images were acquired for 5 minutes. Regions of interest from displayed images were drawn around the tumor sites and quantified as photons/s/cm²/sr using the Living Image software (Version 2.20; Xenogen).

Tumors were removed from mice, frozen immediately with dry ice, and then embedded with Tissue-Tek OCT. Ten-micrometer cryosectioning was prepared using a cryostat microtome (Leica) and fixed in acetone for 10 minutes. After washing, the slides were mounted in mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories). The slides were scanned with Olympus FV1000 laser confocal microscopy (Olympus), and the fluorescent signal was analyzed using FV10-ASW 1.6 Viewer (Olympus).

Student t test was used to evaluate the significance or P values between groups (*, P < 0.05; **, P < 0.01). SEM values were depicted as error bars in all figures.

The expression of TGF-β and IDO in E.G7 cells treated with IFN-γ and/or curcumin was assayed to examine the effect of curcumin on immunosuppressive factors of tumor cells. The protein level of IDO was increased in response to IFN-γ induction but inhibited by the presence of curcumin (Fig. 1A). TGF-β was also suppressed by curcumin in a dose-dependent manner (Fig. 1B). To examine whether TCR-based T-cell function could be improved by curcumin via inhibition of TGF-β and IDO of tumor cells, CD8⁺ T cells derived from OT-1 transgenic mice, which expressed TCR specific to the chicken OVA peptide at residues 257–264 in the context of H2-K^b, were used. E.G7 cells as the target were pretreated with 10 μmol/L curcumin for 24 hours and pGBeLT-transduced CD8⁺ T cells were cocultured with the target at a 1:50 (E:T) ratio for 4 hours. The activation of T cells was accessed by detection of tomato fluorescent signal and IFN-γ production using flow cytometry. The amount of activated and IFN-γ–producing CD8⁺ T cells was significantly higher (about 2- and 1.5-fold, respectively) in curcumin-treated versus untreated E.G7 cells (Fig. 1C).

To examine the immunomodulation of curcumin in vivo, E.G7 tumor–bearing mice were treated with 70 mg/kg/d curcumin for 5 consecutive days when tumor volume reached 150 mm³, and tumors were excised to analyze the expression of TGF-β and IDO with Western blot. No antitumor effect on E.G7 tumors and no significant change in body weight (<20%) were found at the end of curcumin treatment, suggesting that there was no general toxicity of curcumin in this study (data not shown). The expression of TGF-β and IDO from E.G7 tumors was suppressed by curcumin as compared with those of the control (Fig. 2A). Furthermore, the percentage of CD25⁺/FoxP3⁺ out of CD4⁺ cells in TDLNs of curcumin-treated mice was 1.13 ± 0.15 versus 1.65 ± 0.17 of the control group, that is, 30% reduction (P = 0.043; Fig. 2B).

Because curcumin inhibits not only the expression of TGF-β and IDO but also the induction of Treg in the tumor microenvironment, it is noteworthy to explore whether a small number of CD8⁺ T cells can be used to enhance tumor targeting when combined with curcumin in living subjects. According to our previous finding, 10 × 10⁶ OVA-specific CD8⁺ T cells could lead to the complete tumor regression in E.G7 tumor–bearing mice and was considered as the regular dose in these studies (27). Here, E.G7 tumor–bearing mice were treated as the following 5 groups: control, 70 mg/kg/d curcumin, 10 × 10⁶ CD8⁺ T cells (10 MT), 5 × 10⁶ CD8⁺ T cells (5 MT), and 5 × 10⁶ CD8⁺ T cells + 70 mg/kg/d curcumin (combination of 5 MT + 70 mg/kg/d curcumin) when tumor size reached about 150 mm³. Mice were intraperitoneally injected with curcumin one day before T-cell transfer, designated as day −1, for 5 days, and the treatment protocol is shown in Fig. 3A. Tumors, TDLNs, and spleens were harvested and subjected to flow cytometric analyses to determine the presence and activity of transferred Vα2+Vβ5+ CD8+ T cells on day 3 postadoptive therapy, the time at which there was no significant difference in tumor size among the 5 groups (Fig. 3B). The accumulation of Vα2+Vβ5+ CD8+ T cells in tumors and TDLNs of combination-treated mice was enhanced by 3- to 4-fold and 2-fold, respectively, as compared with those of the 5 MT group, and reached to about the same levels of the 10 MT group (Fig. 3C). The activity of Vα2+Vβ5+ CD8+ T cells was judged by CD69 expression. In Fig. 3D, the percentage of Vα2+Vβ5+ CD8+ T cells expressing CD69 in tumors from combined therapy was 60% higher than that of 5 MT group, but no expression of CD69 was seen in the spleens of all 3 groups. The results suggested that curcumin does not alter the function of transferred CD8⁺ T cells present in the spleen.

Evidence from preclinical studies indicates that alteration of the tumor microenvironment using TGF-β signaling inhibitor or IDO-blocking agent 1-methyl tryptophan (1-MT) results in the restoration of T-cell response (15–18). In our preliminary experiments, we have tested whether BLI coupled with pGBeLT reporter is sensitive to TCR-dependent T-cell function during adoptive therapy combined with either TGF-β receptor I kinase (TβR-I) inhibitor or 1-MT in living subjects. The BLI revealed slight but not significant increase of T-cell activation in the combination group on day 3 postadoptive transfer (Supplementary Fig. S1A and S1B). Flow cytometric analysis also confirmed more transferred T cells (∼2-fold) present in E.G7 tumors from 1-MT–treated mice (Supplementary Fig. S1C).

Because curcumin efficiently increased the number and activation of adoptively transferred T cells within tumors according to flow cytometric analysis (Fig. 3), we further examined whether pGBeLT reporter constructed in these T cells could allow noninvasive and sequential tracking of T-cell targeting and activation, and reflect the improvement of TCR-dependent T-cell function. Similar to the treatment protocol in Fig. 3B, when E.G7 tumors reached about 150 mm³, mice were tail-vein injected with 10 × 10⁶ pGBeLT-transduced CD8⁺ T cells, and 5 × 10⁶ pGBeLT-transduced CD8⁺ T cells plus 70 mg/kg/d curcumin to compare the signal intensity of T-cell activation. As early as 4 hours after adoptive therapy, mice were subjected to BLI for baseline scans, and little bioluminescent signal was detected from E.G7 tumors of both groups after 24 hours (Fig. 4A). The photon signal of pGBeLT in the combined group was dramatically increased and peaked at day 3 postadoptive therapy similar to that of the 10 × 10⁶ T cells group. Sequentially, the decrease in BLI intensity paralleled the corresponding marked tumor shrinkage from day 3 to the end of the experiment (Fig. 4A and B). Because the release of granzyme B from CD8⁺ T cells could induce apoptosis of target cells, the expression of caspase-3 in E.G7 tumors on day 3 was examined. Caspase-3 expression was not obvious in the control and curcumin-treated groups but was increased in the 10 × 10⁶ T cells and combined groups (Fig. 4C). In addition, the BLI signal of the combined group was significantly higher than that of the 10 × 10⁶ T cells group on day 5 postadoptive transfer. We further examined whether curcumin could maintain antigen-specific CD8⁺ cells within tumors. Immunofluorescent staining confirmed the persistence of transferred CD8⁺ T cells in E.G7 tumor of the combined group (Fig. 4D).

For decades, adoptive T-cell therapy has been proposed to mediate cancer regression, however, clinical trials highlight the need to continually improve this strategy mainly due to the suppressive nature of the tumor microenvironment (5, 6). Recent studies indicate that using specific inhibitors, mAbs, or siRNA to inhibit TGF-β signaling, IDO, and Tregs can reverse immune suppression in the tumor microenvironment, resulting in the augmentation of T-cell response, but the safety and effectiveness of such treatments have not been carefully evaluated in clinical trials. For example, small-molecule inhibitors, such as 1-MT may be toxic via promotion of inflammatory conditions (29). Although using siRNA to silence immunosuppressive genes of tumor cells is ideal, efficient siRNA delivery in vivo is still challenging and ongoing (18, 30). Specific mAb used to inhibit immunosuppressive factors may also bind to other cells and tissues expressing normal level of targeted proteins or receptors, for example, anti-CD25 Ab used for reducing Tregs may also target activated CD8⁺ T cells simultaneously (31, 32). Other studies show that combined adoptive T-cell therapy with chemotherapy may induce synergistic antitumor effect in animal models, nevertheless, the effect of combined treatment is not long lasting, and the toxicity of chemotherapy remains a valid concern (2, 33, 34).

Curcumin has been shown to regulate transcriptional factors, multiple protein kinases, antiapoptotic proteins, cytokine signaling receptors, and growth factors, which render it with antitumor and immunomodulatory effects (21, 22). However, little is known about the effect of curcumin when combined with T-cell therapy both in cell culture and in living subjects. In cell culture studies, we found that CD8⁺ T-cell activation was significantly increased when cocultured with curcumin-pretreated E.G7 tumor cells, suggesting that inhibition of TGF-β and IDO of tumor cells could be a major contributor (Fig. 1).

Although curcumin exhibits promising results in various studies in vitro, poor absorption and rapid metabolism continue to be a major concern in preclinical models and clinical trials when it is taken orally (35). Pan and colleagues reported that low plasma concentration of 0.13 μg/mL was detected at 15 minutes, whereas a maximum plasma concentration of 0.22 μg/mL was obtained after 1 hour with oral gavage of 1.0 g/kg of curcuma in mice. After intraperitoneal administration of 0.1 g/kg, the plasma level peaked (2.25 μg/mL) at 15 minutes and declined rapidly within 1 hour (36). To avoid the low bioavailability of curcumin for the clinical application, various advanced drug delivery systems, such as nanoparticles, liposomes, nanoemulsions, and the combination of the adjuvants have been designed, but their biopharmaceutical attributes need to be further validated (35, 37). In our E.G7/OT-1 animal model, we found that the critical time for T-cell activation is the first 3 days after adoptive therapy. To observe the biological effects of curcumin within this short period of time, we applied the intraperitoneal injection, instead of the oral administration, to achieve the higher maximum level in the blood circulation.

Other reports and our previous data have shown that multiple administration of 70 to 120 mg/kg curcumin slightly inhibits tumor growth and attenuates inflammation of normal tissues in small animal models, suggesting that this dosage has low to moderate toxicity to tumors and a modulatory effect on the immune system (28, 38, 39). For the same reason, E.G7 tumor–bearing mice were treated with 70 mg/kg/d curcumin for 5 consecutive days to examine the inhibitory effect of curcumin on immunosuppressors. As expected, curcumin inhibited TGF-β and IDO expression of E.G7 tumors and efficiently reduced the differentiation of Tregs in TDLNs (Fig. 2). We found that the percentages of transferred CD8⁺ T cells both in TDLNs and E.G7 tumors in combined treatment was significantly increased as compared with those of the 5 MT group on day 3 postadoptive therapy. CD69 expression in these intratumoral transferred CD8⁺ T cells was also slightly, but not significantly, elevated as compared with those of the 5 MT group. Furthermore, the augmentation of T-cell response by curcumin in mice was antigen specific because only the baseline number of transferred CD8⁺ T cells was detected in nontarget EL4 tumors (Supplementary Fig. S2). However, effects of curcumin on other immune cells may also exist. Dendritic cells (DC) are essential for the induction of adaptive immunity responsible for tumors, but also promote T-cell tolerance via inhibition of T-cell functions in certain circumstances (40). Although the exact mechanism to suppress T cells used by DCs remains poorly understood, some work has shown that murine and human IDO-expressing DCs in TDLNs are able to degrade tryptophan and create profound T-cell anergy (41, 42). Interestingly, curcumin also inhibits IDO significantly in IFN-γ–stimulated murine bone marrow–derived DCs in vitro, resulting in reversing IDO-mediated suppression of T-cell function (43). Therefore, the effect of curcumin on IDO-expressing DCs of the host may be another mechanism contributing to enhanced efficacy of adoptive therapy and is worthy of further study.

We observed the improvement in accumulation and activation of transferred T cells in target tumors by curcumin, but this important information from flow cytometric analysis only provided a snapshot of the immune response. Recent advances in molecular imaging, including MRI, PET, SPECT, and optical imaging, offer powerful strategies to examine immune cell trafficking patterns and their functional status in living organisms, providing dynamic information which otherwise is unavailable by conventional techniques, such as histology, reverse transcriptase PCR, and flow cytometry. However, most publications about T-cell imaging have been limited to invasive imaging with intravital microscopy or noninvasive imaging of T-cell locations and then functional activity of transferred T cells is implied if tumor size is reduced (44–46). Recently, ¹⁸F-FAC, a new PET imaging probe, has been shown with increased accumulation in proliferating T cells through the deoxycytidine salvage pathway (47–49). Their results indicate that ¹⁸F-FAC is more sensitive to alterations in lymphoid mass and immune status than ¹⁸F-FDG. However, the retention in lymphoid organs and tumor uptake of ¹⁸F-FAC hampers the detection of weak immune responses at these sites (45). Ponomerev and colleagues created another PET imaging probe by placing GFP and herpes simplex virus type 1 thymidine kinase genes under the control of a nuclear factor of activated T cells enhancer element and successfully detected antibody-triggered T-cell activation, but T-cell response was not examined in tumor-bearing animal models (50). Previously, we have established an imaging probe specific to T-cell activation based on the expression of granzyme B, a major mediator of CD8⁺ T-cell cytotoxicity, to monitor the dynamics of antigen-specific T-cell activation over time using BLI (27). In this study, there are 2 key factors to determine the effectiveness of combined therapy in vivo: first, whether transferred CD8⁺ T cells are eliminated by pretreated curcumin before interacting with tumor cells; second, whether those T cells accumulated in target tumors are still functional. According to BLI data, the level of T-cell activation in combined therapy was comparable with that in the regular dose group throughout the entire experiment. The result also shows that multiple administrations of 70 mg/kg/d curcumin is acceptable. Comparison between the tumor growth curves of combined and regular dose treatments shows tumor shrinking was initiated on day 3 postadoptive transfer, which was in coincidence with the peak photon flux signal of T-cell activation (Fig. 4A and B). In addition, tumor regression at early time points was mainly due to CD8⁺ T-cell–mediated apoptosis because active caspase-3 expression of tumors in the curcumin-alone group was not different from that of the control on day 3. The cytotoxicity of T cells toward tumors in the combined and regular dose groups was similar, as indicated by the level of caspase-3 expression (Fig. 4C). The results indicate that the combined treatment has similar antitumor effect as that of regular dose T cells in E.G7/OT-1 mouse model. However, the persistence of T cells in the combined treatment on day 5 postadoptive therapy seems to have no better tumor growth control as compared with that of the regular dose. The reason may be that more homogeneous expression of OVA-specific TCRs on CD8⁺ T cells from OT-1 transgenic mice may exert a strong immune response efficiently. This explanation needs to be further validated using different adoptive transfer models.

In summary, this imaging system has the potential to accelerate the study of adoptive immunotherapy in preclinical cancer models. In addition, our study provides the rationale that the combined therapy of CD8⁺ T cells with multitargeting drugs, such as curcumin, may open the possibility for clinical investigation in cancer patients.

No potential conflicts of interest were disclosed.

This work was supported by grants NSC99-3112-B-010-015 and NSC 100-2321-B-010-013 from National Science Council, Taipei, Taiwan (J-J. Hwang). S.S. Gambhir is funded in part by the U.S. National Cancer Institute In Vivo Cellular and Molecular Imaging Center (ICMIC) P50 CA114747.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.