N-(2-18F-Fluoropropionyl)-L-glutamate(18F-FPGLU) is a potential amino acid tracer for tumor imaging with positron emission tomography (PET). In this study, the relationship between glutamate transporter excitatory amino acid carrier 1 (EAAC1) expression and 18F-FPGLU uptake in rat C6 glioma cells line and human SPC-A-1 lung adenocarcinoma cells line was investigated. The uptake of 18F-FPGLU in C6 cells increased significantly after induced by ATRA for 24, 48, and 72 h, which was closely related to expression of EAAC1 in C6 cells (R=0.939). Compared with the SPC-A-1(NT) control cells, the uptake of 18F-FPGLU on EAAC1 knock-down SPC-A-1(shRNA) cells significantly decreased to 64.0%. In the PET imaging of 18F-FPGLU of SPC-A-1 and EAAC1 knock-down SPC-A-1(shRNA)-bearing mice models, the uptake of 18F-FPGLU in SPC-A-1(shRNA) xenografts was significantly lower than that in SPC-A-1 xenografts, with Tumor/Muscle ratio of 1.67 ± 0.1 vs. 3.01 ± 0.3 at 60 min post-injection. The results suggest that transport mechanism of 18F-FPGLU in glioma C6 and lung adenocarcinoma SPC-A-1 cells lines mainly involves in glutamate transporter EAAC1, which is an important transporter of 18F-FPGLU in tumor cells and may be a novel hallmark of tumor glutamate metabolism PET imaging.
Keywords: N-(2-18F-fluoropropionyl)-L-glutamate; tumor imaging; glutamate transporter; excitatory amino acid carrier 1
As the most commonly used positron emission tomography (PET) tracer for tumor diagnosis, 18F-fluoro-2-deoxy-D-glucose (18F-FDG) also has certain false negative and false positive results(Shreve et al. 1999; Fletcher et al. 2008). It has been reported that 18F-FDG negative tumors may use a different metabolic pathway called glutaminolysis(DeBerardinis et al. 2007; Ward et al. 2012). Glutamine and glutamate play key roles in the adapted intermediary metabolism of tumors(Gao et al. 2009; Rajagopalan et al. 2011; Shanware et al. 2011). Several 18F-labeled glutamic acid and 18F-labeled glutamine have been used for metabolic PET imaging of tumor in humans (Baek et al. 2013; Venneti et al. 2015). High uptake of these amino acid tracers in tumor cells is likely related to the increased expression of amino acid transporters. For example, the upregulated system ASC, especially ASCT2 might contributed to the uptake of 18F-labeled (2S,4R)-4-fluoro-L-glutamine(Ploessl et al. 2012), and 18F-fluoroglutamic acid (BAY 85-8050) transport involved in Na+-dependent XAG- and Na+-independent XC- systems with XC- possibly playing a more dominant role, but both of them showed defluorination in vivo(Krasikova et al. 2011). 18F-labeled (4S)-4-(3-[18F]fluoropropyl)-L-glutamate (BAY 94-9392), another derivative of glutamic acid, whose transport was due mostly to upregulation of system XC-, a potential biomarker for tumor oxidative stressï¼Œcan be useful for detecting system XC- activity in vivo and is considered to be a potential tracer for tumor imaging(Koglin et al. 2011).
Our recently developed novel N-18F-labeled glutamic acid, N-(2-[18F] fluoropropionyl)-L-glutamate (18F-FPGLU), seemed to be a potential amino acid PET tracer for tumor metabolic imaging, with high tumor-to-background contrast in several tumor-bearing mice models. Preliminary studies showed that 18F-FPGLU was
primarily transported through Na+-dependent high-affinity glutamate transporter system XAG-(Hu et al. 2014), but the accurate transport mechanism is unknown. Glutamate transport system includes Na+-dependent excitatory glutamate transporter XAG- system and Na+-independent glutamate transporter XC- system(Avila-Chávez et al. 1997). System XC- (xCT) is overexpressed on tumor c ells and is a potential biomarker for tumor oxidative stress. As an important member of XAG- system, excitatory amino acid carrier 1 (EAAC1), also called excitatory amino acid transporter 3 (EAAT3), localizes to the post-synaptic structure of neurons and surrounding glial cells as regulator of excitatory neurotransmission, and also exists in peripheral tissues, perhaps as metabolic regulators(Bailey et al. 2011). The expression of EAAC1 was known to be regulated by several mechanisms that modify carrier abundance on the plasma membranes and was markedly induced by all tans-retinoic acid (ATRA) in rat C6 glioma cells, which led to strikingly stimulate amino acid influx(Bianchi et al. 2008). However, EAAC1 transporter may be a potential biomarker for tumor molecular imaging. It has not been reported so far. This study investigated the relationship between EAAC1 expression and 18F-FPGLU uptake in C6 rat glioma cells line and SPC-A-1 human lung adenocarcinoma. The uptake of 18F-FPGLU was assessed in ATRA-treated and untreated C6 cells lines, and also in shRNA-mediated EAAC1 knock-down SPC-A-1 cells and the non-targeted (NT) control cells in vitro. Further prospective researches of PET imaging of tumor-bearing mice models with C6, SPC-A-1 and EAAC1 knock-down SPC-A-1(shRNA) xenografts were performed to reveal the correlation between the uptake of 18F-FPGLU and the expression of EAAC1.
Materials and methods
All reagents, unless otherwise specified, were of analytical grade and commercially available. All chemicals obtained commercially were used without further purification. Inveon small-animal PET/computed tomography (CT) scanner was purchased from Siemens (Germany).
Synthesis of 18F-FPGLU
The synthesis of 18F-FPGLU from 4-nitrophenyl-2-18F-fluoropropionate (18F-NFP) via a two-step reaction sequence has been described in detail by the earlier paper(Hu et al. 2014).
Cell Culture and Animal Models
The C6 rat glioma cells, SPC-A-1 human lung adenocarcinoma cells were obtained from Shanghai Institute of Cellular Biology of Chinese Academy of Sciences (Shanghai, China). The cells were cultured in culture flasks containing DMEM medium(for C6 cells) or RPMI 1640 medium (for SPC-A-1) supplemented with 10%
FBS and 1% penicillin/streptomycin at 37oC in a humidified atmosphere of 5% CO2 and 95% air. 24 hours before the experiments in vitro, C6 cells lines or SPC-A-1 cell lines were trypsinized and 2×105 cells per well were seeded into 24-well plates. All animal experimental studies were approved by the Institutional Animal Care and Utilization Committee (IACUU) of the First Affiliated Hospital, Sun Yat-Sen University (approval No.A-173). All efforts were made to minimize animal suffering, to reduce the number of animals used, and to use alternatives to in vivo techniques, if available. The nude mice were obtained from Laboratory Animal Center of the First Affiliated Hospital of Sun Yat-Sen University (Guangzhou, China). The C6, SPC-A-1 and EAAC1 knock-down SPC-A-1(shRNA) tumor models were made using previously described methods(Deng et al. 2011). Tumor cells (2-5-106) were injected subcutaneously and allowed to grow for 1 to 3 weeks. When the tumor reached 6-10 mm (diameter) micro PET/CT scans were done.
C6 induced by ATRA
The rat glioma C6 cells were treated by all trans-retinoic acid (ATRA) 12 h after the passage. Culture medium was substituted with fresh medium (containing DMEM medium supplemented with 10% FBS) in the absence or in the present of ATRA at a concentration of 10 μM from a 10 mM stock solution in DMSO according to the literature16. After the treatment of ATRA for 24, 48 and 72 h, quantitative real-time polymerase chain reaction (qRT-PCR) and western blotting were used to monitored the mRNA and protein expression levels of EAAC1 in ATRA treated C6 and non-treated C6 cells.
Generation of shRNA-mediated EAAC1 knock-down cells.
The method of generation of shRNA-mediated EAAC1 knock-down cells was similar to the literature(Youland et al. 2013). SPC-A-1 human lung adenocarcinoma cells was used for shRNA-mediated EAAC1 knock-down experiment. SPC-A-1 cells were transduced with lentivirus ecoding EAAC1-targeted short hairpin RNAs (shRNA). shRNA sequences were selected from human EAAC1 mRNA NM_004170 and the shRNA fragments were cloned in a lentivirus vector pGLV3 plasmid with the sequence 5′-GCATTACCACAGGAGTCTTGG-3′. A non-specific targeting (NT) shRNA for control was cloned in the same lenvirus plasmid backbone. Lentiviral packaging was performed with trans-lentiviral packaging mix in 293T cells according to the manufacturer’s instructions. SPC-A-1 cells were plated on 6-well plates at 2-105 cells per well. After 24 hours, medium was aspirated and replaced with 100 μL of virus-containing solution was added to each well and incubated at 37oC for 24 h. Cells were selected with puromycin and monitored for green fluorescence protein (GFP) expression. The EAAC1 mRNA expression level was monitored by quantitative real-time polymerase chain reaction (qRT-PCR). The EAAC1 protein expression level was quantized by western blotting.
qRT-PCR for the expression of EAAC1
Relative expression levels of EAAC1 mRNA in C6 and SPC-A-1 cells were calculated using the fluorescence quantitative real-time polymerase chain reaction (qRT-PCR) (Stratagene Mx3000P Real time PCR, Agilent). Total cellular RNA was isolated with the Rneasy mini Kit (TAKARA). 1 μg of RNA was synthesized to cDNA in a 20 μL reaction system with reverse transcriptase buffer, RT Enzyme Mix and primer MIX (Bestar qPCR RT kit, DBI). Conditions for reverse transcription were 5 min at 65oC, 5 min on ice, then 60 min at 37oC and 10 min at 98oC. Oligodeoxynucleotide primers of EAAC1 gene for PCR amplification was
5′-AGTTCAGCAACACTGCCTGT-3′ (forward) and (5′-GTTGCACCAACGGGTA ACAC-3′(reverse). PCR was programmed as follows: 2 min at 94oC, 20 s at 94oC, 20 s at 58oC ï¼Œ then 20 s at 72oC ï¼Œ for 40 cycles. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a initial control and each sample was amplified in triplicate. The relative expression of EAAC1 mRNA compared with GAPDH was calculated by comparative threshold method (2 -ΔΔCt ).
Western blotting for EAAC1
Cells were lysed in a detergent-containing buffer with protease inhibitors for 20 min at 4oC. Glyceraldehyde-3-phosphate dehydrogenase ( GAPDH) was used as a reference protein. After solubilization, cell lysates were collected and centrifuged at 14000 rpm for 10 min. The supernatants were transferred into new tubes, quantification of proteins was performed with Pierce BCA Protein Assay Kit (Thermo) and aliquots of 25 μg were loaded on an 10% gel for SDS-PAGE. After electrophoresis, proteins were transferred to polyvinylidene difluorideï¼ˆPVDFï¼‰membranes (Millipore) . The membranes with EAAC1 or GAPDH were departed at the middle position, and were blocked and incubated with deferent antibody, respectively. Non-specific binding sites were blocked with an incubation in Tris-buffer saline containing 5% of bovine serum albumin (BSA) for 1h at room temperature. Then the blots were exposed to EAAT3 antibody (rabbit monoclonal antiserum, 1:1000, Abcam) or anti-GAPDH rabbit monoclonal antibody(1:3000, Abcam) diluted in blocking solution for at 4oC overnight. After washing, the blots were exposed for1h at room temperature to goat anti-rabbit IgG HRP diluted 1:5000 in blocking solution.
Cellular uptake of 18F-FPGLU
Cells were plated in 24-well plates (2x105cells/well) and uptake studies were performed at 24 h after the passage. The cellular uptake of 18F-FPGLU studies was similar to the methods described previously(Krasikova et al. 2013). The medium was aspirated and the cells were washed 3 times with 1 mL warm PBS. 18F-FPGLU was dissolved in PBS solution and was added to each well (74-111 KBq/0.2 mL/well). After incubated with 18F-FPGLU at 37oC for 30 min, the radioactive medium was removed and the cells were washed 3 times with ice-cold PBS. Then, the cells were dissolved in 0.5 mL of 1 N NaOH and the activity was measured by γ counter (GC-1200, USTC Chuangxin Co. Ltd. Zonkia Branch, China). The cell lysate (25μL) was used for determination of protein concentration by BCA protein assay. The uptake data are based on the amount of activity added to each well and the total amount of protein in each well. Each experiment was done in triplicate, averaged and was repeated 5 times on different days. The uptake of 18F-FPGLU was assessed on the ATRA-treated or untreated C6 cells, and on EAAC1 knock-down SPC-A-1(shRNA) cells or SPC-A-1(NT) control cells. The relative uptake ratios were calculated compared to the control cells.
Small-animal PET-CT imaging
Small-animal PET-CT imaging studies with tumor-bearing mice were carried out using the Inveon small-animal PET/CT scanner (Siemens). 3.7-7.4 MBq of 18F-FPGLU were injected intravenously in conscious animals via the tail vein. The mice were anesthetized with 5% chloral hydrate solution (6 mL/kg) and were kept warm throughout the procedure. Imaging started with a low-dose CT scan, immediately followed by a PET scan. PET images were acquired at 30, 60, 90, 120 min post-injection. For a comparative study, mice were kept fasting for 4 h and were anesthetized with 5% chloral hydrate solution (6 mL/kg) and imaged with 18F-FDG (3.7 MBq) at 60 min after intravenous injection. The images were reconstructed by two-dimensional ordered-subsets expectation maximum (OSEM). For each small-animal PET scan, ROIs were drawn over the tumor and muscle of the thigh on decay-corrected whole-body coronal images using Inevon Research Workplace 4.1 software. The quantification was performed according the methods described previously(Hu et al. 2014). Radioactivity concentration within a tumor or other tissue was converted to MBq/g and then divided by the administered activity to obtain an imaging ROI-derived percentage of injected dose per gram of tissue (% ID/g). Then, the ttumor/muscle (T/M) and tumor/brain (T/B) uptake ratios were calculated, respectively.
Expression of EAAC1 was assessed by immunohistochemistry on formalin-fixed paraffin embedded rat brain tissues and C6 xenograft samples. Immunohistochemistry experiments were carried out according to the literature(Wang et al. 2013). Normal rat brain tissues and C6 glioma tissues were fixed in 10% neutral buffered formalin overnight at room temperature. Tissues were then dehydrated, embedded in paraffin, and cut into 3-μm sections. After antigen retrieval, tissue sections were subject to immunohistochemical incubated with antibodies against EAAC1(Abcam), DAB was stained before mounted onto microscope slides. Tissues were analyzed with a Nikon E800M microscope.
Data were expressed as mean+/-SD. Statistical analysis was performed with SPSS software, version 16.0 (SPSS Inc.), for Windows (Microsoft). Student t test was used to assess differences in the magnitudes of samples from two measurements. A P values of less than 0.05 was considered to indicate statistical significant. A scatter plot was drawn with the relative mRNA expression and the relative uptake of 18F-FPGLU in C6 cells treated with ATRA for 24h, 48h, 72h. Spearman correlation analysis and a linear regression analysis was performed between them.
EAAC1 expression and 18F-FPGLU uptake in C6 cells induced by ATRA
The EAAC1 mRNA relative expression levels in ATRA-treated C6 cells assessed by quantitative real-time polymerase chain reaction (qRT-PCR) are shown by Figure 1A. Compared with the untreated C6 cells, the EAAC1 mRNA relative expression level in ATRA-treated C6 cells treated with ATRA at 10 μM for 24, 48 and 72 h was increased to 1.72 ± 0.11ï¼Œ3.22 ± 0.22ï¼Œ4.0 ± 0.21 times, respectively( Fig. 1A). Meanwhile, the western blotting results also showed that EAAC1 protein expression in ATRA-treated C6 cells was increased gradually(Fig. 1B). Corresponding with the high EAAC1 expression in ATRA-treated C6 cells, 18F-FPGLU uptake was significantly increased to 1.47 ± 0.11ï¼Œ2.14 ± 0.29ï¼Œ2.12 ± 0.16 times in C6 cells treated by ATRA for 24, 48 and 72 h, respectively(Fig. 1C). There was a high correlation between the relative EAAC1 mRNA expresion and the relative 18F-FPGLU uptake in ATRA treated C6 cells (R = 0.939, Fig. 1D). To summarize, EAAC1 expression was markedly induced by ATRA in C6 cell lines. As a result, there was more 18F-FPGLU uptake in ATRA-treated C6 cells line which has more EAAC1 expression at both mRNA and protein levels.
PET imaging on C6 glioma-bearing mice
Small-animal PET-CT scan was performed on C6 glioma-bearing nude mice models 1h post-injection of 18F-FPGLU. PET-CT fusion imaging of the mice models demonstrated that 18F-FPGLU could intensely accumulate in C6 glioma (Fig. 2A). The tumor/brain uptake ratio of 18F-FPGLU on C6 glioma-bearing mice was higher than that of 18F-FDG at 1h post-injection of radiotracers(n = 3, P < 0.05, Fig. 2B). However, the tumor/muscle uptake ratio of 18F-FPGLU in C6 glioma-bearing mice was lower than that of 18F-FDG (n = 3, P < 0.05). Immunohistochemistry showed that widely diffuse EAAC1 transporter staining was shown in C6 glioma, however there was minimal EAAC1 staining in normal rat brain write matter tissue (Fig. 2C).
EAAC1 expression and 18F-FPGLU uptake in EAAC1 knock-down SPC-A-1
human lung adenocarcinoma cells
The influence of EAAC1 expression on 18F-FPGLU uptake was specifically investigated using RNA interference-mediated EAAC1 knock-down SPC-A-1 human lung adenocarcinoma cells. Lentivirally delivered shRNA significantly reduced EAAC1 mRNA expression in SPC-A-1(shRNA) cells, as compared to the non-targeted (NT) shRNA control cells (SPC-A-1(NT) cells), EAAC1 shRNA reduced EAAC1 mRNA expression by 72% in SPC-A-1(shRNA) cells (P < 0.01) (Fig. 3A). At the protein expression level, EAAC1 shRNA significantly decreased EAAC1 expression in SPC-A-1(shRNA) cells by 59.6% (P < 0.01) (Fig. 3B). Knock-down of EAAC1 expression was associated with a significantly lower 18F-FPGLU uptake by 36% in SPC-A-1(shRNA) cells (P<0.05)ï¼ˆFig. 3Cï¼‰. These results demonstrate that EAAC1 is an important mediator of uptake 18F-FPGLU in SPC-A-1 in vitro.
PET imaging on EAAC1 knock-down lung adenocarcinoma-bearing mice
Small-animal PET-CT scan using 18F-FPGLU was performed on SPC-A-1 human lung adenocarcinoma and EAAC1 knock-down SPC-A-1(shRNA)-bearing nude mice models (n =3 per group). PET and CT fusion images of 18F-FPGLU in SPC-A-1 and SPC-A-1(shRNA) tumor-bearing nude mice models were shown in Figure 4A. The tumors were clearly visible with high contrast to the contralateral background (muscle) within SPC-A-1 lung adenocarcinoma animal models and the uptake of 18F-FPGLU in SPC-A-1 lung adenocarcinoma was significantly higher than that in EAAC1 knock-down ones. Regions of interest (ROIs) from the whole tumors on the coronal images were measured so that the accumulation of the radioactivity in the small-animal PET scans could be quantified. The tumor/muscle and tumor/brain ratios of 18F-FPGLU in SPC-A-1-bearing models were higher than that in EAAC1 knock-down ones during the 2 hrs protocol (Fig. 4B). The Tumor/Muscle ratio in the EAAC1 knock-down SPC-A-1(shRNA) lung adenocarcinoma was 1.67 ± 0.1 at 60 min post-injection, which was significantly lower than that in SPC-A-1 ones (3.01 ± 0.3) (n=3, P<0.05). These findings confirmed the potential importance of EAAC1 as a mediator of 18F-FPGLU uptake in SPC-A-1 lung adenocarcinoma.
Malignant tumor cells highly accumulate amino acids and their close analogues due to increased expressing levels of amino acid transporters in the tumor cells(Kobayashi et al. 2008; Li et al. 2011). Adaptations of amino acid transport systems in cancer cells can offer opportunities to provide novel diagnostic and therapeutic targets(Wise et al. 2010; Dang et al. 2011; Bhutia et al. 2015). The Na+-dependent XAG- system is an important transport system for anionic amino acids(Kanai et al. 2013). EAAC1 is an glutamate transporter which localizes in partial peripheral tissues such as lung, muscle and kidney proximal tubules S3 segments to facilitate uptake of glutamate, which is involved into glutamine synthesis and followed by its release into the blood(Massimiliano et al. 2014). Moreover, because it can also transport cysteine, EAAC1 is believed to be important for the synthesis of intracellular glutathione and subsequent protection from oxidative stress(Aoyama et al. 2015).Mature astrocytes are generally believed to be EAAC1-negative, but astroglial tumor cells, such as C6 rat glioma cells and several human glioma cell models express the EAAC1 transporter(Palos et al. 1996). Widely diffuse EAAC1 transporter staining in C6 glioma and minimal EAAC1 staining in normal rat brain white matter tissue in our study also confirmed this phenomenon. Moreover, EAAC1 was positive on the human PC-3 prostate tumor cells(Pissimissis et al. 2009). So, EAAC1 transporter may be a potential biomarker for tumor molecular imaging. Additional biochemical studies suggested that the expression of EAAC1 markedly induced by ATRA in C6 glioma cells, especially in C6 cells cultured with 0.1% fetal bovine serum (FBS) (Bianchi et al. 2008). The results of our experiments demonstrated that EAAC1 also markedly induced by ATRA in C6 glioma cells cultured with 10% FBS. Furthermore, the EAAC1 expression was correlated with the uptake of 18F-FPGLU in C6 glioma cells. Expression levels of EAAC1 in ATRA-treated and untreated rat glioma C6 cells cultured significantly varied, which was positively correlated with the 18F-FPGLU uptake between them. Corresponding with the high EAAC1 mRNA expression in ATRA-treated C6 cells, 18F-FPGLU uptake was significantly increased. On the contrary, compared with non-targeted control cells, expression level of EAAC1 obviously decreased in EAAC1 knock-down SPC-A-1(shRNA) cells, and 18F-FPGLU uptake was strongly suppressed in vitro. Furthermore, changes in EAAC1 expression have a significant impact on 18F-FPGLU uptake in SPC-A-1 lung adenocarcinoma PET imaging. Obviously, these results suggest that EAAC1, as an important mediator of glutamate transport across cell membranes, is an important transporter of 18F-FPGLU in C6 glioma and SPC-A-1 lung adenocarcinoma. Nonetheless, there are likely additional factors that facilitate 18F-FPGLU uptake and retention within cells. Despite robust knock-down of EAAC1 expression in SPC-A-1(shRNA) cells, approximately 59.6 % suppression of EAAC1 protein level lowered 18F-FPGLU uptake only by 36 % in SPC-A-1(shRNA) cells. Radiotracer accumulation within tumor cells may reflect both transporter activity across the cellular membrane and metabolism of the tracer within the cells. The persistent accumulation of 18F-FPGLU in the tumor cells might relate to residual EAAC1 activity within the cell membrane or activity of other XAG- transporters. Of course, additional studies are required to clearly define all the components that mediate 18F-FPGLU uptake in tumors. Nevertheless, the results of this study demonstrated that an important mechanism of 18F-FPGLU uptake associated with EAAC1 expression in tumor, which suggested that 18F-FPGLU PET imaging could be used to diagnose tumors with high EAAC1 expression or to select tumors for using EAAC1-based novel therapeutic strategies.
This study is the first report of a relationship between 18F-FPGLU uptake and EAAC1 expression in C6 glioma and SPC-A-1 lung adenocarcinoma. The results demonstrate that EAAC1 is an important transporter of 18F-FPGLU in tumor cells and may be a novel hallmark of tumor glutamate metabolism PET imaging. Furthermore, these data will provide researchers a more clear understanding of how 18F-FPGLU PET can be optimally used in the field of oncology and motivate additional efforts to perform on the mechanism of 18F-FPGLU transport.