Relationship between [ 14 C]MeAIB uptake and amino acid transporter family gene expression levels or proliferative activity in a pilot study in human carcinoma cells: comparison with [ 3 H]Methionine uptake

Introduction: To clarify the difference between system A and L amino acid transport imaging in PET clinical imaging, we focused on the use of α -[ N -methyl- 11 C]-methylaminoisobutyric acid ([ 11 C]MeAIB), and compared it with [ S -methyl- 11 C]- L -methionine ([ 11 C]MET). The aim of this study was to assess the correlation of accumulation of these two radioactive amino acid analogs with expression of amino acid transporters and cell proliferative activity in carcinoma cells. Methods: Amino acid uptake inhibitor studies were performed in four human carcinoma cells (epidermal carcinoma A431, colorectal carcinoma LS180, and lung carcinomas PC14/GL and H441/GL) using the radioisotope analogs [ 3 H]MET and [ 14 C]MeAIB. MeAIB was used to inhibit the A system and 2-amino-2-norbornane-carboxylic acid (BCH) was used to inhibit the L system. The carcinoma gene expression levels of a number of amino acid transporters were measured by microarray and quantitative polymerase chain reaction. Carcinoma proliferative activity was assessed using accumulation of [methyl- 3 H]-3'-deoxy-3'-fluorothymidine ([ 3 H]FLT). Results and Conclusion: [ 14 C]MeAIB uptake occurred principally via a Na + -dependent A type mechanism whereas [ 3 H]MET uptake, which occurred predominantly via a Na + -independent L type mechanism although other transporters were also utilized depending on cell type. There was no correlation between [ 3 H]MET uptake and total system L amino acid transporter (LAT) expression. In contrast, [ 14 C]MeAIB uptake strongly correlated with total system A amino acid transporter (SNAT) expression and proliferative activity in this preliminary study using four human carcinoma cell lines. Carcinoma proliferative activity also correlated with total SNAT expression. between the accumulation of [ 14 C]MeAIB and the gene expression level of total SNAT as well as the accumulation of [ 3 H]FLT, it is suggested that use of the analog [ 11 C]MeAIB in PET may provide an indication of tumor cell proliferative activity. [ 11 C]MeAIB is therefore expected to be very useful in PET imaging. characterize the expression of the mRNAs encoding SNAT1, SNAT2, SNAT4, ASCT1, ASCT2, y + LAT1, y + LAT2, LAT1, LAT2, LAT3, LAT4, and 4F2hc. We then examined the correlation between accumulation of [ 3 H]MET and the quantitative mRNA expression of total system L transporter (total LAT= LAT1 + LAT2 + LAT3 + LAT4) and total system A amino acid transporters (total SNAT SNAT1 SNAT2 SNAT4) in these human carcinoma cell lines. Finally, we examined the correlation between accumulation of [ 3 H]FLT uptake and [ 14 C]MeAIB uptake as well as the correlation between [ 3 H]FLT uptake and the quantitative gene expression of total SNAT transporters.


2-[ 18 F]fluoro-2-deoxy-D-glucose ([ 18 F]FDG)
, an analog of glucose, is the most commonly used radiopharmaceutical in positron emission tomography (PET)-CT imaging [1]. PET-CT imaging is based on the preferential uptake of [ 18 F]FDG in tumor cells as compared to normal cells, because glucose metabolism is increased in tumor cells. [ 18 F]FDG PET-CT has been found to be useful in lesion detection and characterization, evaluation of tumor stage, assessment of treatment response and detection of recurrent disease [2,3]. However, the specificity of this technique is low in patients with active infections and inflammatory diseases (because of high FDG uptake in macrophages) and in the brain (because of high background FDG uptake) [4,5]. Therefore, the development of post-FDG radiopharmaceuticals is needed.
After glucose transport, amino acid transport is another important pathway in cellular energy metabolism. Therefore, natural or artificial amino acid analogs have been widely studied clinically as potential post-FDG radiopharmaceuticals for PET imaging; One of the most important radiolabeled amino acids in this regard is [S-methyl-11 C]-L-methionine ([ 11 C]MET) [6,7]. Since both amino acid transport and protein synthesis rates are enhanced in tumors, [ 11 C]MET has been as widely used in brain tumor imaging as O- (2-[ 18 F]-fluoroethyl)-L-tyrosine ([ 18 F]FET) [8,9].
Numerous amino acid transporters have been identified at the molecular level and have been characterized in mammalian cells [10][11][12]. The main transport systems for the uptake of neutral amino acids are the A, L, and alanine-serine-cysteine (ASC) amino acid transport systems. System L amino acid transporters are Na + independent, and are the main transport mechanism for methionine, tyrosine, phenylalanine, and their analogs such as FAMT (3-fluoro-α-methyl-tyrosine) [13][14][15][16]. The system A and ASC amino acid transporters are Na + -dependent, however, compared to the system L transporters, their involvement in the transport by radiolabeled amino acids in nuclear medicine has not been studied in detail.
The artificial amino acid radiopharmaceutical α-[N-methyl-11 C]-methylaminoisobutyric acid ([ 11 C]MeAIB) is a promising specific substrate of system A amino acid transport. Compared with [ 11 C]MET, [ 11 C]MeAIB is metabolically stable [17] and it has been studied both pre-clinically and clinically. For example, [ 11 C]MeAIB has been shown to be useful in the measurement of amino acid uptake into skeletal muscle and in the diagnosis of malignant lymphoma and head and neck cancers [18][19][20]. In our institute, [ 11 C]MeAIB PET has proven useful in the diagnosis of chest diseases, especially in the differential diagnosis between sarcoidosis and metastasis [21].
Previous studies in carcinoma cells have shown that there is a high correlation between both MET and FAMT uptake and the gene expression levels of system L amino acid transporters [15,16,22]. However, the relationship between the accumulation of radiolabeled amino acids and the gene expression levels of system A amino acid transporters has not been examined.
In this study, we explored the amino acid transport systems in four different human carcinomas cell lines using α-[  (20 min). We also investigated the gene expression profiles of numerous amino acid transporters in these four types of human carcinoma cell lines using microarray analysis. Following on from this initial screen we used quantitative reverse transcription polymerase chain reaction (qRT-PCR) to characterize the expression of the mRNAs encoding SNAT1, SNAT2, SNAT4, ASCT1, ASCT2, y + LAT1, y + LAT2, LAT1, LAT2, LAT3, LAT4, and 4F2hc. We then examined the correlation between accumulation of [ 3 H]MET and the quantitative mRNA expression of total system L transporter (total LAT= LAT1 + LAT2 + LAT3 + LAT4) and total system A amino acid transporters (total SNAT = SNAT1 + SNAT2 + SNAT4) in these human carcinoma cell lines. Finally, we examined the correlation between accumulation of (18.5 kBq/mL). After incubation with the radiolabeled amino acid analog, the medium was aspirated and the monolayers were rapidly rinsed twice with 500 µL of ice-cold incubation medium. Cells were solubilized in 500 µL of 0.1N NaOH, and the radioactivity (either 3 H and 14 C) of an aliquot (400 µL) was measured by addition of Clear-Sol II (Nacalai Tesque Inc., Kyoto, Japan) and scintillation counting using a LSC-5100 liquid scintillation counter (Hitachi Aloka Medical, Ltd., Tokyo, Japan ).
To characterize relative contributions of each type of transport system to overall amino acid analog uptake, we performed inhibition experiments with inhibitors as described above. BCH was used as a system L inhibitor and MeAIB was used as a system A inhibitor. To calculate the relative contributions of amino acid transporter systems, we used the methods reported by Shikano et al. [23] and Nakajima et al. [24]. In brief, uptake of mRNA levels of neutral amino acid transporters were determined by conducting qRT-PCR assays.
Amplification and real-time fluorescence detection were performed using a model Mx3005P Real Time QPCR system (Stratagene Products Division, Agilent Technologies). All data were normalized using the geometric mean of β-actin and GAPDH. All reactions were performed in triplicate.

Uptake of [ 14 C]MeAIB, [ 3 H]MET and [ 3 H]FLT transport in human carcinoma cells.
As shown in Fig. 1

Competitive inhibition studies using [ 3 H]MET and [ 14 C]MeAIB.
The majority of [ 3 H]MET transport appeared to occur in a Na + -independent manner in all four carcinoma cell lines (Fig. 2). The inhibitor BCH caused a large significant inhibition of this Na + -independent transport indicating the importance of system L. Although, the system A inhibitor MeAIB showed significant inhibition of Na + -dependent transport in H441, PC14, and LS180 cells, the absolute magnitude of inhibition was very small relative to the magnitude of inhibition seen with BCH. The contribution of system A to MET uptake is therefore small relative to the contribution of system L.
By comparing the uptake in MeAIB treated cells in the presence of Na + with uptake in control medium lacking Na + the relative contribution of system ASC and/or other systems to MET uptake can be calculated. The ASC system was found to have a significant contribution to MET transport in PC14 and LS180 cells but not in A431 or H441 cells, although as for system A, the relative magnitude was less than system L.
In contrast, in all four carcinomas, the majority of [ 14 C]MeAIB uptake was Na + dependent ( Fig. 3) and as expected there was a large and significant inhibition of this Na + dependent uptake of [ 14 C]MeAIB by MeAIB indicating the importance of system A. The system L inhibitor had either no or minimal effect on the remaining Na + -independent transport.
3.3. Relative contribution of amino acid transport systems for MET and MeAIB in human carcinomas.
The relative contributions of amino acid transport systems A, L, and, ASC/other to the uptake of  [23] and Nakajima et al. [24] and the results are shown in Table 1. Although the dominant uptake mechanism for [ 3 H]MET in human carcinomas was through system L amino acid transport, system A, ASC, or other Na + -dependent system(s) also played a role. The dominant uptake mechanism of [ 14 C]MeAIB in human carcinomas was Na + -dependent and occurred via system A.

Correlation between [ 3 H]MET and [ 14 C]MeAIB uptake and transporter gene expression or [ 3 H]FLT
in human-derived tumor cells.
With regard to the contribution of system L amino acid transport to the Na + -dependent uptake of [ 3 H]MET the A431 and H441 cells had the highest system L contribution (76.3%and 48.9% respectively).
( Table 1). When the expression of numerous sodium-dependent and -independent amino acid transporters were analyzed using microarray gene expression profiling, as well as by qRT-PCR, the expression of the system L transporter LAT1 and the coupling factor 4F2hc were found to be abundantly expressed in both A431 and H441 cells compared to the PC14 and LS180 cells ( Table 2 and Table 3). Based on the results of the qRT-PCR analysis, the expression of LAT1 varied across the four different carcinoma cell lines (Table 4). When the [ 3 H]MET uptake in the presence of Na + for all four carcinomas was plotted against total LAT (system L) expression no correlation was found (Fig. 4A). On the other hand, there was a significant correlation between accumulation of [ 14 C]MeAIB and total SNAT (system A) gene expression in the carcinoma cells (Fig. 4B). Moreover, there are significant correlations between accumulation of

Statistical Analysis
Data are presented as means and ± SDs. P values were calculated using a two-tailed paired Student t test for comparison between two groups. A P value less than 0.05 was considered significant.

Discussion
The amino acid analog MeAIB is an inhibitor and specific substrate for system A amino acid transport and the main transport mechanism for MeAIB uptake is thought to occur via system A [27][28][29][30][31][32].
A study in Chinese hamster ovary (CHO) cells showed that more than 90% of MeAIB transport occurred through system A [9,33]. A study in cultured human erythroleukemic (K562) cells using radiolabeled [ 14 C]MeAIB also gave the same result [34]. Given all of data, it is thought that system A is the principal [ 11 C]MeAIB transport pathway in in vivo human PET studies [19,20], although no human PET study has been conducted in the presence of amino acid transport inhibitors. In our study using four different types of human carcinoma cells we have shown that system A contributes more than 70% to [ 14 C]MeAIB uptake. In contrast the uptake of [ 3 H]MET was less specific; although system L was the predominant mechanism there were contributions of both system A and system ASC and the relative contributions of all systems varied between the different cell lines (Table 1). Stability studies have shown that more than 95% of [ 11 C]MeAIB remained unchanged in human plasma 30 minutes after administration. Since it is an unnatural amino acid, it also cannot be used for protein synthesis [17]. Therefore, [ 11 C]MeAIB imaging may more accurately represent the uptake of amino acid transporter system A compared to [ 11 C]MET.
Since [ 11 C]MeAIB has higher selectivity for system A compared to [ 11 C]-2-aminoisobutyric acid ([ 11 C]AIB, another system A substrate) [27,31], [ 11 C]MeAIB is expected to be very useful in system A amino acid transport PET imaging. We have previously reported the utility of [ 11 C]MeAIB in the diagnosis of chest diseases, especially in the differential diagnosis between sarcoidosis and metastasis [21]. [ 11 C]MeAIB has comparable sensitivity to [ 18 F]FDG PET for the diagnosis of prostate cancer [35] and the tumor/non-tumor ratio obtained with [ 11 C]MeAIB precisely distinguishes the malignant group from the benign group in patients with brain tumors [36].
On the other hand, the main transport mechanism of [ 3 H]MET in tumor cells has been reported to be system L [21,35], which is consistent with our study results. The radiolabeled amino acid [ 11  used in brain tumor imaging [8,9].

We investigated the correlation between accumulation of [ 3 H]MET and [ 14 C]MeAIB in four types
of human carcinoma cells and gene expression, which led to the conclusion that there was no correlation between [ 3 H]MET uptake and LAT expression (Fig. 4A). A weak relationship between [ 3 H]MET uptake and total LAT expression has previously been reported [22,26]. The reasons for this lack of correlation are not completely clear but could perhaps be related to the fact that LAT1 biology is complex and involves a number of other proteins. One such protein is the amino acid transporter activating factor 4F2hc with which LAT1 forms disulfide-linked heterodimers thereby becoming active. A second protein is ASCT2. LAT1 transports neutral amino acid with long side chains, such as leucine, and simultaneously counter transports intracellular glutamine [38,39]. The intracellular glutamine supply is maintained via the intracellular Na + -dependent amino acid transporter ASCT2 that helps facilitate LAT1 function. The interplay between LAT1, ASCT2, and 4F2hc is thought to be important in controlling cancer cell metabolism [40]. For example, co-expression of LAT1 with ASCT2 has been shown in lung cancer, and this co-expression, but not the sole expression of LAT1 or ASCT2, is strongly related to prognosis [41].

In our study, [ 14 C]MeAIB accumulation in four types of human carcinoma cells was correlated
with total SNAT expression (Fig. 4B). Since SNAT does not require an activating factor, there is a high possibility of a consistent relationship between SNAT expression and cellular uptake. System L, which transports amino acids into cells, has a counter transport mechanism associated with it. In contrast, system A, which also transports amino acids into cells, has no counter transport activity. Amino acid transporter function can therefore be evaluated directly using MeAIB (system A substrate), allowing differentiation from system L [42,43]. The correlation between the uptake of radiotracers and the gene expression of amino acid transporter in tumor cells has been evaluated in a number of recent nuclear medicine studies. [15,16,26], a correlation between [S-methyl-3 H]-D-MET uptake and LAT + ASCT expression [22], and a correlation between [25,44] have all been recently reported. However, it should be borne in mind that gene expression levels may not directly correlate with protein levels. In addition, cell-surface localization and activity of amino acid transporter proteins may not necessarily be predicted by gene expression levels.

Trans-1-amino-3-[ 18 F]fluorocyclobutanecarboxylic acid ([ 18 F]FACBC) uptake and ASCT expression
Our data also showed that [ 3 H]FLT accumulation was strongly correlated with SNAT expression and [ 14 C]MeAIB accumulation (Fig. 5C and D) Glucose metabolism, as well as amino acid metabolism, increases as a requirement to supply an energy source for cell proliferation in cancer cells [48]. Moreover, amino acids and amino acid transporters play important roles other than in energy metabolism, such as in macromolecular synthesis, mTOR activation, and ROS homeostasis beyond energy metabolism [49]. There are approximately fifty different types of amino acid transporters, but only LAT1 [50], LAT3 [51], ASCT2 [52], ATB 0, + [53] and xCT [54] have been reported to be expressed at high levels on the surface of cancer cells. Recently, there have been numerous reports in cancer cells related to glutamine transport via ASCT2 and LAT1 [55][56][57].
Moreover, the expression levels of the SNAT amino acid transporter, which belongs to the SLC38 family and like ASCT2 and LAT1 is related to glutamine transport, have been examined in a range of different cancers. Overexpression of SNAT was observed in gastric cancer [58], human hepatocellular carcinoma [59], breast cancer [60,61], hilar cholangiocarcinoma [62], C6 glioma [63], prostate cancer [25], HeLa epithelial cervical cancer cells, and 143B osteosarcoma cells [64]; and SNAT expression in stomach and breast cancers is also related to Ki-67 [61] and PCNA (proliferating cell nuclear antigen) expression [58] as indicated by proliferative activity.
Reportedly, the accumulation of [ 11 C]AIB, an unnatural amino acid and amino acid transporter system A substrate, accumulation in cancer cells post-radiation therapy is correlated with changes in SLC38A1 expression [65]. Data from the Oncomine analysis (a gene expression database of 947 human cancer cell lines, http://www.broadinstitute.org/ccle) revealed that ASCT2, SNAT1 and SNAT2 were overexpressed, which could lead to a new treatment for preventing proliferation of cancer cells [64]. The studies reported here highlight the potential importance of SNAT in cancer. As described earlier, there have been an increasing number of studies highlighting the role of SNAT in cancer cells. Since      Table 2. Microarray gene expression profiling of sodium-dependent and -independent amino acid transporters in four human carcinoma cell lines. Table 3. Absolute quantification of sodium-dependent and -independent amino acid transporters.