Characterization of monocarboxylate transporters (MCTs) expression in soft tissue sarcomas: distinct prognostic impact of MCT1 sub-cellular localization

Soft tissue sarcomas (STSs) are an extremely heterogeneous group of rare tumors that arise predominantly from the embryonic mesoderm [1]. STSs can occur in any soft tissue in the body, and may have different etiologies, including external radiation therapy and occupational exposure to certain chemicals such as herbicides [1, 2]. Despite some therapeutic improvements, metastasis and death remain a significant problem in half of STS patients, who present advanced disease [1]. Recent advances on the knowledge related to the oncogenic mechanisms underlying sarcomagenesis will hopefully translate into more effective therapies [1, 2].

It has been shown that most sarcomas exhibit a strong glycolytic phenotype and this phenomenon supplies the rationale for positron emission tomography imaging of STS using ¹⁸F-fluorodeoxyglucose (FDG-PET) [3]. FDG-PET measurements in STS allow primary and recurrent detection of intermediate and high-grade tumors [4], being associated with high histological grade [3‐5], cellularity, proliferative activity, MIB labeling index, and p53 overexpression [5]. Additionally, FDG-PET is proposed as a modality to monitor treatment response in high-grade STS patients [3, 6, 7], as quantitative FDG-PET is significantly more accurate than size-based criteria at assessing STS response to neoadjuvant therapy [6, 7].

The specific changes that occur in tumor microenvironment have been recently identified as key components in carcinogenesis. In fact, the reprogramming of energy metabolism, with special focus on the Warburg phenomenon, i.e., the preference for the glycolytic phenotype even in the presence of oxygen, has been recently included as a “new” hallmark of cancer [8]. Since the hyper-glycolytic phenotype will generate high amounts of lactate inside cancer cells, monocarboxylate transporters (MCTs) play an important role in the extrusion of lactate, contributing to the maintenance of the intracellular pH of tumor cells. MCTs are transmembrane proteins, which are encoded by the family of genes SLC16A, which presently include 14 members. However, only the first four members, MCT1-4, are able to transport lactate and other monocarboxylates across membranes, coupled with a proton [9]. MCT1 has a ubiquitous distribution in human tissues, being involved in both uptake and efflux of monocarboxylates from cells, and is considered an intermediate affinity isoform. MCT2 is a high affinity transporter, being adapted to the uptake of monocarboxylates into cells, and is mostly found in tissues that use lactate as a respiratory fuel. MCT3 has a very restricted distribution, having been identified in the retinal pigment and choroid plexus epithelia. MCT4 is known as a low affinity transporter and has been observed particularly in highly glycolytic tissues, where is essentially responsible for lactate efflux. MCT1 and MCT4 are the best studied isoforms in cancer, as being responsible for lactate transport across the plasma membrane [9].

However, the role of MCTs in solid tumors is still far from being fully characterized [10]. Although pointed out as potential therapeutic targets [11‐15], more information on MCTs’ expression in human solid tumors is needed to further translate knowledge into the clinic context. Some information about the expression and clinical-pathological significance of MCTs is provided for various human tumors [10] like tumors from brain [16], colon [17, 18], breast [19], uterine cervix [20], prostate [21] and lung [22], and the vast majority of studies are focused on tumors of epithelial lineage, but similar studies are lacking for other types of tumors. In sarcomas, MCT1 expression was found in the precrystalline cytoplasmic granules of 7 out of 10 alveolar soft part sarcomas [23], while MCT1, MCT2 and MCT4 mRNAs were found in the human rhabdomyosarcoma cell line RD [24‐26]. Therefore, studies on MCT expression and clinical-pathological significance in sarcomas may contribute to the knowledge on the biology of this tumor.

Therefore, the aim of this work was to assess the expression of MCT1, MCT2 and MCT4, and MCT1/4 chaperone CD147, in a series of sarcomas, and evaluate their clinical-pathological significance.

Recent studies show that FDG-PET may contribute for precise grading and prognosis in different solid tumors, including soft tissue sarcomas [5], as high grade tumors show a much higher uptake of ¹⁸F-FDG due to a higher glycolytic phenotype [3]. Also, it was recently described that FDG-PET should be considered an important imaging modality for therapeutic monitoring in patients with high-grade STS [6, 7]. In this context, metabolic characterization of STS emerges as a possibly relevant approach for STS management, with therapeutic implications as early treatment decisions such as discontinuation of chemotherapy in non-responding patients could be based on FDG-PET criteria [7]. Importantly, the hyperglycolytic phenotype present in this type of tumors, similarly to other solid tumors, may be the basis for the use of new directed therapeutic strategies which are currently in clinical trials [29]. Therefore, tumor metabolic characterization, including MCTs as responsible for lactate efflux from highly glycolytic cells, will have a relevant impact on predicting the group of patients that will benefit the most from this recent therapeutic approach and may have a prognostic value.

STSs are a very heterogeneous type of cancer, comprising over 50 histological subtypes, which are often associated with unique clinical, prognostic and therapeutic features [30]. Although some authors point to a lack of association between FDG uptake and histological type [4], others show that FDG uptake may differ significantly among histologic subtypes, when tumors of all grades were included [3]. Importantly, a low glycolytic phenotype was described in certain STS subtypes including tumors from the lipogenic lineage [3], which supports the low expression of MCT1 and MCT4 we found in the plasma membrane of this sarcoma subtype. This is in accordance with the low tumor cellularity characteristic of these tumors, which probably results in lower total tumor glucose consumption [3]. The low glycolytic rates shown by tumors from the lipogenic lineage could anticipate the expression of MCT2 in the plasma membrane in a way to allow cancer cells to obtain energy from oxidative phosphorylation, via uptake of other substrates like lactate itself or pyruvate; however, MCT2 was expressed in the cytoplasm of the majority of tumors, with only one tumor showing plasma membrane expression of this MCT isoform. Therefore, it seems that MCT2 lactate transport does not have a relevant role in sarcomas, as observed for other types of tumors [17, 20]. When comparing tumor grade, MCT1, MCT4 and CD147 were able to distinguish between low and high grade sarcomas, reinforcing the description of a lower glycolytic phenotype in lower grade than in high grade sarcomas [3‐5].

The molecular mechanisms underlying MCT expression in sarcomas were not under investigation in the present study, however, other studies have addressed this matter using the human rhabdomyosarcoma cell line RD [24, 26]. Both MCT1 and MCT4 expressions were shown to be induced by the protein kinase C (PKC) signaling pathway [24, 26], while MCT1 expression was shown to be inhibited by PKA signaling pathway [24]. However, additional studies are warranted to unveil the detailed downstream signaling pathway involved in MCT regulation, as well as other regulatory mechanisms.

It has been described that the hypoxia-induced emergence of a hyperglycolytic and acid-resistant phenotype will contribute to the invasiveness of cancer cells [31]. In this context, MCTs perform a dual role as lactate transporters, by extruding the end product of glycolysis, and pH regulators, by extruding a proton. Accordingly, in the present study, patients with disease progression have a higher likelihood to harbor tumors with plasma membrane expression of MCT1 and MCT4. Importantly, both MCT1 and MCT4 were able to distinguish between a group with worse survival (positive group) and a group with better survival (negative group), pointing to a possible value of MCTs in prognosis. This is the first study showing an association of MCTs with soft tissue sarcoma patient survival, which is in accordance with the role of these transporters in cancer as well as with their possible use as cancer therapeutic targets.

CD147, the MCT1 and MCT4 chaperone, is a much largely studied protein, due to its parallel function as a matrix metalloproteinase inducer [32]. In fact, CD147 expression has already been previously studied in a cohort of high grade soft tissue sarcomas, however, with no significant prognostic value [33]. In the present study, as the cohort also included low grade tumors, CD147 was significantly associated with high grade tumors. Importantly, when co-expressed with MCTs, CD147 was also significantly associated with other poor prognostic variables and patient lower survival, strengthening the hypothesis already raised in gastric cancer [34], that the prognostic value of CD147 is associated with its co-expression with MCTs, especially with MCT4 in the case of sarcomas. This suggests that the prognostic value of CD147 is mainly associated with its function as chaperone of MCTs, contributing in this way to the hyperglycolytic and acid-resistant phenotype.

Another major finding in the present study was the nuclear expression of MCT1, which was confirmed by immunofluorescence using a sarcoma cell line, and by cell fractionation, followed by Western blot. To the best of our knowledge, this is the first study showing nuclear expression of this protein. As the cellular localization does not fit with the classic role of this protein as a transmembrane transporter, this finding suggests the existence of an additional, not yet described, role of MCT1. Although interactions with other proteins are not well described for MCT1, some interactions can be found in the literature that can help to explain the mechanism underlying MCT1 expression in the nucleus. In fact, large-scale mapping of human protein-protein interactions by mass spectrometry shows an interaction of MCT1 with AP1S1 (clathrin-associated/assembly/adaptor protein, small 1) [35], which supports a mechanism of endocytosis (clathrin dependent endocytosis) that is shared by other proteins, like Notch1 and EGFR, that may alternatively lead to, instead of degradation, trafficking into the nucleus [36, 37]. Therefore, similarly to other plasma membrane proteins which, upon a stimulus, can be directed to the nucleus and play additional roles as transcription modulators, we anticipate that MCT1 may be trafficked to the nucleus to perform an alternative function, not related to lactate transport activity. Other evidence on the factors governing MCT localization like the presence of specific sorting signals [38] or substrate-induced increase in MCT1 plasma membrane expression via GPR109A [39] may help to elucidate the mechanisms leading to MCT1 nuclear localization. Importantly, the presence of MCT1 in the nucleus appears to have a very different biological role than the one currently known, as tumors with nuclear expression of MCT1 show a completely opposite behavior when comparing to tumors expressing MCT1 in the plasma membrane; MCT1 nuclear expression was found to be associated with the lipogenic lineage, low grade tumors and higher overall survival, pointing to a probable role of this protein as tumor suppressor. Further studies are warranted to elucidate this possible new role of MCT1.

This work gives an important contribution for the comprehension of the metabolic alterations occurring in STS, herein demonstrated to be dependent on the cell lineage, showing the value of MCT expression as a predictor of poor prognosis in these tumors. Importantly, the results herein presented provide the first evidence for a different type of MCT1 expression, which may change the current state of the art in what concerns monocarboxylate transporters, as it seems to be associated with a new biological role leading to a poor prognosis in STS.