Hypoxia and cellular metabolism in tumour pathophysiology

Abstract

Cancer cells are optimised for growth and survival via an ability to outcompete normal cells in their microenvironment. Many of these advantageous cellular adaptations are promoted by the pathophysiological hypoxia that arises in solid tumours due to incomplete vascularisation. Tumour cells are thus faced with the challenge of an increased need for nutrients to support the drive for proliferation in the face of a diminished extracellular supply. Among the many modifications occurring in tumour cells, hypoxia inducible factors (HIFs) act as essential drivers of key pro‐survival pathways via the promotion of numerous membrane and cytosolic proteins. Here we focus our attention on two areas: the role of amino acid uptake and the handling of metabolic acid (CO₂/H⁺) production. We provide evidence for a number of hypoxia‐induced proteins that promote cellular anabolism and regulation of metabolic acid–base levels in tumour cells including amino‐acid transporters (LAT1), monocarboxylate transporters, and acid–base regulating carbonic anhydrases (CAs) and bicarbonate transporters (NBCs). Emphasis is placed on current work manipulating multiple CA isoforms and NBCs, which is at an interesting crossroads of gas physiology as they are regulated by hypoxia to contribute to the cellular handling of CO₂ and pH_i regulation. Our research combined with others indicates that targeting of HIF‐regulated membrane proteins in tumour cells will provide promising future anti‐cancer therapeutic approaches and we suggest strategies that could be potentially used to enhance these tactics.

General introduction

Mutational transformation of normal cells to uncontrolled growing tumour masses results in unique conditions not experienced in typical physiology. Of note, gas physiology changes occur with moderate to extreme hypoxia (decreased O₂), being a defining feature of solid tumours due to incomplete vascularisation. Poor tumour blood perfusion further results in acidosis of the extracellular compartment due to increased production of metabolic acids in the form of H⁺ (lactic acid) and CO₂ (hypercapnia) stemming from enhanced rates of cellular metabolism. In this symposium report, we focus on the influence of hypoxia and hypercapnia on tumour cell pathophysiology and therapeutic developments targeting protein alterations in these conditions.

Hypoxia has been studied with a particular vigour in the last two decades following the discovery of the hypoxia‐inducible transcription factor (HIF) (Semenza & Wang, 1992; Wang & Semenza, 1993 a, b ) and revelation of the extensive gene network that it regulates (Manalo et al. 2005; Schodel et al. 2011). This broad‐reaching transcription factor has served to revive and transform metabolic studies in cancer cells as HIF has been proven to control key components of proliferative and metastatic mechanisms. Here we will limit our discussion to the influence that hypoxia imposes on the tumour cell's ability to obtain amino acids (and ultimately regulate mechanistic target of rapamycin (mTOR) signalling), utilise glucose for exacerbated glycolytic metabolism and finally enhance H⁺ secretion and pH regulation via CO₂ ventilation to deal with altered tumour cell physiology. We further suggest research directives that could potentially enhance the development of anti‐cancer strategies currently on‐going for HIF‐regulated proteins that promote tumour survival and growth.

Amino acid transport and cancer

Cancer cells experience an increased demand for nutrients to support accelerated proliferation. Thus processes related to the supply of glucose, lipids, amino acids and other intermediates are often up‐regulated in rapidly growing tumours. Essential amino acids (EAAs) are a critical class of nutrient since they are absolutely required for protein synthesis but they cannot be de novo synthetised by human cells. Recent studies investigating the branched chain EAA subclass (leucine, isoleucine and valine) have demonstrated that their catabolism is essential for in vivo brain and lung tumour growth (Tonjes et al. 2013; Mayers et al. 2016) Moreover, a stable level of intracellular leucine is required in order to sustain activation of the master regulator of cell metabolism and growth, mTORC1 (Hara et al. 1998; Wolfson et al. 2016). Therefore, cancer cells clearly rely on an efficient mechanism to obtain EAAs, which, like all hydrophilic nutrients, cannot cross the plasma membrane without transporter proteins.

Among the different EAA carriers, the large neutral amino acid transporter 1 (LAT1, SLC7A5) has been the primary candidate linked with cancer progression (Fig. 1). LAT1 is a Na⁺‐independent obligatory exchanger (stoichiometry 1:1) that promotes EAA uptake combined with glutamine efflux (Verrey et al. 2004). LAT1 forms a heterodimer with the CD98 glycoprotein that acts as a chaperone promoting stabilisation, trafficking and functional insertion of LAT1 into the plasma membrane (Yanagida et al. 2001). Overexpression of CD98–LAT1 has been shown to be a negative prognostic factor in a wide range of tumour types including cutaneous melanoma (Shimizu et al. 2015), gliomas (Haining et al. 2012), prostate (Sakata et al. 2009), lung (Kaira et al. 2010) and renal clear cell carcinomas (Betsunoh et al. 2013). This global LAT1 overexpression pattern in almost all cancers may be explained by the fact that its expression is under control of signalling pathways that are deregulated during carcinogenesis such as the Hippo (Hansen et al. 2015; Park et al. 2016) and Myc (Hayashi et al. 2012) pathways. In addition to oncogenic activation, adaptive pathways are able to promote tumour LAT1 expression. Recently, LAT1 regulation by HIF2α was shown to promote mTORC1 activity in hypoxic kidney tumours (Elorza et al. 2012). This study prompted our interest in studying the reliance of cancer cells on the CD98–LAT1 complex in order to sustain their proliferation in oxygen‐ and nutrient‐deprived microenvironments. Use of a combined genetic and pharmacological dissection enabled us to demonstrate in multiple cancer types (colon, lung and kidney) that the EAA transport activity of the complex is the key limiting step in cancer cell proliferation in vitro and in vivo by promoting EAA homeostasis and mTORC1 activity (Cormerais et al. 2016). Moreover, these data showed that LAT1 activity is required for tumour growth under both hypoxic and nutrient starved conditions as well as in normoxia and an environment rich in EAAs. Furthermore, despite the existence of other EAA transporters in human physiology (i.e. SLC7a6‐8 (Fotiadis et al. 2013), SLC43a1‐2 (Bodoy et al. 2013) and SLC6a14 (Bhutia et al. 2015)), the genetic or pharmacological inhibition of LAT1 revealed a lack of functional EAA transport redundancy. This study further increases the attractiveness of LAT1 as a therapeutic target and encourages continued development of specific inhibitors such as JPH203.

Figure 1. Generalised model for cellular metabolism, nutrient supply and metabolic waste handling in hypoxic tumour cells.

Gene induction via the hypoxia‐inducible factor (HIF) leads to modulation of various cellular pathways that may be exploited for anti‐cancer strategies. AA, amino acids; GLUT1, glucose transporter 1; LAT1, large neutral amino acid transporter 1; MCT1/4, monocarboxylate transporter 1 and 4; mTOR, mammalian target of rapamycin; NBC, Na⁺–HCO₃ ⁻ co‐transporter; NHE1, Na⁺/H⁺ exchanger 1.

Prospective mechanisms to improve EAA targeted disruption as an anticancer strategy

First and foremost, pharmacological development of LAT1 inhibitors for clinical use will have to account for potentially damaging off‐target effects. LAT1 knockout mice are embryonic lethal (Poncet et al. 2014) and a recent study has suggested a critical role of LAT1 in the transport of AAs through the blood–brain barrier (Dolgodilina et al. 2016). Furthermore, LAT1 plays an essential role in activation and function of the immune system (Hayashi et al. 2013; Sinclair et al. 2013). Therefore, identification of potential adverse effects stemming from LAT1 inhibition must occur and this is currently on‐going with a phase I clinical trial of the LAT1 inhibitor JPH203 in Japan.

Additionally, LAT1 inhibition in our work on solid tumours has been shown to be cytostatic and not cytotoxic (Cormerais et al. 2016) in contrast to what was recently published in a T lymphoma model (Rosilio et al. 2015). Therefore, future studies should investigate potential synergy between LAT1 targeting and other commonly used chemotherapeutics. Moreover, involvement of the two amino acid sensing pathways, mTORC1 and general control nonderepressible 2 (GCN2) kinase (for review see Efeyan et al. 2015), in promoting cancer cell survival during JPH203 treatment must be investigated. Although LAT1 inhibition arrests tumour growth via mTORC1 inhibition and induction of the GCN2 amino acid stress pathway, these alterations may lead in parallel to adaptive and pro‐survival mechanisms. Indeed, in response to EAA starvation these pathways inhibit global protein synthesis (Gingras et al. 2001; Howell et al. 2013) and promote autophagy (Hosokawa et al. 2009; Koren et al. 2010; B'Chir et al. 2013), macropinocytosis (Palm et al. 2015) and the increased translation of specific stress response proteins such as AA transporters (Harding et al. 2003; Sato et al. 2004). We could thus predict that disruption of these adaptive pathways may lead to cell death during JPH203 treatment. As proof of concept, a recent investigation of the interactions between the GCN2 and mTORC1 demonstrated that upon AA deprivation, activation of the GCN2 pathway sustained cell survival by induction of the stress response protein Sestrin2, which is required to maintain repression of mTORC1 (Ye et al. 2015).

The global strategy of ‘starving tumour cells to death’ has demonstrated encouraging results (McCracken & Edinger, 2013; Selwan et al. 2016) with many preclinical models suggesting the specific involvement of AA transporters (Babu et al. 2015; Bhutia et al. 2015; Rosilio et al. 2015; Cormerais et al. 2016). However, it appears that an opposing approach to take advantage of EAA transporter overexpression in cancers as a means for toxic drug delivery is appealing. For example, LAT1 has been reported to transport the anti‐cancer molecules melphalan and acivicin (Yanagida et al. 2001; Geier et al. 2013). This suggests that chemotherapy could be designed or chemically modified for recognition as substrates in up‐regulated tumour AA transporters to increase tumour uptake while limiting deleterious effects on healthy tissue. The Ganapathy lab was one of the first to explore chemical modification of certain drugs for recognition by AA transporters. They demonstrated that the esterification of the antiviral drug acyclovir (γ‐glutamylacyclovir) allows transportation by the AA carrier SLC6A14 (Hatanaka et al. 2004). Similarly, derived versions of naproxen comprising the side chain structure of certain AAs had a greater efficacy in the prevention of colon cancer, most likely due to improved absorption by tumour cells (Aboul‐Fadl et al. 2014). Finally, glutamine‐ and tyrosine‐based AA conjugates of monocarboxylate transporter (MCT) 1 and 2 inhibitors improved their anti‐tumour efficacy (Nair et al. 2016). Combined, these studies confirm the possibility of generating anticancer molecules as specific substrates for certain AA transporters thus promoting their therapeutic activity in tumours.

Hypoxia, glycolysis and lactic acid transport

Tumour metabolism has proven to be predominantly glycolytic even in the presence of oxygen as originally proposed in the pioneering work of Otto Warburg (Warburg, 1956). Hypoxia, and particularly HIF1 was shown to further push tumour cells towards a glycolytic phenotype by restricting pyruvate oxidation. Indeed HIF1 was shown to induce pyruvate dehydrogenase kinase 1 (PDK1), thus inhibiting the conversion of pyruvate into acetyl‐CoA by the pyruvate dehydrogenase complex (Kim et al. 2006; Papandreou et al. 2006). Hypoxia induced AKT accumulation in mitochondria, and phosphorylation of PDK1 has recently been demonstrated to be an additional control mechanism in hypoxic tumour cell metabolism (Chae et al. 2016). Exacerbated glycolysis leads to accumulation of lactate that must be dealt with to ensure continued cellular function, which led to an interest in understanding the role of monocarboxylate (MCT) lactate/H⁺ membrane symporters in tumour cells (Fig. 1).

Transport of lactic acid occurs primarily via MCT1 (which favours import) and the HIF1‐regulated MCT4 (which favours export) (Ullah et al. 2006; Halestrap, 2013). Both the hypoxia induction of MCT4 and requirement of the CD147/Basigin chaperone (Kirk et al. 2000) for proper function presented potential drug candidates at the tumour cell membrane, resulting in intense study over the past 10 years. A summary of an extensive body of work using a battery of techniques to disrupt MCT1, MCT4 and CD147 in isolation or combination has revealed that an impressive redundancy exists in the lactic acid excretory machinery to perpetuate cellular metabolism (Schneiderhan et al. 2009; Le Floch et al. 2011; Chiche et al. 2012; Doherty et al. 2014; Polanski et al. 2014; Granja et al. 2015; Marchiq et al. 2015; Pelletier et al. 2015). The combined inhibition of MCT1 and MCT4 has demonstrated extremely strong efficacy in arresting tumour growth in cell lines derived from colon, lung, kidney, lymphoma and melanoma (for an extended summary see Marchiq & Pouyssegur, 2016; Parks et al. 2016 a). Since this blockade of glycolysis reactivates oxidative phosphorylation, current efforts are focused on combined use of MCT inhibitors and mitochondrial inhibitors (phenformin) to achieve sufficient depletion of cellular energy stores to induce rapid tumour cell killing (Marchiq et al. 2015; Marchiq & Pouyssegur, 2016). The targeting of MCT or of the upstream enzyme lactate dehydrogenase A also directly highlights the role of extracellular lactic acid in blunting of the tumour immune cell response (Brand et al. 2016). Whether this action is mediated by lactate, acidosis or both necessitates continued interest, understanding and innovative approaches for future clinical applications.

Hypoxia and pH regulation

Carbonic anhydrases

As stated in the introduction, tumour acidity is a hallmark feature of the pathophysiological environment present around tumour cells (Gatenby & Gillies, 2004; Parks et al. 2013 a). Acidosis in the tumour microenvironment can reach levels in the low pH 6 range (for an extensive review of acidic tumour pH_e see Gillies et al. 2002). This has been attributed primarily to enhanced lactic acid export in order to maintain glycolytic metabolism; however, it is important to note that the same degree of tumour acidosis has been observed in glycolysis‐deficient tumours (Newell et al. 1993; Yamagata et al. 1998) indicating the important contribution of cellular CO₂ production and subsequent hypercapnic acidosis in the extracellular space of tumours.

Confirmation of the role of CO₂ in acidifying the tumour microenvironment stemmed from the discovery that carbonic anhydrase 9 (CA9) is robustly regulated by HIF1 (Wykoff et al. 2000) in poorly differentiated cancers. Carbonic anhydrases are found in six different gene families and primarily catalyse the hydration of CO₂ to form H⁺ and HCO₃ ⁻ (for a recent review see Supuran, 2016). CA9 in particular, which normally has limited expression in regions of the gastrointestinal tract, becomes robustly expressed and serves as a marker of poor prognosis in the majority of solid tumours tested so far (Chiche et al. 2010; Neri & Supuran, 2011; Parks et al. 2013 a). Thus, the cancer‐specific and hypoxia‐induced expression of extracellular facing CA9 represented an explanation for acidification of tumour pH_e via conversion of metabolically produced CO₂, and initial confirmation of this concept was provided by Pastorekova and colleagues (Svastova et al. 2004). Somewhat surprisingly, CA9 also then proved to actively contribute to regulation of pH_i (Swietach et al. 2008, 2009; Chiche et al. 2009). This pH regulatory function was attributed to its impact on promoting tumour proliferation (Chiche et al. 2009). The currently accepted model is that CA9 (and also extracellular facing CA12) function to effectively ‘vent’ the acidity generated by extracellular CO₂ towards the distant vasculature (Swietach et al. 2014). This protection of pH_i provides further explanation for observations of hypoxia‐increased tumour cell viability during acidosis via the protection of cellular energy stores (Parks et al. 2013 b).

Advances have been made to develop CA9 as a drug target with current on‐going clinical trials (McDonald et al. 2016). Key studies that have supported the drug development for CA9 include a syngeneic mouse breast cancer study showing dramatic effects with shRNA and small molecule inhibitors (Lou et al. 2011). Furthermore, synergistic effects were observed with combined CA9 inhibition and classical oncology treatment strategies such as radiotherapy and anti‐angiogenic treatment (Dubois et al. 2011; Doyen et al. 2012; McIntyre et al. 2012). More recently, studies have implicated CA9 in the determination of cancer stem cell populations and the fatal ability of cancers to metastasise (Lock et al. 2013; Papi et al. 2013; Ledaki et al. 2015; Pore et al. 2015). Thus there is reason to be optimistic that the clinical strategies targeting CA9 may prove beneficial in the future.

Despite intense research attention, many questions remain regarding CA function in tumours. Of particular interest is the regulatory and compensatory network that exists for CA isoforms to promote an accelerated cellular progression. Potential CO₂/HCO₃ ⁻/H⁺ sensing and signalling pathways involved are largely unknown. Recent reports of the interesting role of receptor protein tyrosine phosphatase‐γ in sensing extracellular CO₂/HCO₃ ⁻ in the renal proximal tubules and cerebral arteries (Boedtkjer et al. 2016; Zhou et al. 2016) may provide important directives for future work in tumour cells. In addition, we are currently taking advantage of CRISPR‐cas9 technology to disrupt expression of CA isoforms (both intra‐ and extracellular) in a hope to provide models to answer these questions (unpublished data). In addition we hope that these CA knockout models will assist in drug development, as currently to this point no models have existed to verify in vitro specificity. We expect that CA functions in tumour cells will remain an important topic for years to come in the field and will have applications for a wide range of other health‐related questions.

Bicarbonate transport

An integral component of the CA9‐driven pH_i regulatory model in tumour cells lies in the effective recapture of bicarbonate (HCO₃ ⁻) generated at the extracellular surface to buffer intracellular acidity (Parks et al. 2013 a; Swietach et al. 2014) (Fig. 1). Despite early suggestions for this mechanism potentially occurring via reversal of Cl⁻/HCO₃ ⁻ exchangers (for review see Parks et al. 2013 a) minimal experimental evidence has existed for specific bicarbonate transport mechanisms until recently.

We approached this question with a quantitative PCR survey of all members of the bicarbonate transporting family upon hypoxia exposure in multiple cancer cell types in the hope of illuminating a candidate to pursue in the midst of this diverse membrane transporting family. Interestingly, the electrogenic Na⁺/HCO₃ ⁻ co‐transporter (SLC4A4) showed robust and HIF1‐dependent induction in hypoxia, however this was unfortunately limited to one colon cancer cell line (Parks & Pouyssegur, 2015). We observed that the constitutive expression of SLC4A4 indeed plays an important role in breast and colon tumour cell proliferation, pH_i regulation and migration (Parks & Pouyssegur, 2015). These SLC4A4 results were recently confirmed by others (McIntyre et al. 2016) and further focus on SLC4A4 will be required for integration into overall tumour cell pH_i regulating models. Furthermore, the recent paper from Harris's group reveals a complex induction pattern of HCO₃ ⁻ transporters in response to prolonged 0.1% O₂ exposure, which has led them to propose that the SLC4A9 transporter plays an essential role in tumour progression (McIntyre et al. 2016). It appears we are currently just at the ‘tip of the iceberg’ with respect to our understanding of HCO₃ ⁻ transport in tumour cells under hypoxic conditions.

Although hypoxia was used as a tool to identify HCO₃ ⁻ transport targets due to the presumed association with hypoxia‐regulated CA9, non‐hypoxia‐dependent HCO₃ ⁻ transporters have been shown to be relevant in breast cancers. This work was initiated by a genome‐wide association study that implicated the electroneutral NBCn1 (SLC4A7) with breast cancer susceptibility (Ahmed et al. 2009). An extensive body of work from Pedersen's group and colleagues has since complemented this initial finding. NBCn1 expression is associated with a truncation in the ErbB2 receptor and contributes to pH_i regulation (Lauritzen et al. 2010, 2012; Boedtkjer et al. 2013; Gorbatenko et al. 2014). More recent work has demonstrated an increased importance for NBC activity in breast cancer patient organoid studies (Lee et al. 2015). Further confirmation of NBCn1 importance was provided this year with NBCn1‐ko mice exhibiting a significant delay in the development of chemically induced breast cancer (Lee et al. 2016). Therefore it appears that as for carbonic anhydrases, numerous HCO₃ ⁻ transporting family members (SLC4 A4/A7/A9) can play an important role in tumour progression. Uncovering these interactions and strategies to effectively target the multiple isoforms will form challenges in future research.

H⁺ extrusion mechanisms and interactions with CO₂/HCO₃ ⁻ balance

As we noted above, tumour acidity is generated via a combination of metabolically produced H⁺ and CO₂. Thus far, we have focused primarily on CO₂‐dependent mechanisms but H⁺ extrusion is obviously something that cannot be ignored (Fig. 1). Indeed, pioneering work on Na⁺/H⁺ exchanger 1 (NHE1) mutants resulted in the development of the cellular H⁺‐suicide technique allowing the cloning of human NHE1 and subsequent identification of NHE isoforms 1–9 (Pouyssegur et al. 1984; Sardet et al. 1989; Counillon & Pouyssegur, 2000). This concept of cellular killing via excessive cellular H⁺ accumulation led to mouse models, which formed the proof of principle that disruption of tumour cell pH_i regulation via NHE inhibition could be exploited as an effective anti‐cancer strategy (for an extensive summary see Pouyssegur et al. 2001; Parks et al. 2013 b). Despite promise in pre‐clinical trials, NHE inhibitors have not gained traction in clinical applications for cancer therapy. Genomic editing (zinc finger nucleases and CRISPR‐cas9) is enabling new research in precise ablation of NHE expression with noticeable impacts being reported for breast cancer in vivo xenografts (Amith et al. 2015) and in vitro spheroids (Andersen et al. 2016). Currently we are investigating NHE1‐ko cells in combination with CA‐ko to determine relative importance in tumour growth (Parks et al. 2016 b). It appears that inhibition of NHE mechanisms may be essential in combination with targeting of CO₂/HCO₃ ⁻ regulating proteins to effectively disrupt tumour pH_i.

Incidentally, the focus on MCT function for cellular metabolism has also revealed their relative contribution to H⁺ extrusion. MCTs are not considered as pH_i regulators in the classical sense as their function is not driven by the need to regulate pH but rather by the need to transport monocarboxylic acids. A good example of this is observed in normal muscle fibres where high‐intensity exercise can lead to pain and fatigue due to H⁺ accumulation (pH_i decrease) which results due to a low‐affinity MCT extruding mechanism for lactate/H⁺ (Halestrap, 2013). Thus in normal cells, MCTs do not respond to a pH_i disturbance directly. However, tumour cells continually produce both lactate and H⁺ providing the two substrates required for MCT function. Experimental observations of pH_i in MCT‐ko models both in vitro and in vivo have revealed pH_i acidification in the absence of MCTs (Chiche et al. 2012; Marchiq et al. 2015). Therefore we now consider that MCTs play an important role in pH_i regulation, primarily through control of the set‐point for resting pH_i. Complementary actions of NHE and MCT H⁺‐extruding mechanism in tumour cells must therefore be considered in future efforts to exploit tumour acidity.

Perspectives: CO₂/pH balance in the tumour microenvironment and immune response

A point of interest raised in the Physiology 2016 (Dublin) ‘Gas exchange and disease’ symposium related to alterations in $P_{\mathrm{C}{\mathrm{O}}_2}$ levels and subsequent changes in disease susceptibility. In particular, the issue of increased mortality rates in response to increased $P_{\mathrm{C}{\mathrm{O}}_2}$ levels experienced during pulmonary disease was raised in the context of altered susceptibility to infection. Mechanistically, many unknowns still exist in this process, but progress is being made at the level of the host immune response. Of note, impairment of immune cell function by increased CO₂ has now been well established (Cummins et al. 2010; Wang et al. 2010; Gates et al. 2013). Interestingly, certain events are proposed to occur as a direct response to CO₂ (i.e. nuclear factor‐κB (NF‐κB) signalling; Cummins et al. 2010) while others involve the acidosis associated with hypercapnia (i.e. HIF signalling; Parks et al. 2013 b; Selfridge et al. 2016) raising the legitimate possibility of a role for direct CO₂ sensing. Recently, the capacity for macrophages to perform autophagy and bacterial killing was directly linked to inhibition by CO₂ and not acidosis (Wang et al. 2010; Casalino‐Matsuda et al. 2015). Since tumour microenvironments present increased $P_{\mathrm{C}{\mathrm{O}}_2}$ in a pathophysiological context as compared with normal tissues, it raises an intriguing question about the potential role of CO₂ in the tumour immune response. Currently, immune therapy is dominating the oncology research discussion due to its curative effect, particularly in cancer types (i.e. melanoma) that historically have resisted any advanced therapeutics. Unfortunately, impressive advances in immunotherapy (PD‐1, CTLA4, etc.) are thus far limited to a subset of patients ranging from ∼20–40% depending on the cancer type (for recent review see Topalian et al. 2016). It is tempting to predict that tumour‐associated carbonic anhydrase expression patterns could be an influencing factor in the tumour immune response. Control of CO₂ via CAs could act to potentially stunt the immune response, and impede efforts of immune‐checkpoint therapies. CO₂ is normally rapidly converted to H⁺ + HCO₃ ⁻ in the tumour microenvironment due to external facing CA9/12 activity. Thus could CA inhibition potentially hinder a tumour immune response via elevation of $P_{\mathrm{C}{\mathrm{O}}_2}$? On the contrary, tumour acidosis can supress other important components of the immune response (T and NK‐cells) as was reported recently for lactic acid (Brand et al. 2016). Thus, could counteracting tumour acidosis with CA inhibition promote immune response mechanisms favourable to tumour suppression? Evidence is being generated to support the counteraction of tumour acidosis via buffer therapy as an important mechanism to improve anti‐tumour effects of immunotherapy, in particular T‐cell activity (Pilon‐Thomas et al. 2016). It appears that the combined effect of CO₂ and pH changes induced by CAs in the tumour environment may provide important starting points in future research with respect to immune cell function and efficacy of tumour immunotherapy.

Summary

Tumour‐associated hypoxia clearly provides a growth and survival advantage for cancer cells. Mounting evidence continues to illuminate hypoxia targets as potentially effective anti‐cancer drugs. Here we have expanded on the particular role of amino acids and pH homeostasis in tumour cells within the context of pathophysiological changes in O₂ and CO₂. It appears that many interesting concepts remain to be discovered, in particular for tumour carbonic anhydrase activity, which is at the interface of O₂ and CO₂ balance within the tumour microenvironment. Moving forward we predict that future efforts to disrupt nutrient import and handling of metabolic waste will prove to be effective individually and in synergy with other promising therapeutic developments such as immunotherapy.

Additional information

Competing interests

None declared.

Author contributions

All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

S.K.P. and Y.C. are funded by the Centre Scientifique de Monaco (CSM). J.P. in addition to CSM funding and GEMLUC received support from the Institute for Research on Cancer and Aging (IRCAN), CNRS, INSERM, Centre A. Lacassagne, University of Nice‐Sophia Antipolis and from the Ligue Nationale Contre le Cancer (LNCC).

Biographies

Scott Parks' work has focused on cellular membrane‐transport physiology, first during his PhD at the University of Alberta on aquatic organisms with Dr Greg Goss, followed by transitioning to France to work with Dr Jacques Pouysségur on hypoxia and tumour cell metabolism initially at the University of Nice and now at the Centre Scientifique de Monaco.

Yann Cormerais has recently completed his PhD in the Pouysségur lab in Monaco where he has become an expert in tumour amino‐acid control pathways and mTOR signalling.

Jacques Pouysségur obtained his PhD in 1972 on bacterial genetics at the University of Lyon followed by a two year postdoc at the National Cancer Institute of the NIH. He established his research group in 1978 at the CNRS Biochemistry Centre of the University of Nice. After directing the CNRS ISDBC institute (1997‐2008), his team joined the IRCAN institute in Nice and the Centre Scientifique de Monaco. His research has spanned the areas of bacterial and somatic cell genetics, Na⁺/H⁺ exchange, pH regulation, G protein‐coupled receptors and MAP kinase signalling in the context of growth control in mammalian cells. His group has been interested for the last 15 years in hypoxia signalling, angiogenesis, cancer metabolism, nutrient sensing and amino‐acid transport.