An appraisal of the current status of inhibition of glucose transporters as an emerging antineoplastic approach: Promising potential of new pan-GLUT inhibitors

The necessity of glucose for neoplastic cells

It has remained a hot and debated topic if sugars feed cancer. Among the diverse types of sugars, glucose is the simplest one, the most assimilable form of carbohydrates (Parker, 2020). Cancer cells metabolize glucose at a comparatively faster rate for which they need to have a high uptake of glucose. This feature of neoplastic cells was initially recognized by a well renowned German scientist Sir Otto Heinrich Warburg, Nobel laureate of 1931 in Physiology or Medicine, who noted that neoplastic cells consume a relatively higher amount of glucose (20 times more) than normal healthy counterparts (Warburg et al., 1927; Warburg, 1956). Further, Warburg also presented evidence demonstrating that cancer cells metabolized most of their consumed glucose via lactate fermentation instead of utilizing glucose metabolism through the TCA cycle. Figure 1 depicts the typical pathway of glucose metabolism in cancer cells in which glucose is metabolized to lactate compared to normal cells, which metabolize it through the TCA cycle. The role of glycolysis in neoplastic cells is accelerated on an average by 100 times (range: 20–300 times) (Vaupel and Multhoff, 2021). Though the glycolysis is inferior compared to the Krebs cycle for the generation of ATP, cancer cells still evolved to opt for glycolysis even in the presence of O₂ owing to two possible reasons: 1) The rate of ATP production in cancer cells is approximately 100 times faster than normal cells and 2) glycolysis provides precursors for biosynthetic machinery (Lunt and Vander Heiden, 2011; Zhou et al., 2018). The metabolic switching of cancer cells is mainly attributed to HIF and its downstream signaling pathways (Semenza, 2010; Zhao T et al., 2014; Masoud and Li, 2015; Nagao et al., 2019; Lee et al., 2021).

Glucose transporters (GLUTs)

Glucose uptake of cancer cells is facilitated by glucose transporters which belong to the solute carriers (SLC) family. The most prominent GLUTs are highly overexpressed in cancer cells and encoded by the SLC2A gene family (Thorens and Mueckler, 2010; Mueckler and Thorens, 2013; Navale and Paranjape, 2016; Galochkina et al., 2019; Echeverría et al., 2021). In addition, another group of membrane-associated glucose transporters belongs to the Na⁺/glucose cotransporters (SGLT) gene family designated as the SLC5A, which carries out the active transport of glucose (Wright et al., 2011).

The credit for discovering glucose transporters goes to LeFevre (1948), followed by advanced research to decipher their structure and functions primarily in the erythrocyte membranes (Widdas, 1952). Fourteen members now constitute the GLUT family with different affinities for glucose transport (Uldry and Thorens, 2004; Thorens and Mueckler, 2010; Mueckler and Thorens, 2013). Figure 2 depicts the typical structural organization of GLUT as downloaded from PDB Data Bank. A GLUT molecule is a glycoprotein with N-linked oligosaccharides composed of 500 amino acids arranged in an array of 12 transmembrane alpha-helices. Depending on the sequence differences, the nomenclature of the various members of the GLUT family was proposed (Joost and Thorens, 2001; Joost et al., 2002; Holman, 2020). According to the Heidelberg Unix sequence analysis, GLUTs are grouped into three classes: Class I, II, and III. Class I comprises GLUT1, GLUT2, GLUT3, GLUT4 and GLUT14. The class II members are GLUT5, GLUT7, GLUT9, and GLUT11. Class III constitutes GLUT 6, GLUT8, GLU10, GLUT12, and GLUT13/HMIT-1 (Ancey et al., 2018). The GLUT isoforms also display tissue-specific need-based variations in their expression pattern (Boado et al., 1994; Uldry and Thorens, 2004; Calvo et al., 2010). As neoplastic cells display essential dependency on glucose, they are reported to overexpress various isoforms of GLUTs (Table 1), particularly GLUT1, GLUT3, GLUT4, and GLUT12 (Martell et al., 1997; Szablewski, 2013; Barbosa and Martel, 2020; Pliszka and Szablewski, 2021). However, the expression of GLUT1 and GLUT3 is most ubiquitous in all cancers (Krzeslak et al., 2012; Barron et al., 2016); hence the following description of the literature review is focused on GLUT1 and GLUT3.

The GLUT1 and GLUT3 display about 64% resemblance of amino acid composition. Deng and Yan (2016) have elegantly described the crystal structure-dependent details of GLUT1 and GLUT3. GLUT1 comprises 492 amino acids from aa 9–455 GLUT constitutes prominent canonical facilitator super family (MFS) protein folds (Deng et al., 2014; Galochkina et al., 2019; Drew et al., 2021). Twelve transmembrane α helices are linked with extracellular and cytoplasmic linker amino acids. The structural organization of GLUT1 and GLUT3 has been reported to exist in three conformational states.

Despite amino acid differences, all GLUT isoforms possess a similar group structure, The central aqueous channel of GLUT1, through which glucose is passaged, comprises transmembrane domains 3,5,7, camp8 and 11 with amphipathic helices (Mueckler et al., 1985). Groups of Alvarez et al. (1987), Jung (1996), and Heinze et al. (2004) aided in the understanding of the structural organization of GLUT1 using specialized techniques like circular dichroism and Fourier transform infrared spectroscopy, mass spectroscopy, and scanning glycosylation mutagenesis (Alvarez et al., 1987; Jung, 1996; Heinze et al., 2004). Subsequently, the chemically predicted structures were reconfirmed using computer modeling techniques (Deng et al., 2014), followed by final corroboration of the predicted structures of GLUT1 by X-ray diffraction of crystal structures, with 12 transmembrane domains clustered into two units of six helical units (Deng et al., 2014; Holman, 2020). GLUT1 contains a single glucose binding site near the C terminal (Deng et al., 2015). Glutamine residues form the glucose binding site at 161, 282, 283, and 288 and tryptophan at 412 positions (Mueckler and Makepeace, 2009). The in silico structure and docking prediction tools have greatly added to deciphering the binding sites of glucose and inhibitors on GLUT and predicting the ‘Modus Operandi’ of the transport process (George Thompson et al., 2015). The two most acceptable models predict the transport of glucose by GLUT1. In the first model, it is predicted that glucose binding sites are located on two sides of the aqueous channel; one is located towards the exterior and the other towards the cytoplasmic side. Hence named as two site/flexed site transport model (Custódio et al., 2021). The glucose transport depends on a simple ligand exchange at these two binding sites. However, GLUT1 is demonstrated to transport glucose unidirectionally at a given time and hence is a uniporter. The second popular glucose transport model by GLUT1 is named the “alternating access model” (Lloyd et al., 2017). According to this model, GLUT1 undergoes transformational oscillations by alternatively opening on either side. Hence, GLUT1 is predicted to be an antiporter in this model (Chadt and Al-Hasani, 2020). The pioneering study of Deng et al. (2014) published in ‘Nature’ demonstrated the existence of four conformational states of GLUT1 based on its crystal structure. These four conformational states are designated as 1) outward open conformation with one substrate; 2) a ligand-bound state and occluded state; 3) an inward open state; and 4) a ligand-free and occluded state (Deng and Yan, 2016). A recent review by Galochkina et al. (2019) has summarized the current state of knowledge on the mechanism of glucose transport by GLUT1 by molecular dynamic simulation studies. According to the model proposed by this study, glucose transport by GLUT1 comprises three stages of a highly cooperative process. The movement of glucose involves its transition via a rotational movement with H-bonds (Galochkina et al., 2019), which is followed by the transformation of the outside open conformational state to the ‘inside open’ conformational state, through which the queued glucose molecule is released in the cytoplasm (Galochkina et al., 2019). The cytoplasmic glucose gets phosphorylated immediately by the action of hexokinase; hence, the concentration of free glucose in the cytoplasm remains at a very low concentration, which in turn becomes the driving force for the passive transport of glucose by GLUT (Medina and Owen, 2002). Figure 3 depicts various conformational changes which facilitate glucose transport across the cell membrane. This mode of glucose transport by GLUT is also designated as gated pore and rocker switch mechanism (Nomura et al., 2015), membrane phospholipids associated cell signaling has been reported to play a crucial role in regulating the signaling events responsible for the oscillation of GLUT conformational states (Hresko et al., 2016).

Nevertheless, in addition to glucose transport, GLUT1 also mediates the uptake of mannose, galactose, glucosamine, and ascorbic acid by cells (Uldry et al., 2002; Pliszka and Szablewski, 2021), which also work through the same operating system of altered conformational states (Pragallapati and Manyam, 2019). Moreover, the expression of GLUT on cell membrane displays clustered and focal distribution patterns, in which cell junctions and cytoskeletal elements play an essential role, as revealed by direct stochastic optical reconstitution microscopy (Yan et al., 2018). Almahmoud et al. (2019) performed a series of in silico studies to comprehend the mechanism of the aforementioned conformational changes in GLUT1. Their systemic molecular docking investigation revealed that the Phe291, Phe379, Gln380, Trp388, and Trp412 amino acid residues mediate the binding of glucose with GLUT1 and determine its conformational oscillation (Almahmoud et al., 2019). Further, the 12 transmembrane domains of GLUT1 are clustered in a replicated trimeric organization, as revealed by crystallographic studies (Holman, 2020). Crystallographic studies on GLUT1 have also revealed additional insight into the conformational states (Holman, 2020). Some amino acids are crucial in maintaining various conformational states, particularly the hydrophobic residues at 291, 292, 293, 294, 386, and 387 positions (Holman, 2020).

The expression of GLUTs in normal healthy cells and transformed cells is a highly regulated process, depending on the glucose needs of specific cell types (Navale and Paranjape, 2016). However, the precise mechanism of the gene expression and membrane transporters of GLUT1 and GLUT3 have still not been fully deciphered. Neoplastic cells largely depend on cAMP-dependent secondary messenger signaling for regulating the expression of GLUT1 and GLUT3 (Meneses et al., 2008). In addition, PI3K/Akt pathway has also been demonstrated to facilitate GLUT1 expression (Hoxhaj and Manning, 2020).

Regulation of GLUT expression

A survey of literature strongly indicates the involvement of multiple signaling messengers in the regulation of GLUT expression in cancer cells in which the prominent ones include Ca²⁺ (Diaz-Ruiz et al., 2011; Echeverría et al., 2021), CAMK (White et al., 2018), PI3K/Akt (Hoxhaj and Manning, 2020), mTOR (Shin and Koo, 2021), HIF1-α, p53 (Monde, 2018; Schwartzenberg-Bar-Yoseph et al., 2004), Matrix metalloproteinases (MMP) (Ito et al., 2002), RAS, PKC (Lee et al., 2015), NF-κB (Zha et al., 2015), AMPK (Zambrano et al., 2019), Phosphate and Tensin Homolog (PTEN) (Ancey et al., 2018), and Thioredoxin-interacting protein (TxNIP) (Ancey et al., 2018). Many of these critical signaling and regulatory molecules are integrated into a specific loop to regulate the expression of various GLUT isoforms. Zhao and Zhang (2016) have elegantly described critical regulatory and canonical pathways regulating GLUT expression: GLUT-PI3K, GLUT-mTOR, GLUT-HIFs, GLUT-RAS, GLUT-MMP, and GLUT-p53 (Zhao and Zhang, 2016). Upstream to these cell-associated signaling mediators regulating GLUT expression, several extrinsic factors are also key players in modulating GLUT expression. Among these, proinflammatory cytokines have been reported to play a crucial role in regulating the expression of GLUT1 and GLUT3. GLUT regulatory cytokines include TNFα (Shikhman et al., 2001; Straus, 2013), IL-1β (Phillips et al., 2005), TGF-β (Andrianifahanana et al., 2016), IFN-y (Freemerman et al., 2014), IGF1 (Actis Dato et al., 2021), and IL-13 (Wieman et al., 2007). These cytokines indirectly regulate GLUT expression via their ability to modulate one or more signaling pathways. For example, IL-1β regulates GLUT expression by modulating PKC, PI3K, p38, and cJUN activation. Likewise, proinflammatory cytokines like IL-13 and IFN-y modulate GLUT1 expression via NF-κB activation (Freemerman et al., 2014; Tan et al., 2018). In addition to cytokines, several regulatory RNAs are also reported to regulate GLUT expression. Non-coding RNAs, including several miRNAs, have been demonstrated to regulate GLUT expression (Kong et al., 2016; Hu et al., 2019). Long non-coding RNA, HOTAIR, has been shown to regulate GLUT by its ability to modulate NFkB (Obaid et al., 2021). Extrinsic factors like iron availability also affect GLUT expression (Potashnik et al., 1995).

Importantly several hormones also play a crucial role in regulating GLUT expression. Gender-specific hormones like estrogen (Medina et al., 2004; Nualart et al., 2009), testosterone (Wilson et al., 2013; Mitsuhashi et al., 2016), and progesterone (Medina et al., 2003; Medina et al., 2004; Frolova et al., 2009; Nualart et al., 2009) whereas non gender-specific hormones like thyroid hormone (Torrance et al., 1997; Brunetto et al., 2012; Ding et al., 2016), insulin (Brosius et al., 1992; Stuart et al., 2009; Maria et al., 2015), and pituitary hormones (growth hormone, gonadotropin releasing hormone (GnRH), luteinizing hormone (LH), follicle stimulating hormone (FSH), and somatotropins) (Kilgour et al., 1995; Harris et al., 2012; De Los Santos, 2013; Kim and Park, 2017; Nicholas et al., 2020) have been implicated in the regulation of GLUT expression. However, like cytokines, the modulatory effect of hormones in GLUT expression is mediated via the involvement of one or more aforementioned signaling mediators and pathways (Kilgour et al., 1995; Shikhman et al., 2001; Phillips et al., 2005; Freemerman et al., 2014; Santos et al., 2014; Maria et al., 2015; Ding et al., 2016; Takaguri et al., 2016; Kim and Park, 2017). Nevertheless, the blend of cytokines, hormones, and specific signaling mediators is also likely to manifest cancer cell-specific differences in the expression of various GLUTs. Nevertheless, physical activities like exercise also modulate GLUT expression via CAMK and AMPK-associated signaling pathways (Röckl et al., 2007; Richter and Hargreaves, 2013).

Further, the role of hypoxemic conditions and HIF1α in regulating GLUT expression needs a little more attention as HIF1α is the master regulator of tumor metabolism (Nagao et al., 2019). Experimental reports strongly indicate a correlation of HIF1α to the expression of various GLUT isoforms (Semenza, 2010; Mattmiller et al., 2011; Semenza, 2013; Sadlecki et al., 2014; Seleit et al., 2017; Kierans and Taylor, 2021). The modulation of GLUT expression by HIF1α is also mediated by the co-involvement of other signaling messengers like RAS (Moldogazieva et al., 2020; Ghanavat et al., 2021; Shin and Koo, 2021). Hypoxic conditions of the tumor microenvironment also regulate GLUT gene expression via modulation of HIF1α and its downstream signaling events (Roy et al., 2020; Li et al., 2021). Nevertheless, even glucose levels in the external milieu serve as essential triggers of GLUT1 and GLUT3 expression via HIF1α mediated regulation of GLUT gene expression (Hayashi et al., 2004; Yang et al., 2014; Zhao and Zhang, 2016; Seleit et al., 2017). Hypoxia and glucose levels have been reported to modulate mRNA levels of GLUT1, GLUT3, GLUT8, GLUT9, GLUT10, and GLUT12 (Frolova et al., 2009; Frolova and Moley, 2011; Kido et al., 2020), indicating the importance and central role of these extrinsic factors on pan-GLUT expression in which HIF1-α has a central role. Nevertheless, some evidence also indicates HIF1-α independent regulation of GLUT expression (Macheda et al., 2005). Such variations could be dependent on the etiology of various cancer cells. Figure 4 shows the primary regulators of GLUT expression and their downstream signaling mediators.

GLUT inhibitors of natural origin

Small molecule GLUT inhibitors mainly belong to natural, seminatural and synthetic chemical origins (Shriwas et al., 2020). have elegantly summarized the GLUT inhibitors of herbal origin. Vick et al. (1973) introduced the first herbal GLUT inhibitor named “Phlorizin,” which was followed by a series of discoveries of several other GLUT inhibitors of plant origin which belong to various chemical categories, including alkaloids, flavonoids, and other oxygen heterocyclic and phenolic compounds (Shriwas et al., 2020). Among these diverse categories of GLUT inhibitors of plant origin, those identified with antineoplastic action include more than 25 compounds, to name a few of them: apigenin (Gonzalez-Menendez et al., 2014), curcumin (Gunnink et al., 2016; Soni et al., 2021), genistein (Vera et al., 1996; Pérez et al., 2011), naringenin (Memariani et al., 2021), oridonin (Yao et al., 2017), phloretin (Wu et al., 2018; Wang et al., 2022), phlorizin (Vick et al., 1973; Kwon and Levine, 2007; Shriwas et al., 2020), quercetin (Schmidl et al., 2021a), resveratrol (Jung et al., 2013; Gwak et al., 2015; Zambrano et al., 2019; Samec et al., 2020), silybin (Zhan et al., 2011), and vinblastine (Shriwas et al., 2020). Although many of these GLUT inhibitors display a promising antineoplastic potential, the precise mechanism of the inhibitory action of most remains elusive, which needs to be deciphered for their optimal utilization in antineoplastic therapeutics.

Cytochalasin B

A cell-permeable mycotoxin named cytochalasin B, which is also a microfilament poison and used to regulate cytoskeletal elements is yet another GLUT inhibitor of natural origin that is extensively investigated for its mechanism of GLUT inhibition and utility for antineoplastic applications (Jung and Rampal, 1977; Devés and Krupka, 1978; Klip and Pâquet, 1990; Granchi et al., 2016; Kapoor et al., 2016). Kapoor et al. (2016) published an elegant article regarding the mechanism of the GLUT-1 inhibitory action of cytochalasin B along with two phenylalanine amides. This study also showed that the GLUT-1 inhibitory action of cytochalasin B could be manifested in mM concentration ranges. The study utilized crystallography and in silico docking tools to decipher the mechanism of GLUT1 inhibition and identified the amino acids and binding sites through which cytochalasin B exerted its inhibitory action. However, being a blocker of microfilament polymerization, its utility for antineoplastic application in vivo needs to be verified vis à vis validation of safety for other organs and tissues. Moreover, the cytoskeleton plays a crucial role in the functioning of many tumor-infiltrating cells like macrophages and dendritic cells, where microfilament poisoning by cytochalasin B needs to be controlled while utilizing it for anticancer therapeutic approaches.

Glucopiericidin A

Glucopiericidin A is yet another highly potent GLUT inhibitor of natural origin with an IC50 value of 22 nM compared to 500 nM for cytochalasin B (Kitagawa et al., 2010). The group of glucopiericidin A members are produced by streptomyces of the actinomycetes family. Glucopiericidin A has been remarkable for its low LD50 values and is predicted to have less toxicity (Zhou and Fenical, 2016). Though initially discovered as a filopodia protrusion inhibitor (Kitagawa et al., 2010), it was later discovered that glucopiericidin A could simultaneously inhibit both mitochondrial respiration (Hall et al., 1966) and glycolysis (Kitagawa et al., 2010) by its ability to inhibit GLUT leading to declined ATP production. Moreover, glucopiericidin A is also shown to interfere with tyrosine kinase-based signaling (Zhou and Fenical, 2016). It has been demonstrated that glucopiericidin A synergies with low glucose levels for inducing cell death in the pancreatic, lung, and other cancer cells (Palorini et al., 2013). However, the precise mechanism of GLUT inhibition by glucopiericidin A remains elusive. Moreover, its antineoplastic action needs to be assessed in a broader spectrum of targets, and its specificity on cancer cells remains undetermined. It remains elusive whether it can directly inhibit GLUT from interfering with glycolysis or via glucose phosphorylation. The GLUT inhibiting potential of glucopiericidin A solely depends on maiden experiments displaying its ability to inhibit the uptake of 2-deoxyglucose (2DG) (Imoto, 2019). Thus, the antineoplastic potential of glucopiericidin A needs further validation.

Polyphenols

Polyphenols are a group of natural compounds found in many plants with reported antineoplastic potential (Keating and Martel, 2018). Polyphenols like apigenin, silibin, kaempferol, gossypol, naringenin, phloretin, genistein, resveratrol, herpertin, quercetin, myricetin, and catechin are reported to inhibit glucose uptake in several neoplastic cell lines, by their ability to inhibit GLUTs at both gene and protein expression level (Keating and Martel, 2018; Ji et al., 2019; Kang et al., 2019). On the other hand polyphenols like quercetin are reported to competitively inhibit GLUT1 (Hamilton et al., 2018; Salehi et al., 2020), whereas for some other polyphenols the mechanisms of GLUT inhibition remain unclear. Johnston et al. (2005) and Farrell et al. (2013) reported that dietary polyphenols decrease glucose uptake in the Caco-2 cell line, possibly depending on their ability to inhibit GLUT and SGLT1. Moreover, polyphenols like curcumin, genistein, and quercetin can improve the antineoplastic action of glycolytic inhibitors in cell lines of myeloid leukemia origin (de Blas et al., 2016). Moreover, phloretin, a polyphenol derived from apple, was shown to inhibit the proliferation of colorectal cancer cell lines by inhibiting GLUT2 and activating the p53-dependent signaling pathway (Lin et al., 2016). In vitro experiments on Caco-2 cells revealed that exposure to anthocyanin-rich plant extract inhibited the expression of SGLT1 and GLUT2 (Alzaid et al., 2013). However, for many polyphenols, the precise mechanism of their inhibitory action on GLUTs remains to be clarified. Among the most worked out polyphenols, apigenin and resveratrol are known to inhibit GLUT1 (Melstrom et al., 2008; Gwak et al., 2015), whereas phloretin, and quercetin are reported to inhibit GLUT1, and GLUT2 (Schmidl et al., 2021a) on the other hand silybin inhibits GLUT4 (Zhan et al., 2011). These studies on polyphenols also corroborate that eating fruits and a vegetable-rich diet abundant in polyphenols decreases cancer risk. It can be attributed mainly to the GLUT inhibitory potential of several dietary polyphenols. In addition, polyphenols manifest their anticancer action through other mechanisms, including modulation of cell signaling, cell survival, angiogenesis, immunopotentiation, hormonal regulation, and enzyme modulation (Niedzwiecki et al., 2016). The target cancers investigated for the anticancer action of polyphenols include prostate, colon, breast, lung, bladder, pancreas, spleen, and leukemia (Niedzwiecki et al., 2016; Montané et al., 2020).

Small inhibitory molecules

The main group of small inhibitory molecules of various GLUTs are principally aromatic compounds with unique imidazole, pyridine and pyrazole rings and most of them have a common overall structural organization with minor differences as depicted in Figures 5A–F.

WZB 117

WZB 117 (Figure 5A) is a polyphenol-derived small molecule inhibitor of GLUT1 (Liu et al., 2012; Ojelabi et al., 2016; Chen et al., 2017). It exerts antineoplastic action with an IC50 value in the 10 µM concentration range on Hela, RKO, A549, and MCF7 cells of the cervix, endometrial carcinoma, lungs, and breast origin, respectively (Liu et al., 2012), accompanied by enhanced radio-sensitization (Zhao et al., 2016). Some mechanisms of the antineoplastic action of WZB117 have also been worked out, showing the implication of a declined GLUT1 expression along with a decrease in ATP levels and glycolytic enzymes (Liu et al., 2012). Moreover, WZB117 triggers an increase in the ATP sensing AMPK and a decline in cyclin E2 (Liu et al., 2012). The antineoplastic activity of WZB117 has also been evaluated in vivo (Liu et al., 2012). Interestingly WZB117 is also reported to inhibit cancer stem cells (Shibuya et al., 2014). However, one of the disadvantages of WZB117 is its instability owing to its ester bonds under aqueous physiological conditions, potentially the most significant limiting factor in its use in cancer therapeutic strategies. Moreover, it is described as inhibiting only GLUT1, which can be another limitation of its use for antineoplastic applications.

DRB18

In a recent study, another promising GLUT inhibitor named DRB18 (Figure 5B), which belongs to a synthetic small inhibitory molecule, has been reported with the ability for GLUT inhibition as evaluated on A549 tumor-bearing nude mice and HER293 cell lines (Shriwas et al., 2021). One of the reported advantages of DRB18 is its relative stability compared to WZB117 (Roberts et al., 2020). So far, DRB18 has been shown to exert cytostatic action on around 60 cancer cell lines of diverse origins. The primary mechanism of the antineoplastic action of DRB18 was shown to be via increased endoplasmic reticulum oxidative stress (Shriwas et al., 2021). The pan-GLUT inhibitory ability of DRB18 was verified through in silico-based studies. The cytotoxic ability of DRB18 accompanied by inhibitory action on glucose uptake showed IC50 in the concentration range of 900 nM-9.0 µM (Shriwas et al., 2021). Moreover, DRB18 treatment caused a remarkable inhibition of glucose-dependent metabolic pathways like glycolysis, TCA cycle, purine, and pyrimidine synthesis (Shriwas et al., 2021). However, the GLUT inhibitory and antineoplastic action of DRB18 has not been reported for any hematological malignancies. The wide range of low to high IC50 values indicates the need for testing its efficacy and safety for antineoplastic applications against a wider variety of preclinical cancer models. It is also unclear if DRB18 can modulate the cytotoxic action of other antineoplastic agents. Safety issues on normal tissues and organs of the tumor-bearing host also remain largely unverified. Moreover, DRB18 is water insoluble, which could be another limitation of its use in anticancer strategies.

BAY-876

A quinoline derivative named BAY-876 (Figure 5C) was initially identified for its highly selective GLUT1 inhibitory potential by (Siebeneicher et al., 2016b), which also exerts its actions in nM ranges but has limited ability to inhibit other GLUT isoforms like GLUT2, GLUT3, and GLUT4. Additional advantages of BAY-876 include its aqueous stability in vitro and adequate bioavailability upon administration through the oral route (Siebeneicher et al., 2016b). BAY-876 manifests antineoplastic action mainly via its ability to inhibit GLUT1 leading to multiple actions on neoplastic cells, including suppression of proliferation, drug-resistance, EMT, and metabolism (Yang et al., 2021). BAY-876 was reported to manifest increased NADP⁺/NADPH simulating glucose starvation conditions (Chen et al., 2022). However, the antineoplastic activity of BAY-876 needs to be validated on a broader range of neoplastic cells. However, docking studies have been performed on BAY-876 for binding to GLUT1 (Ojelabi, 2017), and its exclusive binding to only GLUT1 limits its therapeutic applications. Moreover, the safety of its in vivo applications needs further verification.

Fasentin

Fasentin N-(4-chloro-3-(trifluoromethyl)phenyl)-3-oxobutanamide (Figure 5D) is a small molecule inhibitor of GLUT that was initially reported as a sensitizer to induction of cell death via FAS and TNF (Kraus et al., 2018; Ocaña et al., 2020) later it was reported that Fasentin leads to glucose deprivation by its ability to bind to GLUT1 and GLUT4 as revealed by docking studies (Wood et al., 2008). However, despite the promising GLUT1 inhibitory potential alongside sensitization of cells to death stimuli, not much has been worked out concerning its antineoplastic potential. A recent study reported that fasentin could inhibit the proliferation of endothelial cells and hence may have an implication in antiangiogenic therapeutic strategies. However, its effect on other normal healthy cells must be explored to predict safety issues (Ocaña et al., 2020). Moreover, the effect of fasentin is apparent in about 100 µM range (Ocaña et al., 2020) which is on the higher side than that desired for an effective and safe therapeutic agent.

STF-31

Another promising GLUT1 inhibitory agent with antineoplastic potential is STF-31 (Figure 5E) (Chan et al., 2011; Adams et al., 2014; Xintaropoulou et al., 2015; Kraus et al., 2018). It was shown initially to manifest renal cell carcinoma (RCC) specific cytotoxic action (Chan et al., 2011). Additionally, STF-31 has been shown to modulate HIF and target NAMPT to manifest its antineoplastic action (Adams et al., 2014; Xintaropoulou et al., 2015). Moreover, Kraus et al. (2018) reported that the anticancer action of STF-31 is apparent only at higher concentrations, and hence there is a need for further optimization and testing on a broad range of neoplastic target cells both in vitro and in vivo to explore its dual inhibitory action on GLUT1 and NAMPT. However, the utility of the same will depend on the expression levels of GLUT1 and NAMPT, which may vary in a tumor-to-tumor manner. Moreover, it is also reported that if cancer cells express GLUT2, they can overcome the inhibitory action of STF-31 and that it does not harm normal cells (Chan et al., 2011).

Glutor, a Pan-GLUT inhibitor

Recently a pan-GLUT inhibitor named glutor has been demonstrated to exhibit a potent antineoplastic action in the nanomolar concentration range against cancer cells of diverse origins (Reckzeh and Waldmann, 2020a; Reckzeh and Waldmann, 2020b; Temre et al., 2022). Glutor is (S)-6-Methyl-5-(4-morpholinobenzyl)-4-oxo-2-phenyl-N-(pyridin-3-ylmethyl)-4,5,6,7-tetrahydropyrazolo [1,5-a] pyrazine-6-carboxamide, with the empirical formula C₃H₃₂N₆O₃, and molecular weight of 536.62 (Figure 5F) (Reckzeh et al., 2019).

Glutor was identified to inhibit GLUT1, GLUT2 and GLUT3 in HCT116 cells (Reckzeh et al., 2019). The inhibition of the uptake of 2DG was determined to have an IC50 value of 11 nM in most sensitive cell lines, including urinary cancer-derived cell lines, which are most glucose-addicted (Lea et al., 2015). Interestingly glutor was found to impart no cytotoxicity on normal healthy peripheral blood mononuclear cells and Institute for medical research-90 (IMR-90) embryonic lung cells (Reckzeh et al., 2019). Further, the elegant study of Reckzeh et al. (2019) also screened the antineoplastic action of glutor on a battery of about 94 cancer cell lines of diverse origins, and it was observed that the antineoplastic action was exerted with IC50 values of less than 100 nM. Moreover, the inhibitory action of glutor was also evaluated in multicellular spheroids of HCT-116 cells (Reckzeh et al., 2019). Interestingly glutor treatment minimized glucose deprivation in a dose-dependent manner. Hence, glutor displays the potential for effectively inhibiting glucose uptake in solid tumors. However, it was also reported that some cell lines like BxPC-3 exhibit resistance to glutor. The sensitivity or resistance to glutor depends on the cancer cells’ metabolic background (Reckzeh et al., 2019). The glutor resistance cell lines can stitch between glycolytic and OXPHOS phenotypes (Reckzeh et al., 2019). Depending on the metabolic make-up of the target cells and selected dose, glutor could usher a complete glucose deprivation condition, even leading to upregulation of GLUT1 and GLUT3 expression due to hypoglycemia. Moreover, it is also likely that by its ability to cause inhibition of GLUT1, GLUT2, and GLUT3, glutor possesses an extraordinary capacity to overcome the rescue and compensatory mechanisms of highly adaptive neoplastic cells. It is also noteworthy that the study of Reckzeh et al. (2019) also indicated that neoplastic cells’ cytostasis and cytotoxicity could be synergistically upregulated by using glutor along with glutaminase inhibitor CB-839.

Thus, the results obtained based on screening of glutor for its antineoplastic action on 94 cancer cell lines derived from highly malignant cancers revealed the extraordinary ability of glutor to inhibit glucose uptake, glycolytic flux, and cell survival (Pliszka and Szablewski, 2021).

However, despite the evaluation and certification of the antineoplastic action of glutor on cancer cell lines of diverse origins, the same has not been evaluated in any in vivo tumor model, which will be necessary to assess the translational value of glutor’s antineoplastic potential for therapeutic applications. Moreover, the issues of bioavailability and ideal administration routes also need to be assessed in an appropriate in vivo model.

Glutor treatment of tumor cells inhibited cell survival and increased cell death. It also caused a decrease in glucose uptake associated with altered expression of GLUT1 and GLUT3. HIF-1α, HK-2, LDH-A, and MCT1 also decreased with diminished lactate production and deregulated pH homeostasis. Moreover, the expression of cell survival regulatory molecules p53, Hsp70, IL-2 receptor CD25, and C-myc was modulated upon treatment with glutor. Additionally, glutor treatment triggered mitochondrial membrane depolarization, accompanied by high ROS generation and the modulated ratio of Bcl-2/BAX. Chemosensitivity of tumor cells increased following exposure to glutor and decreased MDR1 expression (Temre et al., 2022). Hence, these observations indicate that glutor has the potential to be used in antineoplastic therapeutic applications.

Other inhibitors

Other crucial small molecule inhibitors of GLUT include GLUTi1 and GLUTi2 (Kapoor et al., 2016), compound 3 (Siebeneicher et al., 2016a), compound 15b (Liu et al., 2020), PUG1 (Ung et al., 2016), KL-11743 (Olszewski et al., 2022), and NV5440 (S. A. Kang et al., 2019) which belong to diverse families of chemicals and exert limited GLUT isoform inhibitory potential and have high IC50 values, so the possibilities of in vivo toxicity remains a concerning issue. Further, the neoplastic cell inhibitory potential in most of these remains limited to only a few targets investigated. Tuccinardi et al. reported the GLUT1 inhibitory potential of oxime-based inhibitors with compatibility to inhibiting the growth of the H1299 lung cancer cell line (Tuccinardi et al., 2013). However, not much has been reported further on the anticancer potential of this category of inhibitors except for their utility in developing novel GLUT inhibitory agents.

Combinatorial approaches

Studies have indicated the promising potential of a combinatorial approach for GLUT inhibition instead of a single inhibitor approach (Tilekar et al., 2020). Such approaches mainly involve a combination of agents with 1) the ability to directly bind to GLUT isoforms and cause their inhibition; 2) approaches to inhibit expression of GLUT, and 3) use of agents which compete with glucose for binding to GLUTs and hence cause inhibited glucose transport. Further to achieve success in such approaches, GLUT inhibitors with pan-GLUT inhibitory potential can be of great use to usher the inhibition of multiple GLUT isoforms. Moreover, a combination of multiple GLUT inhibitors of natural and synthetic origin can also be considered and has shown hope during in vitro experiments on cancer cell lines of various origins ((Kast et al., 2016; Sawayama et al., 2019; Tilekar et al., 2020). However, adding several inhibitors together may cause toxicity and can likely lead to certain harmful effects on normal healthy cells, which also use GLUT isoforms for glucose uptake. Further, most of such combinations still lack the much-required in vivo testing for evaluating efficacy and safety parameters. Moreover, WZB117 (Chen et al., 2017), 2DG (Lee et al., 2016), and BAY876 (Mori et al., 2019) have also been sporadically used in combination with chemotherapeutic agents under in vivo experimentation. Moreover, several studies have been conducted to evaluate the outcome of combining various GLUT inhibitors for achieving optimum inhibition of glucose uptake (Tilekar et al., 2020). The most promising of such combinations with successful antineoplastic outcomes include combination of GLUT inhibitors like cytochalasin B, WZB117 and its derivatives, BAY-876, silibinin, glutor, STF-31 and phloretin with other conventional anticancer drugs like doxorubicin, curcumin, etoposide, cisplatin, 5-fluorouracil, vincristine, cytarabine, oxaliplatin, paclitaxel and antimycin A (Liu et al., 2012; Tsakalozou et al., 2012; Li et al., 2019; Trendowski, 2015; Trendowski, 2015; Kapoor et al., 2016; Chen et al., 2017; Reckzeh et al., 2019; Zambrano et al., 2019; Barbosa and Martel, 2020; Joly et al., 2020; Tilekar et al., 2020; Wu et al., 2020; Shriwas et al., 2021; Olszewski et al., 2022; Temre et al., 2022; Weng et al., 2022) to name a few such representative anticancer drugs in cancers of diverse origins. Nevertheless, inhibitors of other molecules and/or pathways have also been combined with GLUT inhibitors to achieve fruitful antineoplastic effects (Tsakalozou et al., 2012; Tilekar et al., 2020; Weng et al., 2022). For example, the combination of glutor with glutaminase inhibitor CB-839 resulted in successful antineoplastic manifestations (Reckzeh et al., 2019). Additionally, combination with GLUT inhibitors with radiotherapeutic strategies had also displayed improved antineoplastic potential (Zhao et al., 2016). Moreover, GLUT inhibitors, combined with other antiglycolytic agents, also result in superior inhibition of tumor cell metabolism, compared to such action of these agents when applied alone (Raez et al., 2013; Barbosa and Martel, 2020; Tilekar et al., 2020). These studies are also limited in their long-lasting impact in designing novel therapeutics due to the lack of extensive in vivo-based investigations to assess safety and toxicity issues on normal healthy cells and tissues. However, despite the limitations above, combinatorial approaches hold promise for formidable antineoplastic therapeutic applications to overcome drug- and radio-resistance and synergize with the anticancer action of individual agents. Importantly, a combination of GLUT inhibitors with chemotherapeutic agents will also aid in reducing the cytotoxic dose of chemotherapeutic agents and hence minimize their side effects, which need experimental validation. Moreover, combining GLUT inhibitors with other antineoplastic therapeutic agents to manifest a robust and multifaceted inhibition of tumor metabolism could also be a promising antineoplastic strategy. However, such possible combinations are yet to be fully explored. The combinatorial approach of multiagent strategies must be carefully explored to optimize anticancer action while minimizing the side effect. These promising results are also being explored for translational therapeutic potential in clinical trials.

GLUT antibodies

For a long, attempts have been made to make the best utilization of anti-GLUT antibodies to block glucose transport by GLUT. Anti-GLUT1 antibodies have been demonstrated to inhibit glucose uptake in resealed erythrocyte membrane ghosts (Afzal et al., 2004). Nevertheless, a study using Cal27 cells showed that anti-GLUT1 antibodies could inhibit proliferation, induce apoptosis, and chemosensitization for cisplatin (Wang et al., 2013). A study by Rastogi et al. (2007) reported that anti-GLUT1 antibodies could arrest cell growth accompanied by induction of apoptosis in cell lines derived from breast and non-small cell lung cancer (NSCLC). Additionally, anti-GLUT1 antibodies also chemosensitized MCF7 cells to cisplatin, paclitaxel, and gefitinib (Rastogi et al., 2007). Further, anti-GLUT1 antibodies have also been explored for the targeted delivery of chemotherapeutic agents (Barbosa and Martel, 2020). However, a literature search reveals that the spectrum of cancers against which anti GLUT1 antibodies have been tested is limited to realizing their therapeutic potential.

miRNAs in anti-GLUT strategies

There have been attempts to use the potential of miRNAs to target various aspects of tumor metabolism, including GLUTs (Chen et al., 2012; Pedroza-Torres et al., 2019; Shiah et al., 2021). miRNAs like miR-133 and miR-195–5P have modulated gene transcription of various GLUT isoforms (Macheda et al., 2005; Krüger et al., 2008). In this quest, miRNA to GLUT1, GLUT3, and GLUT4 has shown promising antineoplastic potential in prevalent cancers like lung, prostate, breast, colorectal carcinoma, bladder cancer, and pancreatic adenocarcinoma (Lu et al., 2010; Chen et al., 2012; Fei et al., 2012; Chen et al., 2015; Zhao et al., 2017; Pedroza-Torres et al., 2019; Azizi et al., 2021; Shiah et al., 2021). Moreover, a direct inverse correlation has been reported between miRNA-144 and GLUT1 in breast and non-small cell lung cancer (NSCLC) (Azizi et al., 2021). Similarly, the role of miR-22, miR1291, and miR-195–5P have been associated with GLUT1, and GLUT3 in breast, RCC, and T24 cells, respectively (Fei et al., 2012; Yamasaki et al., 2013; Chen et al., 2015; Alamoudi et al., 2018). Given such intimacy between various miRNAs and GLUT isoforms, there is a strong logic to utilize them in therapeutic applications. Similarly, a study on the role of MiR-218 and MiR340 in the expression of GLUT1 is demonstrated in oral squamous cell carcinoma (OSCC) (Xu et al., 2018; Wang et al., 2020; Shiah et al., 2021). Combinatorial use of miRNAs with other conventional GLUT1 inhibitors is also needed to be explored to circumvent GLUT at both gene expression and protein function levels.

Short hairpin RNA

Short hairpin RNA (shRNA) to GLUT1 has shown a promising inhibitory effect on M.D. Anderson - Metastatic Breast 231 (MDA-MB-231) and Homo sapiens-578 tumor cells (HS578T) triple-negative breast cancer (TNBC) cell lines (Oh et al., 2017; Shriwas et al., 2018; Wu et al., 2020). Similarly, the use of shRNA to GLUT1 was observed to show promising outcomes in colon cancer cells (Bai et al., 2019). Wang et al. (2020) demonstrated that shRNA inhibits the survival of laryngeal carcinoma HEP2 cells via beclin-1-associated autophagy. shRNA to GLUT1 is also reported to effectively block glucose uptake in various cancer cell lines (Pliszka and Szablewski, 2021). Thus, the shRNA approach is also promising for effective combat against GLUT expression and function in neoplastic cells by silencing gene expression of various isoforms of glucose transporters.

Antisense cDNA

Approaches to transfecting neoplastic cells with GLUT1 antisense cDNA have been demonstrated to usher in a decline in the gene expression of GLUT1 and an increase in radiosensitivity of neoplastic cells (Chan et al., 1999; Ito et al., 2000; Yan et al., 2013; Pliszka and Szablewski, 2021). The common cancers used to test this approach include leukemia, gastric, rhabdosarcoma, breast, and glioblastoma (Ito et al., 2002; Chan et al., 2004; Yang et al., 2019; Pliszka and Szablewski, 2021). However, not much breakthrough is yet reported using this approach, and its clinical applications need to be evaluated. The approaches mentioned above of ‘nipping in the bud’ of GLUT expression level have still not come into vogue for preclinical and clinical trials as more basic data needs to be collected before embarking further. However, the most popular and explored approaches are still to identify the best pan-GLUT inhibitor, as other approaches to gene intervention can have potentially harmful outcomes. In this quest, a newly recognized pan-GLUT inhibitor named glutor is emerging with the hope of effective GLUT1 and GLUT3 inhibition.