The Warburg effect was initially thought to be a way for cancers to compensate their energy production because the mitochondria were damaged or dysfunctional. Numerous research has challenged this notion, revealing a far more complex scenario with mitochondrial defects not always being at the forefront of this phenotype [4, 5, 6]. Some studies have shown mutations in mitochondrial enzymes or mitochondrial DNA as possible contributors to the Warburg effect, however, dysregulation of signaling pathways, activation of certain oncogenes, mutations in tumor suppressors and the tumor microenvironment (TME) appear to be the major contributors on the metabolic reprogramming of cancer [7, 8, 9, 10]. As these intricate mechanisms continue to be explored, it is acknowledged that the observed metabolic phenotype differences vary not only across cancer types but by the degree of malignancy, with more aggressive and invasive cancers generally exhibiting elevated glycolytic fluxes [11, 12, 13, 14]. This metabolic adaptation that favors glycolysis, leads to the production of various glucose byproducts linked to different metabolic pathways, such as the pentose phosphate pathway (PPP), and the synthesis of essential biomolecules including nucleotides, amino acids, and lipids. Similar adaptations can be observed in stem cells or cells undergoing development, because it makes nutrient acquisition more efficient for new cell production [15]. It is proposed that the TME, a network composed of the tumor cells, immune cells, blood vessels, the extracellular matrix, and other, non-tumor cell types, plays a role in selecting and sustaining this altered metabolism. Lactate generated from aerobic glycolysis creates a highly acidic TME that supports metabolic flexibility and genomic instability [16, 17]. A phenomenon known as the “reverse-Warburg effect” discerns the transfer of lactate (and other metabolites) between stromal cells and cancer cells, further contributing to cancer cells exploiting the TME for metabolic adaption [18, 19, 20]. Moreover, as tumors outgrow their blood supply’s capacity, they favor oxygen-independent energy production, stabilizing the transcription factor Hypoxia-Inducible Factor 1-alpha (HIF-1$\alpha$), which orchestrates a response that includes increased expression of glycolytic enzymes and glucose transporters [21, 22]. This distinctive adaptation to oxygen-independent energy production is often accompanied by the upregulation of vascular endothelial growth factor (VEGF), a key player in promoting angiogenesis, the physiological process through which new blood vessels develop from pre-existing vessels. In hypoxic tumor environments, the activation HIF-1$\alpha$ triggers increased VEGF expression, stimulating the formation of new blood vessels [23, 24]. This angiogenic response is crucial for sustaining the nutritional and oxygen requirements of rapidly proliferating tumor cells. Tumor cells that rely on glycolysis are capable of surviving in a consistent hypoxic environment. Hypoxia further induces matrix remodeling, influencing the metabolic profile of TNBC cells [14]. The extent to which breast cancer cells exhibit the Warburg effect and utilize OXPHOS can indeed vary significantly among distinct subtypes of breast cancer. Specific types of breast cancer, including invasive status, are presented in Table 1 (Ref. [25, 26, 27, 28, 29, 30]), along with their corresponding trends in glycolysis and OXPHOS status.

GLUT proteins, encoded by solute carrier 2 (SLC2) genes, belong to the Major Facilitator Superfamily (MFS) of membrane transport proteins and serve as the primary glucose transporters in mammalian cells. In humans, there are fourteen GLUT proteins (GLUTs 1-14). Notably, GLUT 1-4 have been implicated in various malignant tumors, including TNBC [32]. A comprehensive bioinformatics study using public databases such as ONCOMINE, Kaplan-Meier Plotter, and cBioPortal was carried out on a cohort of breast cancer patients to explore the intricate connection between GLUT1, GLUT3, GLUT4, and clinical prognostic factors. The study concluded that GLUT1 is overexpressed in TNBC patients and therefore has potential as a novel prognostic biomarker [33]. Another study indicated that the combined mRNA expression of SLC2A1 (GLUT1) and tumor suppressor, Retinoblastoma protein (RB1), may be significantly correlated, suggesting that tissues with high levels of RB1 mRNA may also exhibit relatively higher levels of SLC2A1 mRNA in TNBC [34]. In vitro studies have demonstrated promising results with specific inhibitors of GLUT proteins in TNBC cell lines. These include the synthetic non-metabolizable glucose analog, 2-deoxy-D-glucose (2-DG), anti-GLUT1 monoclonal antibody, and phytochemicals like resveratrol. These inhibitors have been associated with decreased glucose uptake, increased apoptosis, and enhanced radiation effectiveness [35, 36, 37, 38]. Oncogenes such as Ras and MYC can promote glucose uptake by upregulating HIF-1$\alpha$ and induce GLUT expression. The signaling pathway involved in cell proliferation and survival, phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt)/mammalian target of rapamycin (mTOR) or (PI3K/AKT/mTOR), is often dysregulated in breast cancer and is associated with endocrine therapy resistance [39]. In addition, the PI3K/AKT/mTOR pathway significantly influences the upregulation of glycolytic enzymes and GLUTs [40]. The tumor suppressor p53 is mutated in 20–30% of advanced cancers, including TNBC, and inhibits the expression GLUT1, GLUT3, and GLUT4 [41].

When glucose enters the cell to undergo glycolysis, the first step of glycolysis regulates the intracellular glucose transport through glucose phosphorylation by hexokinases (HK1-4). HK1 is the predominant isoform found in many adult differentiated tissues, meanwhile HK2 is more expressed during development [42, 43]. In vitro and in vivo studies have shown that, compared to HK1, HK2 is highly expressed in cancers, particularly in highly glycolytic cancers, as HK2 can partially enhance glucose flux (Fig. 2) [42, 43, 44, 45]. Elevated HK2 mRNA levels in TNBC cells, relative to non-TNBCs, suggest intensified glycolytic activity and extracellular acidification [46, 47]. A recent in vivo study by Blaha and colleagues [46] uncovered a kinase-independent function of HK2 that contributes to metastasis. HK2 knockdown in 4T1 cells, which are a murine TNBC equivalent cell line, affected the expression of Epithelial-Mesenchymal Transition (EMT) genes and metastasis by reducing SNAIL protein stability, and nuclear localization via Glycogen Synthase Kinase (GSK3$\alpha$/$\beta$), a key enzyme that regulates glycogen metabolism and plays a role in metastasis. GSK3-resistant SNAIL variants restored metastasis in HK2-deficient 4T1 cells, offering valuable insights for metastasis research and metabolism [46]. Both HK1 and HK2 are unique in that they are able to bind the mitochondria, amplifying the role of these enzymes as metabolic sensors by their physical interaction with mitochondria at the outer mitochondrial membrane (OMM) at the voltage-dependent anion channel (VDAC) [42, 48, 49]. VDAC is critical for cell metabolism because it is an entry point for a variety of metabolites such as pyruvate and NADH and can transport calcium ions [43, 48, 49, 50, 51, 52]. By binding VDAC, HK1 and HK2 are able to scavenge mitochondrial ATP and essentially couple glycolysis with OXPHOS, creating a positive feedback loop to maintain the Warburg phenotype [43, 48]. HK2 localizing at the mitochondria enables multifaceted functionalities distinct from its canonical enzymatic activity including autophagy modulation and the inhibiting of apoptosis (Fig. 2) [53, 54]. The release of HK2 from VDAC can trigger a robust apoptotic signaling cascade. This release is thought to be initiated when GSK3$\beta$, a kinase regulated by Akt, phosphorylates a putative HK docking site on VDAC, however, the mechanism by which this occurs is still not completely understood and remains a topic of debate [55, 56]. Ciscato and colleagues [54] expanded on the knowledge of HK2 binding at the mitochondria in varying cancer models, including TNBC, by showing that HK2 localizes at the Mitochondria-Associated Membranes (MAMs), a contact site between the OMM of the mitochondria and the endoplasmic reticulum in vivo and in vitro. MAMs tightly regulates numerous cellular processes ranging from lipid homeostasis to calcium homeostasis by the interaction of protein inositol trisphosphate receptors (IP3Rs) at the ER and VDAC at the OMM which are tethered by the chaperone glucose-regulated protein 75 (GRP75) [57]. The group also observed that HK2 dissociating from the MAMs triggers an influx of calcium in the mitochondria, causing an overload and ultimately cell death (Fig. 2).

There are key enzymatic reactions in glycolysis and one of those key steps is when the glycolytic shunt enzyme 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 4 (PFKFB4) catalyzes and modulates the levels of fructose-2,6-bisphosphate (F26BP). In turn, this affects the activity of the main rate-limiting step in glycolysis: the enzyme phosphofructokinase-1 (PFK-1) [58]. Several studies have suggested that PFKFB4 is upregulated in various cancers, including TNBC, as part of their metabolic adaptation. Elevated PFKFB4 expression can result in increased levels of F26BP which, thereby, enhances the activity of PFK-1. An in vitro study by Dai and colleagues [58] revealed a non-canonical role of PFKFB4 of nuclear translocation in response to hypoxia, activating HIF-1$\alpha$ and promoting metabolic plasticity in TNBC. Importantly, increased PFKFB4 expression correlated TNBC patients with poorer outcomes [58]. Within the PFKFB4 family, PFKFB3 has emerged as a pivotal figure in tumor glycolysis, carrying substantial implications for both tumor growth and metastasis. Its function is intricately controlled by crucial factors such as HIF-1$\alpha$, AKT and phosphatase and tensin homolog (PTEN) [59]. The expression of PFKFB3 has been shown to be highly correlated with VEGF expression in breast cancer, thereby contributing to angiogenesis and distant metastasis. TNBC patients have significantly higher expression levels of VEGF in tumors compared to non-TNBC patients [60].

In anaerobic glycolysis, lactate dehydrogenase (LDH) catalyzes the reversible final step, converting pyruvate into lactate. Accumulation of the metabolic byproduct lactate and the consequent acidification of the extracellular environment is frequently observed in numerous solid tumors [61]. Monocarboxylate transporters (MCTs) facilitate lactate’s transport across cell membranes (Fig. 2). In particular, TNBC exhibits an increased expression of MCT isoforms, notably MCT1 and MCT4, which are crucial for mediating lactate expulsion and the acidification of the TME. Prior in vitro and in vivo studies have demonstrated that increased lactate production within cancer cells inhibits the immune surveillance of tumors by T-cells and natural killer (NK)- lymphocytes, consequently facilitating tumor proliferation [62]. The prognostic relevance of plasma LDH levels has been investigated in several breast cancer studies which concluded that there is a correlation between high serum LDH levels and poor overall survival (OS) and progression-free survival (PFS) in breast cancer patients [61, 63]. Therefore, the notably elevated expression of LDH-A (monomer of the tetramer LDH) in breast cancer tissues could serve as an indicator of the degree of malignancy.

In vitro work in the TNBC cells, MDA-MB-231, demonstrated that IL-6 increases the rate of glucose intake and lactate generation by activating STAT3. Furthermore, overexpression of constitutively activate STAT3, was observed in TNBC and HER2+ patient tissues [71, 72]. STAT3 is crucial in advanced cancers because its activation can promote glycolysis by activating transcription of HIF-1$\alpha$ as well as transcriptionally activating HK2, in vitro [73, 74] A comprehensive study utilizing in vivo and in vitro approaches in addition to patient samples, demonstrated that STAT3 mediated immune cell (NK cells) metabolism to resemble that of tumor cells, resembling the Warburg phenotype. This metabolic reprogramming allowed the NK cells to function more optimally against the tumor cells in the TME [75]. Interestingly, numerous STATs (STAT1, STAT2, STAT3, STAT5, and STAT6) have been shown to localize in the mitochondria in a variety of different cell types to regulate mitochondrial function. For instance, in vitro studies have shown how post-translational modifications, such as serine/tyrosine phosphorylation and acetylation of STAT3 enable translocation to the mitochondria [76, 77]. One study in the mouse TNBC equivalent 4T1 cells showed that serine phosphorylated mitochondrial STAT3 can decrease apoptosis and promote tumor cell growth by decreasing the amount of reactive oxygen species (ROS) that is released from complex I [78]. Extending on these findings, another group used TNBC cells and reported that when STAT3 is reduced, it lowers the activity of complex I dehydrogenase, which impairs levels of NAD+. This functions as a reverse signal, linking mitochondrial metabolism to changes in nuclear gene expression, inducing antioxidant genes that help maintain redox balance and malignant cell proliferation, survival, and migration [79]. The mechanisms triggering mitochondrial translocation to engage this non-canonical role of STAT3 have to be further investigated to provide clearer insights on how canonical and non-canonical function can be utilized effectively in immunotherapy cancer therapeutics [80].

There are several miRNAs involved in breast cancer, however, here, the focus will be on miRNAs implicated in TNBC metabolism. In vitro and in vivo studies have identified miRNAs that exhibit significant correlations with the Warburg effect in TNBC including miR-210-3p, miR-181a, miR-767-5p, and miR-155 [87, 88, 89]. Some of these miRNAs are associated with glycolytic metabolism, resulting in dysregulation of glucose consumption and lactate production by modulating glucose transporters and metabolic enzymes including HK2, PKM2, and LDH-A [90]. Within this context, the miR-200 group stabilizes phosphoglucose isomerase (PGI), which contributes to cell invasion and metastasis, in vitro and miR-210-3p has a pronounced effect on promoting aerobic glycolysis in TNBC. Mechanistically, miR-210-3p was observed to target GPD1L, a glycerol-3-phosphate dehydrogenase 1-like enzyme, thereby preserving the stability of HIF-1$\alpha$, and simultaneously repressing the activity of the tumor suppressor p53, in vitro, specifically on the TNBC cells, MDA-MB-231 and HS578 [88, 90].

Evidence demonstrates that miR-155 functions as an oncogenic miRNA (oncomiR) in human breast cancers, playing a crucial role in tumor development. MiR-155 controls the expression of a number of genes involved in differentiation, inflammation, apoptosis, and transcriptional regulation. Experimentally and through bioinformatics, several miR-155 targets (~149) have been identified. These targets include the tumor suppressor PTEN, Forkhead-box protein O3a (FOXO3a), Cyclin-dependent kinase inhibitor p27 (p27), and Casein Kinase-1$\alpha$ (CK1$\alpha$) among others. In vitro and in vivo studies have demonstrated that miR-155 promotes glycolysis in TNBC cells and enhances glucose uptake in xenograft tumors by suppressing miR-143 and activating STAT3 [91]. Additionally, in a study of 50 TNBC patient specimens and PET images, a significant positive correlation was observed between miR-155 expression and glucose utilization in breast tumors. This association was mediated through miR-155 inhibiting the upstream inhibitor of c-MYC, FOXO3a, thereby initiating activation of c-MYC (PIK3R1-PDK/AKT-FOXO3a-cMYC pathway) and highlighting the pivotal role of miR-155 in regulating glucose metabolism in breast cancer [90]. Although miR-155 has been considered as a potential noninvasive molecular marker for breast cancer screening, it is important to clarify that further research is needed to fully establish its effectiveness [92].

While miRNAs primarily function in the cytoplasm, some studies have suggested the presence of a subset of microRNAs within the mitochondria. These miRNAs, called “mitomiRs” can regulate mitochondrial homeostasis, mtDNA maintenance, and Ca${}^{2+}$ balance and therefore play a major role in cancer metabolic reprogramming [93]. The mitomiRs that have been identified in TNBC include miR-29a, miR-140, let-7a, miR-128, miR-30a as glucose metabolism regulators and miR-4485 as an OXPHOS metabolism regulator [94]. The field of mitochondrial miRNA research is dynamically evolving, underscoring ongoing efforts to unravel precise functions and mechanisms.

A study that characterized the metabolic profile of breast cancer cells, in vitro, revealed that among their panel of cells, TNBC cells presented elevated glycolysis and reduced OXPHOS rates compared to other types of receptor positive breast cancer cells [103]. Another study supplemented these findings with tumor tissues gene expression data, reporting that OXPHOS complex subunit genes were downregulated in TNBC tumor tissue samples compared with non-TNBC tissues [47]. A byproduct of OXPHOS is ROS and interestingly, TNBC cells display higher levels of ROS compared with other subtypes of breast cancer [104]. Excess mitochondrial ROS can trigger apoptotic signals in the cell, but a moderate amount of ROS can be used as a potent signaling molecule in order to adapt to the stress of the TME, especially when a cancer becomes more advanced [50, 98, 101, 102, 105, 106]. For instance, the transcription factor HIF-1$\alpha$ utilizes ROS as a signaling molecule to maintain the transcription of genes favoring metabolic plasticity and angiogenesis [96, 101, 102, 107]. One of the numerous ways that cancer cells maintain their redox state and mitigate the negative effects of excess ROS is by generating nicotinamide adenine dinucleotide phosphate (NADPH) to act as an antioxidant. This helps the cells avoid apoptosis or necrosis, as well as, providing a nucleotide backbone for continued assembly of cellular biomass [96, 101].

Advanced cancers can also avoid unwanted cell fate and maintain survival by modulating their calcium signaling. Calcium ions (Ca${}^{2+}$) are secondary messengers which regulate a variety of cellular functions. The frequency, duration, and volume of Ca${}^{2+}$ concentrations in the cell, elicits a diverse cellular response, ranging from cell death to cell migration, activation of specific transcription factors and stimulating metabolism, processes that attribute to the aggressiveness of certain cancers [105, 108]. There are many different types of Ca${}^{2+}$ channels, transporters, and regulatory proteins involved in facilitating the movement of Ca${}^{2+}$ in the extracellular and intracellular space, making their way through organelles such as the ER and mitochondria [97, 108, 109, 110]. Here we focus on mitochondrial Ca${}^{2+}$ as it contributes to the regulation of metabolism and the invasiveness of TNBC. Since cancer cells rely on the production of metabolic intermediates, such as amino acids precursors, nucleotides, lipids, and sugars over mitochondrial ATP production, they become highly dependent on glycolysis and the tricarboxylic acid (TCA) cycle for the generation of these intermediates (among other pathways). In this way, cancer cells reprogram their metabolism to tightly regulate the amount of substrate that is produced in the TCA cycle and thereby fed into OXPHOS. There are four TCA cycle dehydrogenase enzymes that are sensitive to mitochondrial Ca${}^{2+}$ concentrations for regulating their enzymatic activity and substrate availability. Mitochondrial glycerol-3-phosphate dehydrogenase (GPDH), oxoglutarate (also known as $\alpha$-ketoglutarate) dehydrogenase (OGDH), and isocitrate dehydrogenase (IDH) utilize Ca${}^{2+}$ directly for their enzymatic activity [99, 111, 112]. Pyruvate dehydrogenase (PDH), on the other hand, indirectly relies on its activation through dephosphorylation by the Ca${}^{2+}$ sensitive pyruvate dehydrogenase phosphatase (PDP) [99, 111, 112]. Stimulation of IDH and OGDH enzymatic activity by Ca${}^{2+}$ is crucial for OXPHOS to proceed as both enzymes generate NADH that feeds into the respiratory chain (Fig. 2).

At the OMM, mitochondria uptake their calcium through VDAC, mentioned previously for its role in glycolytic cancers and at the inner mitochondrial membrane (IMM) through the mitochondrial calcium uniporter (MCU) [97, 113]. Several studies have analyzed MCU mRNA expression levels in available cancer patient databases and found that MCU expression is elevated in breast cancer as compared with normal tissues, and certain subtypes such as invasive ductal carcinoma have higher MCU expression when compared with ductal carcinoma [114, 115, 116]. MCU overexpression is also correlated with poor patient prognosis and with lymph node invasion in breast cancer patients [106, 114, 115]. Notably, when MCU expression was compared between 180 breast cancer patients, it was found that MCU is the most elevated in basal-like (TNBC subtype) molecular type of breast cancer [117]. Interestingly, a majority of the studies that have investigated the role of MCU in breast cancer migration and invasion have used TNBC cell lines to address these questions due to the aggressive nature of this cancer [49, 106, 114, 115, 117, 118, 119, 120]. One study looking at three different TNBC cell lines, MDA-MB-231, MDA-MB-468, and BT-549, showed that siRNA silencing of MCU reduced cell migration in all three TNBC cell lines. Furthermore, in MCU deleted MDA-MB-231 mouse xenografts, they observed a significant reduction in tumor progression and lymph node infiltration [106]. Additionally, the same study found that MCU silencing reduced ROS production, HIF-1$\alpha$ transcription, and cell migration [106]. It is crucial to note the complexity of the metabolic fluxes involved in TNBC that go beyond glycolysis and OXPHOS, especially in conjunction with the intricacy of mitochondrial Ca${}^{2+}$. A recent in vitro study showed TNBC cells having sustained IP3 Ca${}^{2+}$ oscillations that are linked to their mitochondrial Ca${}^{2+}$ uptake. This was shown to be linked to regulating fatty acid oxidation and reducing cell migration in TNBC cells [118]. Further research contributions that intersect the topics of metabolic reprogramming and mitochondrial Ca${}^{2+}$ in difficult to treat cancers such as TNBC can create novel opportunities for therapeutic targets.

Advanced cancer diagnoses are often met with limited treatment options. Unlike other subtypes of breast cancer, patients with TNBC pose unique challenges for practitioners due to the lack of target receptors, therefore, targeting metabolic reprogramming in this cancer has emerged as a novel avenue. Some of these novel approaches, include the combination of metabolic inhibitors with current standard treatments, chemotherapy, radiation therapy, and immunotherapy, to improve outcomes for TNBC patients. An innovative approach utilizes nanoparticles for targeted drug delivery in cancer patients, including those with TNBC, to enhance treatment effectiveness while minimizing side effects and dosage. For example, nanoparticles have been employed as dual-delivery systems for combining chemotherapy agents with metabolic inhibitors. Some groups have investigated nanoparticles to deliver paclitaxel, a chemotherapy agent, and lonidamine, a glycolytic inhibitor to cancer cells, aiming to counter the Warburg effect and reduce the highly glycolytic phenotype of certain cancers [121, 122]. Another possible avenue involves diclofenac, a non-steroidal anti-inflammatory drug which reduced cell proliferation and increased apoptosis, particularly in TNBC by uniquely targeting c-MYC and thereby reducing glucose metabolism and lactate transport [123]. Pilot clinical trials have been exploring dietary interventions such as ketogenic diets and controlled fasting as well as the use of natural compounds in conjunction with chemotherapy and radiation treatment for advanced cancers [124, 125, 126, 127]. Some of these pilot studies have suggested that a ketogenic diet alongside chemotherapy could potentially yield better results for specific breast cancer subtypes. However, it is crucial to emphasize that there is currently no conclusive evidence regarding the effectiveness of this approach, and more extensive research and large-scale clinical trials are necessary to assess both its safety and efficacy [128].

Researchers have begun investigating a way to target MCU, as a result of the increased interest in mitochondrial Ca${}^{2+}$ and the correlation between TNBC and MCU overexpression [129, 130, 131]. In fact, 1600 FDA-approved drugs were screened for their ability to modulate mitochondrial Ca${}^{2+}$ without affecting cytosolic Ca${}^{2+}$ [129]. This research group identified a negative modulator of MCU, benzethonium, an antiseptic, which inhibited cell growth and migration in TNBC. Interestingly, another group synthesized an analog of the flavonoid protoapigenone, RY10-4. This drug was selectively cytotoxic to cancer cells and caused an increase in the expression of MCU which led to Ca${}^{2+}$ overload and in turn cell death in TNBC cells [130]. Studies like these are relatively recent and shed light on the intricacy and autonomy that can exist between cancers and even within the same type of cancer. Recognizing the molecular and clinical heterogeneity within TNBC subtypes is crucial, as therapeutic approaches may vary in effectiveness among them. Tailoring metabolic reprogramming targeting to different molecular subtypes within TNBC is key. Using cancer metabolism as a therapeutic target holds promise for enhancing cancer treatments significantly. For a deeper understanding of this growing field, we invite readers to consult recent articles on metabolic therapies [132, 133].