Background

Cancer is a serious public health problem. The incurrence and mortality is increasing year by year [1]. In addition to conventional radiotherapy, chemotherapy, and surgery, there are currently more and more popular neoadjuvant chemotherapy and molecular targeted therapies. These treatment options can cure early and part of the intermediate tumors in certain degrees, but are not ideal for most of cancer in middle and late stages [2]. Among many reasons, the treatment resistance is the one of major drawbacks. Radiotherapy and chemotherapy, as the routine treatment, face substantial challenges of resistance. However, the characters of chemo- and radio-resistance in different kinds of cancers are not exactly the same.

In the early 1920s, German biochemist and physiologist Otto Warburg conducted groundbreaking research and proposed the famous “Warburg effect”: Tumor cells prefer to use glycolysis for glucose metabolism even in oxygen-rich conditions, rather than more efficient mitochondrial oxidative phosphorylation for ATP production [3]. Actually, the entire metabolic network reprograms under the control of oncogenes and tumor suppressor genes, and the flow of nutrient in metabolic networks is also redefined in the process of tumorigenesis. Metabolic reprogramming provides critical information for clinical oncology. The aberrant glucose metabolism is a major kind of metabolic reprogramming in cancer [4], and recent studies have shown that aberrant glucose metabolism regulates cancer proliferation, cell cycle, drug resistance, and DNA repair [5,6,7]. As the molecular mechanisms underlying chemo- and radio-resistance are still poorly understood, the alteration of glucose metabolism in cancer provides new ideas to explain chemo- and radio-resistance. Herein, this review updates the mechanisms of metabolic reprogramming involved in tumor chemo- and radio-resistance.

Main text

The overview of glucose metabolic reprogramming

Metabolic reprogramming refers to the redefinition of the flow and flux of nutrient in tumor cells in the metabolic network to meet the needs of tumor cells for energy and anabolism [8]. Under oxygen-rich conditions, normal or differentiated cells can metabolize glucose and produce carbon dioxide through a tricarboxylic acid cycle (TCA), which produces 30 or 32 mol of adenosine triphosphate (ATP) per mole of glucose and a small amount of lactate during oxidative phosphorylation [9]. Only under hypoxic conditions, normal or differentiated cells produce large amounts of lactic acid by anaerobic glycolysis. However, German scientist Otto Warburg first proposed that tumor cells rely mainly on glycolysis to provide energy under aerobic conditions [3](Fig. 1). Weinberg characterized “aberrant metabolic phenotype” with “autologous proliferation signaling, apoptosis resistance, evasion of proliferation inhibition, continuous angiogenesis, infiltration and migration, unlimited replication capacity, immune escape” in tumor cells.

Fig. 1
figure 1

The energy metabolism of cancer cells. Under aerobic condition, Most of the glucose is first converted to pyruvate via glycolysis in the cytosol. Most pyruvate are mostly processed to lactate via glycolytic pyruvate even in the presence of oxygen, and only a small portion of pyruvates enters the mitochondria to produce CO2 by undergoing TCA cycle. In addition, small proportion of the glucose is diverted into the upstream of pyruvate production for biosynthesis (e.g., pentose phosphate pathway, and amino acid synthesis)

Glucose metabolic reprogramming between aerobic glycolysis and oxidative phosphorylation, previously speculated as exclusively observable in cancer cells, exists in various types of immune and stromal cells in many different pathological conditions other than cancer [6]. It has been well established that tumor cells have elevated rates of glucose uptake and high lactate production in the presence of oxygen, known as aerobic glycolysis (also termed the Warburg effect) [10]. As a matter of fact, high lactate production also remodels the tumor microenvironment (TME) by contributing to acidosis, acting as a cancer cell metabolic fuel and inducing immunosuppression resulting in aggressive proliferation, invasion, migration and resistance therapy [4]. However, the molecular mechanisms involved in the changes of glucose metabolism are complex. Changes in the tumor microenvironment, activation of oncogenes, and inactivation of tumor suppressor genes all contribute to the disruption of metabolism and steady-state metabolism of cells, ultimately leading to aberrant glucose metabolism [11, 12]. Specific oncogenes activation or tumor suppressor genes deactivation can reprogram the underlying metabolism of tumor tissues. Some genes can act as initiators of glucose consumption, include myc, KRAS, and BRCA1 [13,14,15]. Despite the progression, we still do not fully know the metabolic pathways that are reprogrammed by oncogenes or suppressor genes.

Glucose metabolic reprogramming and chemo- and radio-resistance

Tumor cell survival under aberrant metabolism of glucose is a vital step not only for the process of tumorigenesis but also in treatment resistance and recurrence, especially for the occurrence of treatment resistance [4]. Chemotherapy in the form of neo-adjuvant or adjuvant therapy is the dominant treatment for most of cancers; the resistance directly affects the survival and prognosis of cancer patients [16]. Theoretically, the tumor mass, made of distinct chemo-resistant cell populations has been recognized as an important mechanism for chemo-resistance [17]. Actually, inhibition of glycolysis not merely inhibited cell proliferation, but alleviated resistance to chemotherapeutic drugs.

Existing evidence indicates that increased glucose uptake and enhanced aerobic glycolysis are able to induce the intrinsic or acquired resistance to DDP in gastric cancer cells [18]. Elevated lactate levels caused by aberrantly activated glycolysis can reinforce DNA repair and promote cisplatin-resistance in cervical carcinoma cells via the inactivation of histone deacetylase [19]. High-precision radiation therapy enables radiation oncologists to decrease delivery of an excessive dose of radiation to normal tissues and also to administer a high and booster dose of radiation, particularly to small target fractions in a malignant tumor [20]. Previous studies have revealed that the Warburg effect or aerobic glycolysis promotes the radio-resistance of various malignant tumors via generating a chemically reduced milieu associated with the development of radio-resistance in laryngeal carcinoma, prostate cancer, head and neck cancer [21,22,23,24,25,26]. For example, activation of adenosine monophosphate-activated protein kinase (AMPK) mediates metabolic reprogramming in resistant cancer cells through promoting both the Warburg effect and also mitochondrial biogenesis [27,28,29,30]. However, both the gene network triggering metabolic reprogramming and the molecular mechanism linking the reprogramming with radio-resistance remain to be determined.

The mechanisms of glucose metabolic reprogramming-mediated chemo- and radio-resistance

Although increasing evidence has confirmed that glucose metabolic reprogramming can induce tumor radiotherapy and chemotherapy resistance, the specific mechanisms are still not clear [31,32,33,34]. The previously reported resistance mechanisms include mutations or increases in drug targets, changes in drug metabolism, and alterations in DNA repair, overexpression of anti-apoptotic genes, and inactivation of apoptotic gene products, immunosuppression and the formation of CSCs, etc.

With the increasing research understanding on the resistance of chemo- and radiotherapy, the researchers have pointed out that cancer stem cells, tumor microenvironment, autophagy, and exosomes are all closely related to tumor chemo- and radio-resistance. In fact, recent reports have shown that chemo- and radio-resistance acquisition is coupled to deregulate glucose metabolism and glycolysis [35]. Signaling pathways related to chemo-radiotherapy resistance are abnormally activated or inactivated during metabolic stress, such as Wnt, PI3K/AKT, Notch, NF-κB, MAPK [36,114, 115]. During recent years, the interaction between immunosuppression and treatment resistance in different subsets of tumor cells within the TME was increasingly valued by cancer researchers [116,117,118] (Fig. 3).

Fig. 3
figure 3

The immunosuppressive effect of the tumor microenvironment. The hypoxia and acidosis of the tumor microenvironment (TME) contribute to immunosuppression via several mechanisms. These mechanisms include increased accumulation, activation, and expansion of immunosuppressive regulatory T (Treg) cells; recruitment of inflammatory monocytes and tumor-associated macrophages (TAMs) and reprogramming of TAMs towards the pro-tumor M2 phenotype; suppression of dendritic cell (DC) maturation, which results in inhibiting activation of tumour-specific cytotoxic T lymphocytes (CTLs). Importantly, the programmed cell death protein 1 (PD-1)–programmed cell death 1 ligand 1 (PD-L1) pathway is often activated in the TME as a mechanism to evade anticancer immune responses, with up-regulation of PD-L1 expression on TAMs, DCs, and tumor cells. In addition, tumor-infiltrating CTLs typically up-regulate PD-1, limiting their cytotoxic potential against tumor cells. CCL20, C-C-motif chemokine ligand 20; CXCL, C-X-C-motif chemokine ligand; GM-CSF, granulocyte–macrophage colony-stimulating factor; TGFβ, transforming growth factor β; IL, Interleukin

Tumor cells have to adapt their metabolism to survive and proliferate in this harsh microenvironment. Changes in the tumor microenvironment can affect the levels of infiltrating cell-associated chemokines in tumor cells. These chemokines, in turn, recruit Tregs to tumor tissues to exert immunosuppressive effects [119]. For example, under an inflamed microenvironment, the TLR (Toll-like receptor) can increase glucose uptake and lactate production in Treg cells through up-regulating the expression of key enzymes Glut1 (a glucose transporters), which is beneficial to the proliferation of Treg cells [142]. The newest documents have found that the exosomes can induce the formation of cancer stem cells (CSCs) to decrease the effect of chemo- and radio-therapy [145,146,147] (Fig. 4).

Fig. 4
figure 4

The role of the exosomes in the formation of CSCs. The cancer cells with enhanced glycolysis could release a large amount of exosomes contained several of glycolytic enzymes and CSCs markers. These exosomes can be taken up by the recipient cancer cells, and then promote the glycolysis and induce the dedifferentiation of cancer cells to acquire stemness phenotype through transfer their stemness-related molecules

The aberrant glycolytic reaction of CSCs contributes to therapy resistance via preserving stemness and tumorigenic properties of CSCs [148,149,150]. Exosomal LMP1 activates the PI3K/AKT pathway, and then up-regulates the expression of the surface marker CD44+/High, ultimately increasing the populations of CD44+/High cells, which are the putative stem cell in nasopharyngeal carcinoma cells [150,151,152]. Besides, exosomal LMP1 could reduce the phosphorylation of AMPK and changed its subcellular location after irradiation, which appears to occur through a disruption of the physical interaction between AMPK and DNA-PK, and then causes decreasing in AMPK activity which is associated with LMP1-mediated glycolysis and resistance to apoptosis induced by irradiation [126, 153, 154]. Similarity, the resistant cancer cells with enhancing glycolysis can secrete a large amount of exosomes containing EpCAM protein, an epithelial cancer stem-like cell markers and glycolysis enzymes [126, 155,156,157,158,159]. The neighboring non-resistant cells can take up these exosomes and positively regulate mTOR and epithelial growth factor receptor (EGFR) signaling pathways to enhance the glycolysis and promote EpCAM+ tumor cells to ovarian cancer stem cells (CD133+ and CD117+CD44+) and putative drug-resistant tumor cell phenotype (EpCAM+ CD45+) transformation [152, 155, 159,160,161,162]. Besides, the exosomes secreted from resistant tumor cells can be taken up by non-resistant cells and induce the production of ROS via enhancing metabolic reprogramming [163]. The increased level of ROS can activate the Wnt signaling pathway to accumulate the cancer stem-like cells with CD44v8-10high/Fbw7high/c-Myclow or CD44v8-10high/Fbw7low/c-Mychigh, leading to the formation of resistant sites [147, 149, 152, 164].

Transport of exosomal components can contribute to the chemo- and radio-resistance of cancer cells [165,166,167]. Among them, transfer of miR-100, miR-222 and miR-30a from the exosomes derived from adriamycin- and docetaxel-resistant MCF-7 breast cancer cells to drug sensitive MCF-7 cells increased the drug resistance of the sensitive cell line through increasing CSCs proportion in cancer cell populations and promoting the phenotypic transition of non-CSCs toward the CSCs phenotype [168,169,170]. Actually, exosomal HSPs could be involved in the occurrence of EMT and ECM remodeling which were closely associated with the formation of stem cells to mediate the resistance of cancer cells [171]. E.g. exosomal HspDNAJB8, an Hsp40 family member, has a role in maintenance of renal cell carcinoma CSCs/CICs (called cancer stem–like cells/cancer-initiating cells), resistance to chemotherapy and radiotherapy [172, 173]. Similarly, the exosomal lncRNA UCA1 is demonstrated to possibly activate the Wnt signaling pathway and facilitate the malignant transformation of stem cell through the modification of the gene network by tail modification of histone to increase chemo-resistance of cancer cells [174, 175].

Exosomes are speculated as a novel target for solving the radio- and chemo-resistance because they can promote CSCs phenotype. However, the research about the role of exosomes in the treatment resistance of cancer is not much more; it isn’t a good explanation to verify the concrete effect of exosomes and need to more studies to confirm.

Perspectives of metabolic inhibitors

Up to date, the metabolic inhibitors aim to inhibiting the enzymes about tumor metabolism, and then decrease the level of cancer glucose consumption to decrease amount of ATP, attenuating amino acids and nucleotides synthesis, and generate reactive oxygen species (ROS) [126, 176,177,178,179,180,181,182]. Metabolic inhibitors reduce the metabolite levels in glycolysis, PPP and nucleotide biosynthetic pathways to down-regulate the resistant effect of cancer cells via preventing DNA damage repair and enhancing chemotherapy and radiation responsiveness [47, 183]. For example, 3-BrPA (3-bromopyruvate), a special inhibitor of HK-2 kinase, can induce the imbalance of intracellular redox via inhibiting the glycolysis and strengthening the tricarboxylic acid cycle in cancer cells, during which a large amount of ROS is produced and accumulated in the cancer cells, destroying the normal structure inside the cell and causing the cell to gradually die [184]. Therefore, 3-BrPA can sensitize first-line anti-tumor drugs in the resistant cancer cells, such as 5-fluorouracil, doxorubicin, mycin, mitoxantrone and platinum drugs (e.g. cisplatin, oxaliplatin) [185]. In addition, the covalent inhibitor JX06 targeting PDK via structural modification hinders access of ATP to its binding pocket and in turn impairs PDK1 enzymatic activity, which increases the sensitivity of chemotherapy and radiotherapy by promoting cellular oxidative stress and apoptosis [186]. FX11, an LDHA inhibitor, can be capable of blocking aerobic glycolysis via inactivating the CK2/PKM2/LDHA axis to induce oxidative stress, and suppress drug resistance in various cancers [187]. 3PO, a glycolysis inhibitor targeting PFKFB3, can inhibit the glycolysis of nintedanib- and sunitinib-resistant tumor cells via inducing cell-cycle arrest and apoptosis, and thus promote the therapeutic efficacy of chemo- and radio-therapy [188].

Even though some metabolic inhibitors have been approved for clinical treatment, the efficacy is not ideal and rigorous evidence-based medical evidence lacks. There are inextricable links between cell metabolism, tumor immunity, and tumor epigenetics. Metabolic inhibitors can only achieve maximum biological efficacy when combined with targeted inhibitors of macromolecule synthesis, cellular immune-agonists, and agonists or inhibitors associated with metabolic pathways. Furthermore, most metabolic inhibitors lack specificity and cannot target tumor cells and have a killing effect on normal cells. Therefore, the researches on metabolic inhibitors have promising development prospects.

Conclusions

Extensive studies have provided strong evidence for reprogramming of cancer metabolism in chemo- and radio-resistant cancer. Aberrant glucose metabolism could alter many physiological activities(Fig. 5), e.g. inducing DNA damage repair, enhancing autophagy, changing tumor microenvironment and increasing the secretion of exosomes, etc. However, these alterations are not a simple relationship between chemo- and radio-resistance and glucose metabolism. Additional studies are needed to better understand the molecular mechanisms linking resistance to cell metabolism. Additionally, it will be important to understand whether the effects of metabolic inhibitors are cell type-specific. Because changes in treatment resistance can directly or indirectly impact multiple processes--including metabolism, ROS signaling, and calcium signals. The outcome may be critically dependent on cell types. Finally, once the interconnections between the glucose metabolism of cancer cells and resistance to treatments are better understood, we will hopefully be able to harness this information to devise therapies for cancer resistance.

Fig. 5
figure 5

The overview of acquired chemoradiotherapy resistance mediated by metabolic reprogramming in cancer cells