Reprogramming of amino acid metabolism in cholangiocarcinoma: A potential target for metabolic-targeted therapy

Abstract

The incidence of cholangiocarcinoma (CCA), a highly aggressive malignancy of the bile duct epithelia, has been gradually increasing worldwide. However, curative treatments are still limited. Novel therapeutic strategies are urgently needed to improve patients’ survival and quality of life. Metabolic reprogramming has been well recognized as one of the hallmark processes supporting the development of several cancer types, including CCA. Apart from the Warburg effect and high glucose requirement in CCA cells, amino acid metabolism is also found to be essential in CCA development and progression. Upregulation of proteins and enzymes involved in amino acid metabolism is reported in CCA, typically associated with a poor prognosis for patients. Targeting these proteins and enzymes has been shown to retard CCA progression, and thus, they are promising targets for drug development. This article reviews the reprogramming of amino acid metabolism in CCA and its roles in CCA progression, such as aggressive phenotypes. The up-to-date development of therapeutic agents targeting particular proteins in amino acid metabolism is also discussed. A summary of the current knowledge gap and directions for further research are also provided and proposed.

Key Words: Amino acid; Biliary tract cancer; Cancer metabolism; Cholangiocarcinoma; Metabolic reprogramming

Core Tip: Metabolic reprogramming is a malignant hallmark found in many types of cancer, including cholangiocarcinoma (CCA). A higher requirement and dependency of certain amino acids is found in CCA and is generally associated with aggressive phenotypes of cancer cells, leading to poor prognosis of patients. In this article, amino acids with reports of their significance in CCA development and progression are reviewed. The mechanisms underlying amino acid metabolic reprogramming and their networks with intracellular signaling are also discussed. Up-to-date interventions and the opportunity for metabolic targeting in CCA treatment are proposed.

INTRODUCTION

Metabolic reprogramming is one of the malignant hallmarks of cancers[1]. Cancer cells exhibit metabolic plasticity, allowing them to utilize alternative metabolic pathways or alternative energy sources for rapid growth, survive under stressful conditions with high invasive ability[2]. Since the discovery of high aerobic glycolysis rates in cancer cells, also known as the Warburg effect, the metabolic reprogramming of cancers has drawn the attention of scientists and researchers to further investigate this unusual metabolism of cancer cells[3]. Growing evidence suggests that not only glucose metabolism, the primary energy-generating pathway for cells, but also the metabolism of other biomolecules, including fatty acids, nucleic acids, and amino acids, is significantly altered in cancer cells to support their malignant phenotypes[4]. The alterations of cellular metabolism involve not only catabolic pathways that yield high-energy compounds for the cells, but also anabolic pathways that synthesize building blocks and necessary molecules needed for the functional modifications of major biomolecules[5]. Interventions targeting these metabolic pathways are expected to be a promising strategy for cancer treatments[6]. A better understanding of the molecular basis and insights into the mechanisms of the metabolic alterations of these biomolecules is necessary for further development of anti-metabolic strategies to overcome cancer.

The reprogramming of amino acid metabolism has been relatively less studied than glucose and other carbohydrate metabolisms[7]. Amino acid metabolism is closely related to monosaccharide metabolism and could be integrated with the metabolic pathways of other nutrients[8]. The carbon skeleton of amino acids can be converted into intermediates in glucose metabolism and also in energy production pathways, such as the tricarboxylic acid (TCA) cycle. In addition to serving as the energy source, amino acids also function as the building blocks and nitrogenous backbones of other functional biomolecules, such as neurotransmitters and hormones[9]. In certain disorders, reprogramming of amino acid metabolism is associated with the occurrence of diseases[10]. For example, in specific cancer types, amino acids cannot be fully classified as essential or non-essential, because some cancer cell types become highly dependent on generally non-essential amino acids or become auxotrophic under specific circumstances[11,12]. Many types of invasive adenocarcinoma exhibit upregulated L-type amino acid transporters (LATs), which are predominantly associated with a poor prognosis in patients, including those with cholangiocarcinoma (CCA), a malignancy of bile duct epithelia[13-16]. These findings suggest the importance of high amino acid requirements in cancer cells and the potential significance of reprogramming amino acid metabolism as a therapeutic target[17].

CCA is a malignant transformation of epithelial cells lining the biliary trees. It can be classified according to anatomical location as intrahepatic (iCCA) and extrahepatic (eCCA) subtypes, in which the latter can be further subdivided into perihilar (pCCA) and distal (dCCA) types[18]. CCA accounts for 3% of all gastrointestinal cancers and ranks as the second most prevalent primary liver cancer[19]. Considered rare in Western countries, CCA has an exceptionally high incidence in East and Southeast Asia. Notably, its incidence is also increasing worldwide[20]. CCA is a cancer characterized by a highly heterogeneous genetic background and phenotypes[21,22]. Whole-genome and whole-exome sequencing studies across different regions have revealed distinct mutational landscapes among intrahepatic (iCCA), perihilar (pCCA), and distal (dCCA) subtypes. The variety of genetic backgrounds highlights the fundamental biological differences among these anatomical subtypes, which might include a unique metabolic reprogramming[19,23-25]. Actionable alterations, including fibroblast growth factor receptor 2 (FGFR2) fusions and isocitrate dehydrogenase 1 (IDH1) mutations, occur predominantly in iCCA, whereas pCCA and dCCA more frequently harbor human epidermal growth factor receptor 2 amplifications and KRAS mutations. In addition, mutation frequencies vary by geographic region and disease etiology; for example, IDH1 mutations are less common in Asian populations. Liver fluke-associated CCAs show lower IDH1 but higher TP53 mutation rates compared with CCAs of other etiologies[21,26,27]. IDH1 is directly functional in the TCA cycle, whereas KRAS and TP53 can regulate the expression of several genes involved in energy metabolism. These suggest the distinct patterns of metabolic preference and alteration in CCA originated from different etiologies and regions. A different pattern of metabolic reprogramming in CCA might, therefore, be an opportunity as well as a challenge in targeted therapeutic development of CCA.

Despite advanced understanding of pathogenesis and the development of new therapeutic modalities, the curative treatment of CCA remains limited. This resulted in poor 5-year survival rates of 7% to 20% among patients[28]. The challenges of CCA treatment stem from several factors, including the aggressive, silently progressive nature of the disease, which often leads to diagnosis at an advanced stage[18]. The only curative surgical tumor resection can be performed in approximately one-third of patients who have an early diagnosis[29]. Available targeted therapies for the FGFR2 fusion gene benefit only 10%-15% of patients with this mutation[30-32]. In contrast, the majority of patients in the region with the highest CCA incidence do not harbor this mutation[32]. Standard chemotherapy for advanced-stage CCA can prolong survival in those fit for the chemotherapeutic course; however, the improvement is modest, with a median survival of less than a year[28]. Therefore, appropriate and effective treatments for CCA have remained a challenge for decades and require further investigation to identify novel therapeutic modalities for curative treatment and to improve patients’ quality of life.

Metabolic reprogramming is also one of the malignant hallmarks of CCA[33], and aberrant metabolisms of glucose, fatty acids, and amino acids have been gradually reported. As mentioned earlier, upregulated LATs are common among invasive adenocarcinoma, including CCA[13,15,16]. High LAT expression has been reported in several biliary tract cancers, including intrahepatic CCA[16], primary sclerosing cholangitis-associated biliary tract cancer[34], and biliary tract adenocarcinoma[35]. These findings suggest increased amino acid uptake and, hence, a dependency on amino acids in CCA cells. In this context, tumor metabolism is of particular interest because metabolic reprogramming, a hallmark of cancer, is driven by oncogenic signaling and influences gene and protein expression. Metabolic reprogramming then provides metabolic enzymes and metabolites that can serve as diagnostic and prognostic biomarkers and as targets for treatment[36,37]. This review then includes evidence of the amino acid requirement for CCA, which may be targetable for CCA treatment. A comprehensive perspective on each amino acid metabolism and its associated pathways, as well as the potential for clinical translation, is also discussed.

REPROGRAMMING OF AMINO ACID METABOLISM IN CCA

Glutamine: A generally non-essential amino acid that is essential for CCA

Glutamine is recognized as an alternative energy source for cancer cells, in addition to glucose[38]. Many cancers increase glutamine uptake and utilize it as an energy source by replenishing the TCA cycle intermediate, also known as the anaplerotic pathway[39]. This pathway can be achieved through glutaminolysis, the conversion of imported glutamine into a TCA cycle intermediate. Mitochondrial glutaminolysis begins with the transportation of glutamine through the inner mitochondrial membrane via a glutamine transporter, solute carrier family 1 member 5[40], followed by the conversion of glutamine to glutamate by glutaminases (GLSs) that are localized in mitochondria[41,42]. Mitochondrial glutamate can be exported to the cytosol through the solute carrier family 25 member 18 and solute carrier family 25 member 22 transporters and then participate in the biosynthesis of glutathione, or undergo conversion into alpha-ketoglutarate (α-KG) by glutamate dehydrogenase 1 (GLUD1/GDH1). Mitochondrial α-KG enters the TCA cycle, contributing to either the oxidative phosphorylation process or the reductive carboxylation pathway. NADH or FADH₂ derived from glutamine can serve as electron donors for energy production[43]. Citrate, generated by reductive carboxylation of α-KG, is especially crucial for lipid synthesis under low-oxygen conditions[44]. On the other hand, α-KG can also be exported from mitochondria through solute carrier family 25 member 11 to the cytoplasm[45] and participate in fatty acid biosynthesis and NADH generation[46]. Thus, glutamine can serve not only as an energy source for cancer cells but also as a building block for other anabolic pathways required for cancer cell proliferation and tumor growth.

In iCCA, combined data from the Oncomine database with immunohistochemistry suggest overexpression of glutaminase-1 (GLS1) in iCCA tissues compared with peritumoral areas[47]. Overexpression of GLS1 is associated with metastasis in patients with CCA and the increased invasive ability of CCA cells in vitro. High expression of GLS1 is negatively correlated with E-cadherin expression, an epithelial marker highly expressed in epithelial cells. Conversely, GLS1 expression is positively correlated with vimentin expression, a mesenchymal marker, suggesting that GLS1 is associated with epithelial-mesenchymal transition and is associated with the higher metastatic potential of CCA cells when GLS1 is overexpressed. Moreover, GLS1 expression was significantly associated with poor tumor differentiation and lymphatic metastasis in the patients. Patients with high GLS1 expression had poorer overall and disease-free survival than those with low GLS1 expression in tumor tissues. Collectively, GLS1 is, hence, proposed as a promising therapeutic target for iCCA treatment[47]. However, the molecular mechanisms need to be investigated to determine whether the anti-proliferative and anti-metastatic effects of suppressed GLS1 are direct effects of its functional loss in other biological processes or indirect effects from insufficient cellular energy.

Due to the high proliferative rates and active metabolism of cancer cells, significant amounts of energy are consumed to support their rapid growth[48]. When metabolic oxygen demand exceeds supply, oxygen-deficient areas in cancer cells are exacerbated, and their metabolism changes[49]. Substantial depletion of glutamine in the tumor core region is found to be a common event in solid tumors[49]. An in vitro study suggests that CCA cell growth is highly dependent on glutamine[50]. A resistance to cisplatin and gemcitabine, the standard chemotherapy for CCA, can also be overcome by glutamine depletion. Previous studies have shown that CCA cells with cisplatin resistance exhibit enhanced expression of glucose transporter and glutamine transporter, as well as genes associated with cancer progression, including L1CAM, AXL, and ZEB2. Thus, glucose starvation and treatment with the GLS1 inhibitor CB-839 (telaglenastat) exerted synergistic effects in reducing intrahepatic CCA cell proliferation and overcoming the cisplatin-resistant phenotype[50]. The underlying cisplatin resistance mechanisms in CCA cells may also result from adaptation that reprograms glutamine metabolism[51]. A recent study also demonstrated significantly higher glutamine levels in sera from patients with CCA compared with the normal control group. The in vitro and in vivo studies suggested that glutamine is beneficial for CCA cells, preventing ferroptosis and promoting tumor growth. The regulation of glutamine metabolism can also be modulated by gut microbiota via the anaplastic lymphoma kinase/NADPH oxidase 1 signaling axis[52]. Exogenous glutamine supplementation can also protect CCA cells from ferroptosis. Maintaining glutamine levels is thereby vital to CCA cells, in addition to replenishing the TCA cycle via glutamate and α-KG generation. Another study showed that glutamine synthetase (GS) was upregulated in CCA cells by the mediation of sirtuin 6. Inhibiting either GS or sirtuin 6 exerted significant anti-tumor effects and sensitized CCA cells to chemotherapy[53]. These findings affirm a significant role of glutamine in CCA progression and thus targeting glutamine is also promising for CCA cases with documented chemoresistance. In addition to preclinical models, a phase II clinical trial using telaglenastat for solid tumors, including biliary tract cancer, suggests that telaglenastat is safe with a recommended dose of 800 mg twice daily. At a dose with favorable pharmacokinetics and pharmacodynamics, it also demonstrates anti-tumor activity, supporting further clinical development, while the common adverse effects are fatigue (23%) and nausea (19%) among 120 patients enrolled in the trial[54].

Upregulation of glutamate-oxaloacetate transaminase (GOT1) has been associated with malignant phenotypes of various cancers, including pancreatic ductal adenocarcinoma[55], colorectal cancer[56], and prostate cancer[57]. Further, annexin A1 (ANXA1) can promote CCA proliferation via glutamine metabolism through GOT1 stabilization[58]. ANXA1 generally functions as a scaffold protein, connecting ubiquitin-specific protease 5 with GOT1 and facilitating their interaction. This scaffolding activity promotes GOT1 deubiquitylation and stabilization, thereby inhibiting its degradation via the ubiquitin-proteasome pathway[58]. The glutamine metabolism is thereby enhanced, supporting CCA tumor growth. All these studies reiterate the significance of glutamine, an amino acid classified as non-essential in most cells but highly essential in CCA cells for growth and progression. A schematic summary of the glutamine metabolism and its roles in CCA progression is shown in Figure 1.

Figure 1 The alteration of glutamine metabolism in cholangiocarcinoma cells. Glutaminolysis begins with the transportation of glutamine through the inner mitochondrial membrane through solute carrier family 1 member 5, followed by the conversion of glutamine to glutamate by GLSs. Mitochondrial glutamate can be converted to alpha-ketoglutarate by GDH1. Aspartate can be exported to the cytoplasm, undergo transamination, and converted to oxaloacetate by glutamate-oxaloacetate transaminase (GOT1). Annexin A1 generally functions as a scaffold protein, connecting ubiquitin-specific protease 5 with GOT1 and facilitating their interaction. This scaffolding activity promotes GOT1 deubiquitylation and stabilization, inhibiting its degradation via the ubiquitin-proteasome pathway. SLC1A5: Solute carrier family 1 member 5; AGC: Aspartate-glutamate carrier; GLS1: Glutaminase 1; GDH: Glutamate dehydrogenase; GOT: Glutamic-oxaloacetic transaminase; α-KG: Alpha-ketoglutarate; ANXA1: Annexin A1; USP5: Ubiquitin-specific protease 5; GAD1: Glutamate decarboxylase 1; SLC25A11: Solute carrier family 25 member 11; TCA: Tricarboxylic acid.

Gamma-aminobutyric acid: A glutamate derivative serving as a negative regulator of CCA

Gamma-aminobutyric acid (GABA) is a potent amino acid neurotransmitter with physiologic effects throughout the body[59,60]. Although classified as an amino acid containing both an amino and a carboxylic group, GABA is different from the other common amino acids in that its amino group is not attached to the α-carbon. GABA is not translated from genetic codons, but is mainly synthesized by the decarboxylation of glutamate by glutamate decarboxylase[61]. Glutamate, an essential precursor for GABA synthesis, is derived from glutamine and α-KG, the TCA cycle-derived intermediate. Glutamate dehydrogenase (GDH) catalyzes the reaction between glutamate, α-KG, and ammonia, using NAD+ or NADP+ as coenzymes. GDH directly regulates glutamate concentrations and indirectly regulates GABA levels by altering the availability of precursors[62]. A study in extrahepatic CCA models revealed that GDH promoted cell proliferation and metastasis, potentially via Smad-mediated induction of transforming growth factor-β signaling. GDH expression level was also correlated with CD34 expression, cellular differentiation, the presence or absence of capsular and vascular invasion, lymph node metastasis, neural invasion, and patient age[63].

Several studies suggest that GABA exerts antagonistic effects on CCA progression. GABA inhibits in vitro CCA cell proliferation via both cyclic AMP/protein kinase A- and d-myo-inositol-1,4,5-triphosphate/Ca²⁺-dependent pathways, leading to downregulation of phosphorylated extracellular signal-regulated kinase 1/2[64]. The in vivo study, GABA significantly decreased tumor volume, tumor cell proliferation, and vascular endothelial growth factor-A/C (VEGF-A/C) expression, thereby enhancing apoptosis in CCA. Additionally, GABA can inhibit the invasive ability of CCA cells by suppressing the activity and expression of 2 matrix metalloproteinases (MMPs), including MMP-2 and MMP-9[65]. A study in CCA cell lines under diabetogenic conditions revealed that treatment with baclofen, an aminobutyric acid B2 receptor agonist, can inhibit the phosphorylation of glycogen synthase kinase 3, thereby further phosphorylating β-catenin, leading to its degradation[66]. On the other hand, treatment with baclofen also suppresses phosphorylation of signal transducer and activator of transcription 3 (STAT3). Altogether, these inhibitions led to reduced CCA cell proliferation via downregulation of c-Myc and cyclin D1 expression[67]. Another study also suggested that the anti-proliferative and pro-apoptotic effects of GABA on CCA cells are mediated by downregulation of p-STAT3 (Tyr705) expression[67]. Collectively, all these studies have shown that, although GABA is derived from glutamine and glutamate, it exhibits anti-tumor effects in CCA. Therefore, the regulation of glutamine and glutamate metabolism in CCA cells, which maintains a balance between benefits and cytotoxicity, should be another potential point of investigation. Unveiling the metabolic networks of glutamine in CCA cells could help identify a metabolic vulnerability that could inform the development of a therapeutic strategy. The pathways involving GABA metabolism and the effects of its analogs on CCA cell biology are summarized in Figure 2.

Figure 2 Metabolic pathways of gamma-aminobutyric acid and the effects of gamma-aminobutyric acid analogs on cholangiocarcinoma cell biology. Gamma-aminobutyric acid (GABA) inhibits cholangiocarcinoma (CCA) cell proliferation via both cyclic AMP/protein kinase A and d-myo-inositol-1,4,5-triphosphate/Ca²⁺-dependent pathways, leading to downregulation of phosphorylated extracellular signal-regulated kinase 1/2. Moreover, an aminobutyric acid B2 receptor agonist can inhibit glycogen synthase kinase 3 phosphorylation, thereby further phosphorylating β-catenin, leading to its degradation. The up or down direction of the arrows indicates increased or decreased levels in CCA cells. Black arrows denote the pro-tumor effects of the genes or intermediates, while the white arrows denote their anti-tumor effects. GABA: γ-aminobutyric acid; GLS1: Glutaminase 1; GDH: Glutamate dehydrogenase; GAD1: Glutamate decarboxylase 1; cAMP: Cyclic adenosine monophosphate; IP3: Inositol trisphosphate; PKA: Protein kinase A; ERK: Extracellular signal-regulated kinase; GSK3: Glycogen synthase kinase-3.

Metabolic network of glutamine: The interaction among populations in the tumor microenvironment

As mentioned earlier, glutamine is an essential nutrient for cancer cells, including CCA. On the other hand, glutamine is a precursor for glutamate and GABA, derivatives that could be toxic to CCA cells at high concentrations. A balance between pro- and anti-tumor metabolic pathways of glutamine is crucial for tumor growth and survival. Understanding how the glutamine metabolic network in CCA cells regulates the balance between pro-tumor and anti-tumor metabolites is critically important but remains insufficiently explored. Resolving these missing pieces of information is essential for identifying effective metabolic therapeutic strategies in CCA, which requires an integrated analysis of the metabolic tug-of-war between tumor cells and immune cells within the tumor microenvironment (TME). Although direct evidence in CCA remains limited, relevant insights can be inferred from studies in other cancer types. Within the TME, cancer cells show a marked dependence on glutamine to sustain proliferation and metabolic activity, a phenomenon commonly referred to as glutamine addiction[68,69]. However, glutamine is also needed for non-tumor cells and immune cells. Tumor-associated macrophages, one of the most abundant immune populations in the TME, typically display an M2-like phenotype that supports tumor growth and immune evasion. Glutamine metabolism plays a central role in M2 macrophage activation and polarization. Specifically, the ratio of α-KG to succinate generated from glutamine metabolism regulates M2 activation, with an increased α-KG/succinate ratio favoring the M2 phenotype[70]. In addition, GS is required to maintain M2 macrophage phenotype, whereas GS inhibition can reprogram M2 macrophages toward an M1-like phenotype[71], thereby enhancing antitumor immunity and reducing immunosuppression and metastatic potential.

Beyond the macrophages, glutamine is also essential for adaptive immune cells. Activated T lymphocytes upregulate the glutamine transporter, solute carrier family 7 member 5, leading to increased glutamine uptake and entry into the TCA cycle to support cellular function[72]. Glutamine catabolism further promotes glutathione synthesis, thereby influencing T-cell differentiation and function[73]. Similarly, B-cell proliferation and metabolism within the TME are highly dependent on glutamine. Together, these findings indicate that glutamine metabolism orchestrates both tumor-promoting and tumor-suppressing processes by differentially regulating cancer cells and immune cells. Disrupting this metabolic balance may therefore represent a promising therapeutic window in CCA, although tumor-specific investigations are still needed.

Methionine: An essential amino acid for normal cells and CCA

Methionine typically serves as a key metabolite that plays crucial roles in regulating epigenetics through chromatin methylation and nuclear function, mediated by polyamines. It is also needed to maintain the redox status in cells via glutathione synthesis. In mammals, methionine is converted into S-adenosylmethionine (SAM/AdoMet), catalyzed by methionine adenosyl transferase (MAT), which facilitates the transfer of an adenosine group from ATP to the sulfur in methionine[74]. SAM serves as the methyl group donor for methylation processes involving DNA, proteins, and lipids, generating S-adenosylhomocysteine (SAH)[75]. This compound subsequently undergoes hydrolysis, producing adenosine and homocysteine by SAH hydrolase[76]. The methionine cycle is completed when homocysteine is remethylated by homocysteine methyltransferase (HMT) with 5-methyltetrahydrofolate as a methyl donor, linking the folate cycle to the conversion of homocysteine back into methionine. This reaction requires vitamin B12 (cobalamin) as a coenzyme.

The methylation of DNA and histones by DNA methyltransferases and HMTs requires SAM as a cofactor[76]. Methionine adenosyltransferase 1 (MAT1) interacts with and modulates the function of the well-known oncoprotein, c-Myc, which is overexpressed in human CCA[77,78]. On the other hand, methionine adenosyltransferase alpha (MATα1) and c-Myc/Maf proteins reciprocally regulate each other at the promoter level. The MAT1A promoter is hypermethylated from c-Myc-mediated transcriptional suppression. In the murine CCA model, knocking down MAT1A increased tumor growth, invasion, and metastasis. In primary biliary cholangitis, an autoimmune cholangiopathy associated with iCCA and eCCA[79,80], treatment of SAM protects cholangiocytes via its antioxidant and S-glutathionylation properties, thereby improving liver biochemistry[81].

Methionine salvage pathways are universally used to regenerate methionine from 5′-methylthioadenosine[82], a sulfur-containing nucleoside produced from SAM primarily via the polyamine biosynthetic pathway. 5-methylthioadenosine phosphorylase (MTAP) is a key enzyme in the methionine salvage pathway, and the MTAP gene is deleted in several cancers. The deletion of the MTAP gene is hypothesized to result from its chromosomal proximity to the tumor suppressor gene CDKN2A, and both gene deletions are frequently found together in cancers[83]. A study across 390 cell lines suggested that the viability of MTAP-deficient cancer cells is impaired by the depletion of the protein arginine methyltransferase (PRMT5). Loss of MTAP leads to the accumulation of MTA, thereby creating a hypomorphic state of PRMT5, which selectively sensitizes cells to PRMT5 inhibition[83]. PRMT5-targeting drugs significantly inhibited CCA cell proliferation and, in combination with cisplatin and gemcitabine, synergistically suppressed the growth of CCA organoids. A molecular mechanism study suggested that the inhibition blunted the expression of oncogenic genes involved in chromatin remodeling and DNA repair, leading to the formation of RNA loops and promoting DNA damage[84]. These studies suggested that methionine not only serves as a precursor for energy metabolism but also participates in several pro-tumorigenic pathways in CCA cells. Roles of methionine metabolism in CCA cells are summarized in Figure 3.

Figure 3 Metabolic pathways and the alteration of methionine metabolism in cholangiocarcinoma cells, including the methionine cycle and the methionine salvage pathway. Loss of methylthioadenosine phosphorylase in the methionine salvage pathway leads to the accumulation of 5'-methylthioadenosine, thereby creating a hypomorphic state of protein arginine methyltransferase 5 (PRMT5), which selectively sensitizes cells to PRMT5 inhibition. The methionine adenosyltransferase 1 promoter is hypermethylated due to c-Myc-mediated transcriptional suppression, leading to increased tumor growth, invasion, and metastasis. The up or down direction of the arrows indicates whether levels in cholangiocarcinoma cells are increased or decreased. MTR: 5-methylthioribose; MTAP: Methylthioadenosine phosphorylase; MTA: 5'-methylthioadenosine; MAT1: Methionine adenosyltransferase 1; MS: Methionine synthase; AHCY: Adenosylhomocysteinase; MT: Methyltransferase; PRMT5: Protein arginine methyltransferase 5.

Tryptophan: Different metabolic pathways provide different effects on CCA

Tryptophan is an essential amino acid that undergoes metabolism through 3 primary pathways - approximately 95% through the kynurenine (Kyn) pathway, 1% through the 5-hydroxytryptamine (serotonin, 5-HT) pathway[85], and the remaining 4% through the indole pathway[86]. Kyn pathway (KP) is the main pathway for tryptophan metabolism that leads to nicotinamide adenine dinucleotide production. The process begins with the action of indoleamine-2,3-dioxygenase-1 and -2 (IDO-1 and IDO-2), as well as tryptophan 2,3-dioxygenase, which serve as initial rate-limiting enzymes responsible for converting L-tryptophan into N-formyl-L-Kyn. This intermediate is then processed by formamidase, producing L-Kyn[87]. Studies in several diseases suggest that the KP also plays a crucial role in immune system regulation, contributing to immunosuppression during infections or inflammatory responses. Indoleamine 2,3-dioxygenase 1 (IDO1) has a significant impact on immune cell modulation. It helps maintain a balance between immune activation and suppression at sites of localized inflammation[88]. Evidence for tumoral immune resistance revealed that IDO can deplete tryptophan, leading to G1 phase arrest and halting the proliferation of T-lymphocytes[89,90]. A pilot study in CCA examined databases of patients with CCA for expression of 19 immune checkpoints and evaluated their prognostic significance[91]. The study revealed that IDO1 expression was associated with poor overall survival. Additionally, IDO1 gene expression in patients with CCA was associated with advanced stages II and III[91]. Immunohistochemical microarray analysis in extrahepatic CCA revealed that IDO1 expression was associated with decreased numbers of CD8+ tumor-infiltrating lymphocytes, elevated pN category in TNM staging, overall stage, and recurrence[92]. However, the molecular mechanisms linking IDO1 and the progression of CCA remain to be elucidated. Targeting Kyn metabolism as a potential treatment strategy for CCA remains unexplored and warrants further investigation.

Serotonin, or 5-HT, is another metabolite produced from tryptophan, which is first converted to 5-hydroxy-L-tryptophan through a hydroxylation (oxidation) process, and then decarboxylated to form 5-HT[86]. The hydroxylation step is facilitated by tryptophan hydroxylase (TPH), a key enzyme in serotonin synthesis that regulates the rate of production. At the same time, the decarboxylation is carried out by aromatic-L-amino acid decarboxylase. TPH is predominantly found in the brain's raphe nuclei and in the enterochromaffin cells of the gut lining, which are the primary sites for serotonin synthesis[86]. A study of tumor tissues from patients with CCA revealed that TPH expression in CCA cells is statistically significantly higher than in normal liver samples[93]. In contrast, mRNA levels and protein expression of monoamine oxidase (MAO), a mitochondrial-bound enzyme responsible for 5-HT degradation, are significantly decreased in CCA cell lines compared with a non-malignant cholangiocyte cell line, accompanied by increased 5-HT secretion[93]. Treatment with 5-HT promotes the growth of CCA cell lines, whereas treatment with p-chlorophenylalanine, a tryptophan hydroxylase inhibitor, suppresses CCA growth in vivo. Moreover, the use of telotristat ethyl, a prodrug with specific inhibitory activity against TPH and blocking serotonin synthesis[94,95], combined with chemotherapy resulted in reduced serotonin levels, enhanced tumor growth inhibition in vivo, and increased animal survival. However, phase 2 of the clinical trial still did not address significant safety issues noted with telotristat ethyl in combination with gemcitabine plus cisplatin chemotherapy[95]. Therefore, although highly promising at targeting serotonin metabolism for cancer treatment, further comprehensive studies are needed for clinical translation to patients in the real world.

A serotonin degradation enzyme, monoamine oxidase A (MAOA)-another isoform of MAO with an X-linked gene[96] suppresses CCA cells' growth and invasion after forced overexpression[97]. Interestingly, the CpGI 28 promoter region of MAOA is hypermethylated in CCA samples via unclear mechanisms, but not in non-malignant areas. Transcription factor SP-1 serves as a major transcription factor binding to the MAOA core promoter and positively regulates expression and catalytic activity of MAOA[98], with R1 repressor 2 (RAM2)[99] competitively binding to the SP1 site on the core promoter of MAOA and leading to suppression of MAOA gene expression. The balance between SP-1 and RAM2 can modulate MAOA expression. The study suggests that interleukin-6 signaling mediates the balance between SP-1 and RAM2 by increasing RAM2 association with the MAOA promoter, thereby preventing SP-1 from accessing the core promoter of MAOA.

The last metabolic pathway of tryptophan in humans, which accounts for approximately 4%, does not occur in human cells but rather in gut microorganisms[100]. Most dietary tryptophan derived from protein digestion is absorbed in the small intestine; however, a substantial amount of tryptophan is still passed to the large intestine, where commensal bacteria metabolize the remainder[101]. From this metabolic pathway, bioactive indole compounds are generated, including indole-3-propionic acid, indole-3-acetic acid, indole-3-aldehyde, indole-3-lactic acid, and indole-3-acrylic acid; hence, this tryptophan metabolism process is referred to as the indole pathway[102]. The effects of indolic metabolites from intestinal bacteria have been reported in a few studies on colorectal cancer, with immunomodulatory and anti-tumor effects[103,104]. However, the underlying mechanisms have not been clarified in most studies. To date, several studies have demonstrated the potential effects of indoles in hepatobiliary diseases; however, there is a lack of clear evidence on how tryptophan-derived indoles and the indole metabolic pathway influence the progression of CCA[105-107]. Given that the biliary tract microbiota is a homologous community to the intestinal microbiome, it is hypothesized that the production of indole by biliary tract commensal microorganisms may be another factor influencing CCA development and progression, similar to the roles of the intestinal microbiome in colorectal cancer[108-110]. However, systematic studies, both preclinical and clinical, are needed to support this hypothesis and could inform therapeutic designs based on the effects of the resident microbiome. A summary of tryptophan metabolism and its impact on CCA cells is depicted in Figure 4.

Figure 4 Metabolisms of tryptophan and its roles in cholangiocarcinoma cells, including the kynurenine pathway and serotonin pathway. In the serotonin pathway, monoamine oxidase A (MAOA) expression and catalytic activity can be regulated by the transcription factor SP-1, which binds to the MAOA promoter. R1 repressor competitively binds to the SP-1 site on the MAOA core promoter, leading to suppression of MAOA gene expression. The up or down direction of the arrows indicates increased or decreased levels in cholangiocarcinoma cells. MAOA: Monoamine oxidase A; RAM2: R1 repressor; TPH1: Tryptophan hydroxylase; DDC: Dopa decarboxylase; IDO1: Indoleamine 2,3-dioxygenase 1.

Branched-chain amino acids: Essential amino acids for both CCA cells and the TME

Branched-chain amino acids (BCAAs), including valine, leucine, and isoleucine, are essential amino acids[111]. In mammals, BCAA aminotransferases (BCAT) initially catalyze the transamination of BCAAs to branched α-keto acids (BCKAs), with α-KG as the most common nitrogen receiver, thus generating glutamate[112]. BCAT has low hepatic activity; therefore, the initial step of BCAA catabolism occurs in skeletal muscle, unlike other amino acids, whose catabolism occurs in liver cells. Branched chain amino acid transaminase 2 (BCAT2), a member of the BCAT family, encodes a ubiquitously expressed enzyme that catalyzes BCAA metabolism[113]. According to expression analysis of BCAT2 in pan-cancer, the Cancer Genome Atlas and the Genotype-Tissue Expression databases, BCAT2 was found to be upregulated in CCA compared with normal tissues[114]. This implies that the requirement for BCAA is increased in CCA cells and that the primary site of BCAA metabolism might shift from myocytes to malignantly transformed cholangiocytes. Expression of BCAT2 also demonstrates its modulatory potential in regulating immune cell infiltration into the TME[115]. However, a correlation between subtypes of immune cells and BCAT2 levels is largely dependent on cancer type, and the underlying mechanisms regulating tumor immune microenvironments remain to be elucidated[116,117].

Another study in CCA identified upregulation of a pancreatic progenitor cell differentiation and proliferation factor (PPDPF)[118]. This molecule disrupts the interaction between methylcrotonoyl-CoA carboxylase subunit alpha (MCCA) and methylcrotonoyl-CoA carboxylase subunit beta (MCCB). A disruption between MCCA and MCCB consequently leads to inhibited leucine catabolism and activated mammalian target of rapamycin (mTORC1), thereby promoting CCA cell growth and proliferation. Under amino acid starvation, Ariadne RBR E3 ubiquitin protein ligase 2 and OTU deubiquitinase 4 regulate PPDPF stability by modulating its ubiquitination. Moreover, interleukin-10 derived from monocytes and macrophages increases BCAA levels in CCA cells and further stabilizes PPDPF. An in vivo study reveals that PPDPF knockout or dietary leucine restriction significantly suppresses CCA progression[118]. This study suggests that BCAA metabolism is not only essential for intracellular energetic metabolism but also plays a role in the communication among stromal cells in CCA TMEs, where various mechanisms promote cancer aggressiveness and progression. A summary of the BCAA metabolism in CCA cells is shown in Figure 5.

Figure 5 Branch chain amino acid metabolism and its alteration in cholangiocarcinoma cells. The level of pancreatic progenitor cell differentiation and proliferation factor (PPDPF) protein is stabilized by OTU deubiquitinase 4, and leucine catabolism is inhibited, leading to its accumulation in the cytosol. Upon amino acid starvation, the PPDPF protein is ubiquitously degraded upon binding to ariadne homolog 2. The up or down direction of the arrows indicates increased or decreased levels in cholangiocarcinoma cells. The dashed arrow refers to unclarified mechanisms. IL-10: Interleukin-10; IL-10R: Interleukin-10 receptor; ARIH2: Ariadne homolog 2; OTUD4: OTU deubiquitinase 4; PPDPF: Pancreatic progenitor cell differentiation and proliferation factor; BCAT2: Branched chain amino acid transaminase 2; BCKD: Branched chain ketoacid dehydrogenase kinase; MCC: Methylcrotonyl-CoA carboxylase; mTOR: Mammalian target of rapamycin; GLS1: Glutaminase 1; GDH: Glutamate dehydrogenase.

Urea cycle: A nitrogen disposal pathway supporting CCA tumor growth

The urea cycle is the primary pathway for disposing of nitrogen in humans[98], which begins in the mitochondria of hepatocytes and terminates in the cytoplasm. The first step is a rate-limiting step that incorporates carbon monoxide and ammonia into carbamoyl phosphate by the enzyme carbamoyl phosphate synthetase I (CPS1), followed by the combination of carbamoyl phosphate and ornithine to generate citrulline by ornithine transcarbamoylase (OTC). The OTC is then transported into the cytoplasm, where it reacts with aspartate to form argininosuccinate, which is then converted into arginine by argininosuccinate lyase. Arginine undergoes hydrolysis by arginase to finally form urea and ornithine. Urea, as a nitrogen waste, is excreted from the body by the kidneys in urine. Dysregulation and reprogramming of the urea cycle are also reported in cancer cells[119-121]. Downregulation of CPS1, a rate-limiting enzyme that catalyzes the first reaction of the urea cycle, was significantly associated with higher pT status in TNM staging of iCCA and was an independent prognosticator predicting lower survival in patients[122]. However, co-expression of CPS1 and its long noncoding RNA, the CPS1 intronic transcript 1, was found to be associated with poor liver function and a poor prognosis in patients with intrahepatic CCA[123]. Levels of argininosuccinate synthetase, a key enzyme in the urea cycle, were also significantly lower in CCA cells than in non-tumor cells[124]. The in vitro study, depletion of extracellular arginine by pegylated arginine deiminase (ADI-PEG20) reduced cell proliferation and enhanced G0/G1 cell cycle arrest in CCA cell lines[124]. Although highly promising in vitro experiments, clinical investigations of ADI-PEG20 in CCA are still lacking. Early clinical trials in other cancers, such as melanoma, sarcoma, hepatocellular carcinoma, and biliary tract cancer, reported safety profiles and adverse effects of this agent[125,126]. However, phase II and III clinical trials in hepatocellular carcinoma did not demonstrate satisfactory therapeutic efficacy of ADI-PEG20[127,128]. In addition, using ADI-PEG20 can lead to intolerable immunological reactions in patients[125,129]. The immunogenicity of ADI-PEG20 is another concern that needs further investigation to reduce adverse effects while maintaining drug efficacy. The prognostic value of arginase-1 (Arg-1) in CCA was also explored, despite the lack of clarity regarding its underlying mechanism. Immunohistochemistry results reveal that high Arg-1 expression was associated with increased tumor size and shorter overall and disease-free survival in patients with iCCA[130]. These studies suggested that, although arginine is classified as a non-essential amino acid in normal tissues, it can become essential when cells become malignant. This potential metabolic vulnerability of CCA cells warrants further research to identify anti-metabolic targets for therapeutic development. A summary of the urea cycle and the alterations of its enzymes in CCA cells is shown in Figure 6.

Figure 6 The alteration of the urea cycle’s enzymes in cholangiocarcinoma progression. Alteration of enzymes in the urea cycle is associated with aggressive phenotypes of cholangiocarcinoma (CCA) cells and poor clinicopathological characteristics of patients. The up or down direction of the arrows indicates increased or decreased levels in CCA cells. CPS1: Carbamoyl phosphate synthetase 1; OTC: Ornithine transcarbamylase; ASS: Argininosuccinate synthetase; ASL: Argininosuccinate lyase; ARG: Arginase.

CHALLENGES FOR THE FUTURE DIRECTION OF CLINICAL TRANSLATION IN CCA

Despite a challenge and a greater theoretical limitation of inhibiting amino acid metabolism as a target of cancer treatment, several studies have investigated the potential of anti-metabolite agents, especially glutamine metabolism inhibitors, in clinical trials. Apart from telaglenastat, one of the glutaminase 1 inhibitors with the most advanced clinical studies, the trial results from other developed drugs have not yet been satisfactory and were terminated due to intolerable adverse effects. The glutamine antagonist, 6-diazo-5-oxo-L-norleucine (DON), demonstrated promising antitumor activity by inhibiting multiple glutamine-dependent enzymes in preclinical studies, but its clinical development was halted due to dose-limiting gastrointestinal toxicity[131]. Recent efforts to revive DON through tumor-targeted prodrugs aim to preferentially release DON at tumor sites or in protected compartments, such as the central nervous system, thereby minimizing gastrointestinal exposure. When administered at low concentrations but at metabolically effective doses, optimized DON prodrugs achieve sustained glutamine pathway inhibition with markedly improved tolerability, supporting renewed clinical interest in glutamine antagonism[132]. Another example is the selective GLS1 inhibitor bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide (BPTES), whose clinical application has been limited by poor solubility[133] and unfavorable pharmacokinetics[134]. Innovative delivery approaches, such as a biomimetic, erythrocyte membrane-coated nanosystem (R-CM@MSN@BC) may overcome these barriers. This system enables controlled, reactive oxygen species-responsive release of BPTES and the photosensitizer Ce6 in combination with photodynamic therapy, resulting not only in direct tumor cell necroptosis but also in immunogenic cell death and remodeling of the immunosuppressive TME[135]. These examples illustrate that successful clinical translation of glutamine-targeted therapies depends on targeted drug delivery, rational combination strategies, and biomarker-driven trial designs, rather than further development of first-generation inhibitors alone. Nevertheless, because CCA is relatively rare, most clinical studies are initially conducted in other cancers. The safety and efficacy of using these agents in patients remain limited and may require additional information from other cancers before initiating trials in CCA and other biliary tract malignancies.

Reprogramming of amino acids in CCA is another potential avenue to investigate for cancer cells’ vulnerability and as a therapeutic target[136]. The currently available comprehensive view of amino acid alterations is summarized in Figure 7. However, several challenges arise in the research and development of therapeutic strategies targeting amino acids. First, amino acids are generally not a primary source of energy for metabolism, although some studies have shown that CCA is highly dependent on glutamine. Targeting amino acid metabolism may be helpful initially, but cancer cells may then undergo metabolic adaptation. Second, the enzymes involved in most amino acid metabolism in cancer cells are not cancer-specific isoforms and are also expressed in normal tissues. This could result in significant adverse effects after implementation in a clinical study, leading to the early termination of the trial. To address this concern, a delivery system that targets cancer cells precisely would help minimize side effects[137,138]. Third, CCA is a highly heterogeneous malignancy. Different cancer etiologies can give rise to distinct metabolic profiles, and molecular testing of the metabolic background of cancer cells in individual patients would be needed[139,140]. Fourth, comprehensive mathematical modeling approaches, including flux balance analysis, dynamic isotope tracing, and quantitative frameworks to estimate multi-pathway compensatory fluxes of amino acids, have not yet been elucidated in most cancers. At present, most studies focus on static measurements of gene expression or metabolite abundance, which are insufficient for modeling dynamic metabolic rewiring or ranking synergistic drug combinations. Finally, amino acids are essential nutrients required for protein synthesis to support the normal functions of tissues. Patients with cancer usually need protein supplementation to prevent frailty and prepare their health for intensive treatments. Therefore, simply restricting protein intake or the amino acid composition might not be as straightforward as controlling glucose intake or glycemic levels[138]. These challenges remain to be addressed and need further investigation, along with studies to develop treatment strategies targeting amino acid metabolism in cancer, including CCA.

Figure 7 A comprehensive view of amino acid metabolism alteration in cholangiocarcinoma. The up or down direction of the arrows indicates increased or decreased levels in cholangiocarcinoma cells. The black arrows denote the pro-tumor effects of the genes or intermediates, while the white arrow denotes their anti-tumor effects. AGC: Aspartate-glutamate carrier; GLS1: Glutaminase 1; GDH: Glutamate dehydrogenase; GOT: Glutamic-oxaloacetic transaminase; α-KG: Alpha-ketoglutarate; ANXA1: Annexin A1; USP5: Ubiquitin-specific protease 5; GAD1: Glutamate decarboxylase 1; GABA: γ-aminobutyric acid; cAMP: Cyclic adenosine monophosphate; IP3: Inositol trisphosphate; PKA: Protein kinase A; ERK: Extracellular signal-regulated kinase; GSK3: Glycogen synthase kinase-3; MTR: 5-Methylthioribose; MTAP: Methylthioadenosine phosphorylase; MTA: 5'-Methylthioadenosine; MAT1: Methionine adenosyltransferase 1; MS: Methionine synthase; AHCY: Adenosylhomocysteinase; MT: Methyltransferase; PRMT5: Protein arginine methyltransferase 5; MAOA: Monoamine oxidase A; TPH1: Tryptophan hydroxylase; DDC: Dopa decarboxylase; IDO1: Indoleamine 2,3-dioxygenase 1; ARIH2: Ariadne homolog 2; OTUD4: OTU deubiquitinase 4; PPDPF: Pancreatic progenitor cell differentiation and proliferation factor; BCAT2: Branched chain amino acid transaminase 2; BCKD: Branched chain ketoacid dehydrogenase kinase; MCC: Methylcrotonyl-CoA carboxylase; mTOR: Mammalian target of rapamycin; CPS1: Carbamoyl phosphate synthetase 1; OTC: Ornithine transcarbamylase; ASS: Argininosuccinate synthetase; ASL: Argininosuccinate lyase; ARG: Arginase.

CONCLUSION

CCA is a cancer with high metabolic adaptation, and its amino acid metabolism is significantly changed to meet the requirements of its malignant progression. Reprogramming of amino acids in CCA provides several vulnerabilities for developing therapeutic strategies targeting cancer cells. However, several challenges and limitations remain to be investigated. Comprehensive studies, not only focusing on the main metabolic pathways but also on associated metabolic networks, are needed to develop effective targeted therapy for aberrant amino acid metabolism with minimal adverse effects.

ACKNOWLEDGEMENTS

We would like to thank Professor John F Smith for editing this manuscript via the KKU Publication Clinic, Khon Kaen University, Thailand (PCO-1316).