An integrative view on glucagon function and putative role in the progression of pancreatic neuroendocrine tumours (pNETs) and hepatocellular carcinomas (HCC)

higher levels of glucagon, pancreas as a producer and liver as a scavenger. The main objective is to bring to discussion some glucagon-dependent mechanisms by presenting an integrative view on microenvironmental and systemic aspects in pNETs and HCC biology.


Introduction
Cancer metabolism research area has evolved greatly in the past two decades (Serpa, 2020).However, little is known about the impact of systemic metabolism control and diet on cancer.Cancer cells face different stressful conditions during carcinogenesis and disease progression; in the first place, they must be adapted to the organ microenvironment, which upon tumour growth will be modulated towards a tumour microenvironment.Hence, the tumour and organ microenvironment (TOME) must be considered.The regulators of systemic metabolism have a cell-directed action on organs that physiologically benefit from their command dependent on the activation of intracellular signalling.In the pathophysiology of cancer, the action of systemic modulators can occur directly on cancer cells and activate signalling pathways useful to orchestrate the metabolic adaptation needed for cancer cells to resist to TOME and thrive.The most well studied systemic metabolic regulator is insulin, mainly in the context of diabetes.Nevertheless, even in diabetes, glycaemic control does not depend solely on insulin; glucagon, the hypoglycaemia-acting hormone is also fundamental for maintaining glucose systemic balance (Rix et al., 2000).Insulin and glucagon are presented as having an opposite and competitive role on glycaemia control, involving the function of the endocrine pancreatic component and the liver (Fig. 1 A), however it is demonstrated that an interplay between these two hormones is an important control step on each other secretion and consequent function (Wewer Albrechtsen et al., 2017;Quesada et al., 2008a;Zhang et al., 2021).
β-cells of the pancreatic Langerhans islets produce insulin in response to hyperglycaemia, which, in turn, activates the import of glucose by the hepatocytes and increases glucose consumption and storage in the liver, promoting glycolysis and glycogen synthesis, respectively (Rix et al., 2000;Svendsen et al., 2018).Conversely, glucagon is produced by α-cells of the pancreatic Langerhans islets in response to hypoglycaemic, and it promotes the release of glucose into the bloodstream by activating glycogen degradation and gluconeogenesis (Rix et al., 2000;Svendsen et al., 2018).Therefore, while the role of glucagon in glucose homeostasis in the human body is well understood, its specific impact at the cellular level, particularly in cancer cells, remains to be explored.Albeit insulin impact has been addressed in different cancer models, glucagon and hyperglucagonaemia (>100 pmol/mL) impact in TOME and their consequent effects on cancer metabolic remodelling is not well understood.Glucagon has a natural receptor called the glucagon receptor (GCGR) (Zhang et al., 2017), but it can also interact and activate the glucagon-like peptide 1 (GLP-1) receptor (GLP-1R) (Svendsen et al., 2018).Importantly, GLP-1 is a repressor of glucagon exocytosis by α-cells (Rix et al., 2000), thus the binding of glucagon to GLP-1R may subvert the system and contribute for hyperglucagonaemia and the activation of signalling pathways that may favour cancer progression.The subversive loop depends on the glucagon levels in the TOME, the mutational profile of GCGR, and the expression profile of GCGR and GLP-1R.
In this review, we will depict the influence of glucagon on cancer cells functioning and its contribution to disease progression, by reviewing two poor prognosis cancer models (Fig. 1 B).The first model is an adaptive model that focuses on pancreatic neuroendocrine tumours (pNETs), as α-cells hyperplasia, an adaptive phenomenon, is considered a precursor lesion prompted by CGCR mutations and contributes to hyperglucagonemia.This phenomenon is proved in murine models (Lawrence et al., 2011) - (Gelling et al., 2003) in humans, albeit a mechanistic link is needed, some papers claim that α-cells hyperplasia must be considered and explored as a pNET precursor lesion (Smith et al., 2020a;Yu et al., 2011).The association is stronger in patients with familial endocrine alteration; however, some sporadic subsets seem to present a trend between α-cells hyperplasia and pNETs (Ouyang, 2011).The putative link between α-cells hyperplasia and pNETs was extensively reviewed integrating human and models data, as mentioned by Yu et al. (Yu, 2015) , (Yu, 2014).The second model is a responsive model, the hepatocellular carcinoma (HCC), which may benefit from the glucagon-rich TOME, as the liver is the main controller of glucagon levels in the bloodstream.Besides this, the incidence of pNETs and HCC is associated with diabetes (Yang et al., 2020;Hernandez-Rienda et al., 2022), which is a systemic metabolic condition presenting hyperglucagonaemia as a feature (Godoy-Matos, 2014).
The main objective is to bring to discussion some glucagondependent mechanisms by presenting an integrative view on microenvironmental and systemic aspects in pNETs and HCC biology.This way we would like to contribute to fill a gap that still exists between cancer cell and systemic metabolic mechanisms and control.

Glucagon regulation of systemic metabolism
Glucagon is synthesized in the α-cells of the pancreas as a proglucagon polypeptide (Fig. 2 A), which plays a crucial role in regulating glucose homeostasis.The post-translational processing of proglucagon into active glucagon involves the removal of specific amino acid (AA) Fig. 1.Glucagon function in hypoglycaemia and its putative role in pancreatic neuroendocrine tumours (PNETs) and in hepatocellular carcinoma (HCC).(A) Glucagon and insulin are the main controller of glucose levels in the peripheral blood (PB).Hypoglycaemia stimulates the production of glucagon by α-cells in the pancreatic Langerhans islets.Glucagon acts on the liver and stimulates the release of glucose to PB; glucose is synthesized in hepatocytes through gluconeogenesis and also results from glycogenolysis.Hyperglycaemia stimulates the production of insulin by β-cells in the pancreatic Langerhans islets.Insulin acts on the liver and stimulates the uptake and consumption of glucose by hepatocytes, glycolysis and glycogenogenesis are activated.(B) Glucagon is a metabolic controller, though it may act directly on cancer cells and activate the metabolic remodelling needed to sustain cancer progression.Pancreas and liver are the main organs responsible for the bioavailability of glucagon, respectively as a producer and as a scavenger.Therefore, the pancreatic neuroendocrine tumours (PNETs) and the hepatocellular carcinoma (HCC) may benefit from the high concentrations of glucagon in pancreas and liver, and this hypothesis prompted the definition of two cancer models: the adaptive model focused on PNETs since, α-cells hyperplasia, an adaptive phenomenon, is a precursor lesion prompted by CGCR mutations and contributes for hyperglucagonaemia; and the responsive model, HCC, which may benefit from the glucagon-rich TOME as the liver is the main controller of glucagon levels in the blood stream.
An extensive capillary network throughout the islet enables quick detection of shifts in plasma nutrients and signalling molecules, activating subsequent secretory responses (Cabrera et al., 2006).Furthermore, α-cells are precisely regulated by a highly innervated sympathetic and parasympathetic system that extends into the perivascular spaces.This neuronal network is crucial for rapid islet secretory responses, especially under hypoglycaemic circumstances, where α-cells are stimulated to release glucagon, which then activates hepatic gluconeogenesis and glycogenolysis, leading to an increase in blood glucose levels (Ahrén, 2000;Taborsky and Mundinger, 2012).Thus, nutrients are a crucial factor among the multiple signalling levels that modulate the secretory capability of islet cells.
The GCGR is a G-protein-coupled receptor (GPCR), which is a class of over 1000 proteins regulating most aspects of cellular behaviour (de Graaf et al., 2017).GPCRs are classified into six classes based on shared sequence motifs and functional similarity, with Class B (secretin receptor family) binding peptide hormones such as calcitonin, parathyroid hormone, glucagon, and GLP-1 (de Graaf et al., 2017).
The primary function of glucagon in the control of glycaemia is to increase hepatic glucose production (Raskin et al., 1978).The liver is a central organ in glucose homeostasis, as it is responsible for regulating glucose production and release into the bloodstream (Sharabi et al., 2015).
One of the primary effects of glucagon in the liver is the stimulation of glycogenolysis.In addition to this, glucagon promotes gluconeogenesis, which produces de novo glucose from non-sugar sources (Raskin et al., 1978).Overall, glucagon plays a critical role in regulating hepatic metabolism, promoting the production and release of glucose and ketone bodies when energy demands are high, and glucose levels are low.Therefore, glucagon helps to maintain glucose homeostasis and ensures that cells throughout the body have access to the energy and biomass they need to function properly (Jiang and Zhang, 2003).

Regulation of glucose homeostasis
The fall of blood glucose levels below a certain threshold, such as during fasting or exercise, promotes glucagon secretion by the pancreas to restore blood glucose levels (Colberg, 2022;Quesada et al., 2008b).As mentioned earlier, the primary function of glucagon is to stimulate the liver to break down glycogen and release glucose into the bloodstream (Jiang and Zhang, 2003).Glucagon also promotes gluconeogenesis the production of glucose from non-carbohydrate sources such as AAs and fatty acids (FAs) (Quesada et al., 2008b;Cynober, 2002).Gluconeogenesis mainly occurs in the liver and helps to maintain blood glucose levels during prolonged periods of fasting (Adeva-Andany et al., 2019).In addition to these functions, glucagon inhibits insulin secretion from the pancreas, reducing glycogenogenesis and glycolysis, which helps prevent hypoglycaemia (Jiang and Zhang, 2003).

Regulation of glycogenolysis
Glycogen storage is another mean of controlling blood glucose levels, and glucagon plays a role in inhibiting glycogen synthesis (glycogenogenesis) and activating glycogenolysis, by acting on glycogen synthase (GS) and glycogen phosphorylase (GP), respectively.Glucagon is required for the phosphorylation and activation of GP, and it suppresses GS activity by inducing the enzyme phosphorylation, converting it into an inactive state (ANDERSEN et al., 1999).
Glucagon rapidly mobilizes hepatic glycogen reserves in response to short-term starvation, leading to a rapid increase in hepatic glucose production (Jiang and Zhang, 2003).This is achieved through glucagon signalling via PKA, which phosphorylates and activates glycogen phosphorylase kinase (GPK), resulting in the phosphorylation of GP and glycogen breakdown (Habegger et al., 2010).Glucagon has also been implicated in decreasing the acetylation of hepatic glycogen phosphorylase (GP) contributing to its activition (Zhang et al., 2012).In addition to promoting glycogen breakdown, glucagon inhibits GS activity, leading to an overall increase in glycogenolysis (Petersen et al., 2017).Considering these effects, glucagon is an essential counter-regulatory hormone during hypoglycaemia (Sprague and Arbeláez, 2011).

Regulation of gluconeogenesis
Glucagon promotes gluconeogenesis once glycogen reserves are depleted during prolonged fasting to enhance hepatic glucose production and maintain blood glucose levels (Petersen et al., 2017).This is accomplished by allosterically adjusting the activity of multiple enzymes, shifting the metabolic flow from glycolysis to gluconeogenesis Fig. 2. Glucagon production and glucagon-dependent signalling and metabolic pathways.(A) In humans, the processing of proglucagon to glucagon involves posttranslational processing in the α-cells by the prohormone convertase 2 (PC2) releasing glicentin related pancreatic polypeptide (GRPP), glucagon, intervening peptide-1 (IP-1) and the major proglucagon fragment (MPF), which can be cleaved by prohormone convertase 1 and 3 (PC1/3) leading to GLP-1 release.In the enteroendocrine L-cells and in the central nervous system, the post-translational processing of proglucagon is mediated by PC1/3 and releases glicentin, oxyntomodulin, GLP-1, IP-2 and GLP-2.(B) The glucagon receptor (GCGR) is a class B G-protein-coupled receptor (GPCR).Upon glucagon action, the activation of the GCGR interferes with cellular functioning by activating the production of cAMP and related signalling pathways, namely PKA, phosphatidylinositol 3-kinase (Pi3K) and mitogen-activated protein kinase (MAPK) pathways.These pathways contribute for the activation of nuclear regulators such as Cyclic AMP-Responsive Element-Binding Proteins (CREB) and Extracellular Signal-Regulated Kinases (ERK).Glucagon can also activate the phospholipase C (PLC)/inositol phosphate pathway via Gq proteins, leading to Ca 2+ release from intracellular reserves and activation of nuclear CREB-Regulated Transcription Coactivator 2 (CRTC2).(C) The main metabolic pathways related to glucagon action are gluconeogenesis to supply the energetic and biomass needs for glucose; protein degradation to release amino acids that will be used in both bioenergetics and biosynthesis, and fatty acids degradation (β-oxidation).Therefore, the control of gene expression by glucagon is fundamental, including the regulation of genes encoding transporters, such as monocarboxylate transporters (MCT), glucose transporters (GLUT) and sodium-glucose transporters (SGLT).

Regulation of amino acids (AAs) metabolism
Glucagon increases the import of AA by hepatocytes, supplying AA substrates for gluconeogenesis (Almdal et al., 1992).This is achieved by glucagon-stimulated expression of AA transporters for alanine, glutamine, histidine, and asparagine in the liver, resulting in enhanced AA absorption (Lim, 1999;KILBERG et al., 1985).Once inside the hepatocytes, these AAs are further processed to be used as precursors for gluconeogenesis.Deamination is required, after which the amine groups enter the urea cycle for excretion (Schutz, 2011).Glucagon induces rapid deamination of glutamine, resulting in an immediate increase in the ureagenesis and AA metabolism (Lacey et al., 1981).
Furthermore, glucagon increases the mRNA levels of enzymes involved in the urea cycle through the cAMP-PKA-CREB-pathway (JanahKjeldsenGalsgaard et al., 2019).Glucagon specifically increases the transcription of the enzyme N-acetyl glutamate synthetase (NAGS), directing AA flow towards ureagenesis (JanahKjeldsenGalsgaard et al., 2019).
Overall, glucagon prepares hepatocytes for the absorption of AAs from the circulation, which are used as precursors of gluconeogenesis during prolonged fasting (Longuet et al., 2008a).Antagonism of the GCGR leads to hyper-aminoacidemia due to reduced absorption of circulating AAs and impaired ureagenesis, highlighting the relevance of hepatic glucagon signalling in AA metabolism (Dallas-Yang et al., 2004;Zeigerer et al., 2021).Hyper-aminoacidemia has been proposed to play a role in the increase in α-cell mass (Galsgaard et al., 2018;Gong et al., 2023;Hayashi and Seino, 2018).Notably, the increase in circulating AAs after GCGR ablation stimulates glucagon release from α-cells, resulting in a vicious loop of glucagon overproduction (Wewer Albrechtsen et al., 2019).This helps explain the hyperglucagonaemia found following GCGR signalling ablation in the liver.Arginine, alanine, and proline have been shown to promote glucagon release from α-cells (Galsgaard et al., 2020), while glutamine induces α-cell hyperplasia (Dean et al., 2019).Furthermore, the pancreatic amino-acid transporter SNAT5 (encoded by SLC38A5 gene) has been found to be essential in α-cell hyperplasia induced by hepatic GCGR inhibition in mice, and its absence inhibits the development of hyperplasia.
These findings highlight the need for further characterization of the role of glucagon in hepatic AA metabolism and underlying processes that regulate the glucagon-GCGR axis.

Regulation of fatty acids (FAs) metabolism and ketogenesis
Glucagon plays a crucial role in regulating FA metabolism and ketogenesis (Briant et al., 2018).When secreted by α-cells, it modulates FA mobilization by acting on adipose tissue, stimulating hepatic lipolysis and FA oxidation (Pégorier et al., 1989;Perry et al., 2020) while decreasing lipogenesis (Prip-Buus et al., 1990) and the secretion of triglyceride and very low-density lipoprotein (Longuet et al., 2008b;Galsgaard, 2020).In FA metabolism, cells utilize FAs to produce energy for physiological processes.This process involves the hydrolysis of stored triglycerides to produce free FAs, which are then transformed into acetyl-CoA through β-oxidation inside mitochondria (Langin et al., 2005;Peters et al., 2016;Gormsen et al., 2017).Glucagon stimulates hormone-sensitive lipases (HSL), enhancing lipolysis by breaking down triglycerides into glycerol and free FAs that circulate throughout the body (Langin et al., 2005;Peters et al., 2016).
Furthermore, when glucose availability is depleted during fasting or carbohydrate restriction, glucagon stimulates the liver to produce ketones from FAs, primarily β-hydroxybutyrate and acetoacetic acid, through the process of ketogenesis (Ullah et al., 2016;Capozzi et al., 2020).
The primary pathway influenced by glucagon induced GCGR activation is the PKA pathway, which plays a crucial role in cellular signalling.Additionally, glucagon can activate the phospholipase C (PLC)/ inositol phosphate pathway via Gq proteins, leading to the release of Ca 2+ from intracellular stores (Denis et al., 2012).Furthermore, glucagon has been associated with signalling through the MAPK pathway, which is a complex cellular signalling pathway involved in various biological processes, such as cell proliferation, differentiation, migration, survival, and apoptosis (ZHANG and LIU, 2002).The MAPK pathway is activated by extracellular signals, such as growth factors, cytokines, and hormones, and it involves a cascade of phosphorylation events that activate a family of protein kinases known as MAP kinases, including RAF, MEK, and ERK, leading to the activation of downstream targets such as transcription factors, cytoskeletal proteins, and other signalling molecules (ZHANG and LIU, 2002).
The MAPK pathway can be activated not only by stimulation of receptor tyrosine kinases (RTKs) but also by G protein-coupled receptors (GPCRs) like GCGR through multiple intricate cascades.These cascades can involve the induction of RTK phosphorylation, leading to ERK activation (Pierce et al., 2001).In addition to RTK activation, GPCR-stimulated MAPK cascades can also be activated through integrin-based scaffold and the β-arrestin scaffolds (Winston and Hunter, 1996).The activated ERK then translocates to the nucleus and transactivates transcription factors, resulting in changes in gene expression that modulate proliferation by activating differentiation or mitosis (Pierce et al., 2001).There is growing evidence of extensive crosstalk between the PI3K and MAPK pathways, which can occur in a GTP-dependent manner, with PI3K interacting with the Ras GDP/GTP exchange protein (Cuesta et al., 2021).
Beyond its physiological functions, these signalling pathways B. Ferreira et al. regulated by glucagon have also been implicated in carcinogenesis and cancer progression, suggesting that glucagon-dependent signalling may play a role in certain types of cancer, particularly those related to glucagon circuitries such as pNETs and HCC.Further research is needed to fully understand the impact of glucagon signalling on cancer development and progression.

Pancreatic neuroendocrine tumours (pNETs)
Gastroenteropancreatic neuroendocrine tumours (GEP-NETs) are a rare and heterogeneous group of neoplasms that arise from multipotent stem cells in the pancreatic epithelia and the gastrointestinal system (Yalcin et al., 2017;Modlin et al., 2008).pNETs are a subtype of GEP-NETs, which derive from pancreatic endocrine cells accounting for approximately 12% of all GEP-NETs (Cives and Strosberg, 2018).
The incidence of pNETs has increased from 1 in 100,000 to 5 in 100,000 people worldwide, most likely because of improved diagnosis procedures (Das and Dasari, 2021;Haugvik et al., 2012).pNETs are classified into functional and non-functional.Functional pNETs (15%) produce hormones such as insulin, gastrin, or glucagon, leading to specific clinical symptoms associated with hormone hypersecretion.Non-functional pNETs (85%) do not produce hormones or cause related symptoms (Das and Dasari, 2021;Haugvik et al., 2012).However, it is worth noting that the actual prevalence of pNETs may be underestimated, as autopsy studies have revealed the presence of pNETs in approximately 10% of individuals, suggesting that a significant number of tumours may go undetected and undiagnosed (Halfdanarson et al., 2008;Kimura et al., 1991).Non-functional NETs are frequently detected at later stages and are often less well differentiated and are associated with a higher risk of metastasis (Zhang et al., 2013;Bilimoria et al., 2007).In contrast, patients with functional pNETs are often diagnosed at early stages due to the severity of symptoms caused by hormone overproduction.Thus, the pNETs five-year survival rate in non-metastatic tumours is relatively high at 85-95%, but in patients with unresectable disease and liver metastases the five-year survival rate drops to 30-40%, with a median survival of 24 months.Overall, pNETs patients have a poor prognosis, emphasising the critical need for knowing better pNETs biology and develop new and more effective therapies (Lawrence et al., 2011;Frilling et al., 2010;Meeker and Heaphy, 2014;Öberg et al., 2012).

Diagnosis and therapies
The workup of an unexpected pancreatic mass or symptoms compatible with a pNET is centred on reaching a diagnosis and determining the functionality and extent of the disease.Cross-sectional imaging, functional imaging, invasive imaging, such as endoscopic ultrasound (EUS) or arterial stimulation venous sampling (for functional pNETs), and tumour markers testing are used to achieve these goals (Lee et al., 2017;Ito et al., 2012).
Surgical excision is the only known cure for functional or nonfunctional pNETs, however, most patients are diagnosed at advanced stages.Tumour functionality, grade, and stage are all factors that influence which therapy choices are available to patients with pNETs, and there are several therapeutic options for advanced pNETs, but the efficacy is still low (Dumlu et al., 2015;Li et al., 2022a;ZIOGAS et al., 2022).
Many patients with advanced, recurring, or metastatic pNETs are not surgical candidates and require pharmacological therapy, which is applied to reduce excess hormones and symptoms, but also to limit tumour growth.In patients with unresectable functional pNETs, the primary focus is hormonal management, as hormone overproduction can cause severe morbidity and death ( Öberg et al., 2010).

Molecular alterations-genetics and epigenetics
pNETs can arise in the context of hereditary cancer syndromes or as sporadic cases.Inherited cancer susceptibility syndromes, such as MEN1, VHL, and, less frequently, TSC, are associated with a 10-15% portion of pNET cases (Geurts, 2020).
Multiple Endocrine Neoplasia Type 1 (MEN1) inherited germline mutation underlies an autosomal dominant cancer susceptibility syndrome with an estimated incidence of 0.25% (Thakker et al., 2012).The occurrence of hyperplasia, GEP-NETs, and anterior pituitary adenomas defines MEN1 syndrome.As a classical tumour suppressor gene, the full inactivation of MEN1 depends on a second hit, which is most commonly the loss of heterozygosity, and the loss of chromosome 11 heterozygosity (over the MEN1 locus) has been observed in 38.6% of non-functional pNETs and 15-20% of gastrinomas and insulinomas (Jiao et al., 2011a).
The von Hippel Lindau (vHL) syndrome is an autosomal dominant hereditary cancer syndrome caused by germline mutations in the VHL gene and is associated with benign and malignant tumour forms (Findeis-Hosey et al., 2016).pNETs are found in 9-17% of patients with vHL (Ahmad et al., 2021;Blansfield et al., 2007).pNETs connected with vHL are non-functional, and notably, patients with vHL frequently have numerous pancreatic cysts and tumours and present a differential expression of genes involved in angiogenesis and hypoxia-inducible factor signalling (Speisky et al., 2012).
Tuberous sclerosis complex (TSC) syndrome is another autosomal dominant disorder that affects 1 in every 6-10 thousand people.pNETs in TSC are highly uncommon, with only a few examples documented in the literature, with the majority having TSC2 mutations when examined.Tumours are usually well differentiated and can be functional (Northrup et al., 2013).
Epigenetic alterations, such as DNA methylation (Plass et al., 2013) and histone modifications, have also been shown to be important in the development of sporadic pNETs (Tirosh and Kebebew, 2020).Hypermethylation of CpG islands in the promoter regions of tumour suppressor genes, such as cyclin-dependent kinase inhibitor 2A (CDKN2A), B. Ferreira et al. is frequently observed in sporadic pNETs (Tirosh and Kebebew, 2020;Serrano et al., 2000).In addition, alterations in histone modifications, such as the loss of methylation of histone H3 on lysine 27 (H3K27) trimethylation, have been observed in sporadic pNETs and are thought to be associated with aberrant modulation of gene expression (Cejas et al., 2019).
It is important to note that the frequency of genetic and epigenetic alterations in sporadic pNETs can vary depending on the subtype and stage of the tumour (Tirosh and Kebebew, 2020).Therefore, a comprehensive genetic and epigenetic analysis of pNETs cases is necessary to understand the underlying molecular mechanisms and identify potential therapeutic targets.

Hepatocellular carcinoma (HCC)
The most common type of liver cancer is HCC.Having more than 700,000 cases per year, it is one of the major causes of morbidity and mortality worldwide with higher incidence in developing countries, over 80% of these cancers occur in such regions (Ferlay et al., 2008).It is well known that the most major risk factors for the development of HCC include the cirrhosis and the chronic liver disease.These comorbidities can be explained by environmental factors such as viral hepatitis, hepatitis B (HBV) and hepatitis C (HCV) viruses, or alcohol intake (Xu et al., 2021).It has also been reported that the risk of HCC in men is up to 2-7 times higher than women (McGlynn et al., 2001;Parkin et al., 2005), might be taken as explanations the exposure to the referred environmental factors, the anti-inflammatory effect of oestrogen in women via the inhibition of interleukin-6 (Il-6) (Naugler et al., 2007) and the production of testosterone in men increasing androgen receptor signalling and therefore promoting liver cell proliferation (Yu and Chen, 1993).Other risk factors could be the exposure to hepatocarcinogens such as aflatoxin or oral contraceptives (Maheshwari et al., 2007;Liu and Wu, 2010;Chen et al., 2013), increasing the risk of HCC depending on the dose and exposure duration, and metabolic and genetic diseases (Balogh et al., 2016a;Ko et al., 2007), including hemochromatosis, Wilson's disease, α-1 antitrypsin disease, tyrosinemia, glycogen-storage disease types I and II, diabetes and porphyrias, are associated with an increased risk of HCC.A recent meta-analysis has shown that smoking is also associated with a higher risk of developing HCC (Gandini et al., 2008).

Classification and staging
HCC can be classified depending on the genome alterations, pathway signalling activation and other tumour-associated characteristics that will be further described.Even it is possible to measure all these factors, HCC is a very heterogeneous disease turning difficult to provide an accurate prognosis (Morris et al., 2021).Therefore, many staging and scoring systems have been proposed over time for patients with this disease (Maida, 2014).The Okuda staging system (Okuda et al., 1984) combines anatomical features of the tumour, such as size, with the evaluation of liver function.Parameters such as albumin, bilirubin, and the presence of ascites are measured to determine the stage of disease.However, with advancements in technology enabling the detection of small tumours, the Okuda system has become outdated.As an alternative, the Cancer of the Liver Italian Program (CLIP) (A new prognostic system for, 1998) system evaluates liver function based on the Child-Pugh score, and considers tumour morphology (size and nodular involvement), the levels of α-fetoprotein (AFP) and the eventual presence of portal vein thrombosis (PVT).In addition to the Okuda system and CLIP system, there are multiple staging systems for HCC (Chevret et al., 1999;Llovet et al., 1999a;Leung et al., 2002;Kudo et al., 2003;Tateishi, 2005;Yang et al., 2012;Parikh and Singal, 2016;Yau et al., 2014;Burak and Kneteman, 2010) including (GRETCH, BCLC, CUPI, JIS, Tokyo, MESIAH, ITA.LI.CA, HKLC, Alberta algorithm).However, due to the heterogeneity of the disease and differences in methodologies used, there is no globally accepted staging system for HCC.Recent studies (Tellapuri et al., 2018;Marrero et al., 2005) have evaluated the prognostic power of each system, obtaining that the most robust was the Barcelona Clinic Liver Cancer (BCLC) system.This system is characterized for being the only to include a combination of liver disease, tumour extension and physical-related symptoms, providing an indication for a first-line treatment.In this system (Llovet et al., 1999b), the disease is classified in four groups divided from A to D (early HCC, intermediate HCC, advanced HCC and end-stage HCC respectively), each one with its unique characteristics and prognosis.In early HCC (stage A), are included the asymptomatic patients with a maximum of 3 nodules smaller than 3 cm and it can be further stratified into four subgroups: stage A1, single tumours and absence of relevant portal hypertension and normal bilirubin; stage A2, single tumours associated with relevant portal hypertension and normal bilirubin, stage A3, single tumours with both relevant portal hypertension and increased bilirubin, and stage A4, three tumours smaller than 3 cm regardless of their liver function.Intermediate HCC (stage B) includes the asymptomatic patients with multinodular tumours and without vascular invasion or extrahepatic spread.Advanced HCC (stage C) involves the patients with either symptomatic tumours or with tumoral invasiveness reflected by the presence of vascular invasion or extrahepatic spread.Finally, end-stage HCC (stage D) includes the patients with severe cancer-related symptoms reflected by a deteriorated performance status (PS) or very advanced tumours leading to liver failure.

Diagnosis and therapies
Patients with cirrhosis have been described as a well know population to have an increased risk of developing HCC (Balogh et al., 2016b).That is why it is very important to perform screening techniques in order to identify an early stage of disease so we can provide a wider therapeutic approach while having a better cost-effectiveness ratio.In order to do this, the better technique known for surveillance is ultrasound (US) examination with an interval of 6 months.Once it is confirmed a hepatic lesion, techniques such as computed tomography (CT) scan or magnetic resonance imaging (MRI) are both useful to make a more precise diagnosis.In a cirrhotic liver, hepatocytes have an increased proliferation leading to the formation of regenerative nodules (Balogh et al., 2016b).It is then extremely important to distinguish between a regenerative nodule or the possibility of having a tumour.In order to do that, there are some criteria that classify a mass in the liver to be just followed-up or as HCC and then to be treated as soon as possible.It is defined that nodules smaller than 1 cm and detected by US, are just followed by this same technique during 3-4 months.On the other hand, nodules bigger than 1 cm and detected by US should be confirmed by contrast-enhanced radiologic techniques such as CT scan or MRI.It was demonstrated in a recent meta-analysis that MRI is the technique with higher sensitivity, when compared to the CT scan, in terms of pre-patient or pre-lesion and therefore should be the preferred for the diagnosis of HCC (Lee et al., 2015).It is also possible to confirm the diagnosis measuring biomarkers like the serum AFP level, or through biopsies, even though this last option is recommended to be used cautious because there is a small probability (between 2 and 3%) of tumour seeding during the surgery and it usually give false-negative results (Silva et al., 2008).
In both cirrhotic and non-cirrhotic livers diagnosed with HCC the treatment options follow the same principles even though in the second case the results are less predictable.As discussed before, this disease is often detected in a very advanced BCLC stage, and there are some cases where patients do not present a good response in the first line of treatment.One of the most common approaches to treat HCC tumours is throughout surgical resection and liver transplantation.This option is only available in patients with an early stage of disease (stage A) and it pursuit a curative objective (R0 resection with 5-year survival between 40 and 70%).It can be also used as a less invasive alternatives the percutaneous ablation methods, such as percutaneous ethanol injection (PEI) and radiofrequency ablation (RFA), the transarterial chemoembolization (TACE) and transarterial radioembolization (TARE).Stage B HCC can also be treated with TACE as it was proven to increase the survival of the patients.For more advanced stages of disease (stage C), it is used a systemic therapy with the administration of Sorafenib (Llovet et al., 2008) which is an inhibitor of Raf kinase, vascular endothelial growth factor receptor 2 and 3 (VEGFR2 and VEGFR3), platelet-derived growth factor receptor β (PDGFRβ), c-KIT, FLT-3 and RET in order to inhibit angiogenesis and proliferation (Cheng et al., 2009;Plaza-Menacho et al., 2007).Despite all the side effects (EASL-EORTC Clinical Practice Guidelines, 2012; Iavarone et al., 2011), Sorafenib is recommended whenever tumour progression is observed.For end-stage HCC (stage D) there is no specific cancer therapy, and the best option is supportive care.

Molecular alterations-genetics and epigenetics
The recent advances in technology over the past decade, have enabled the identification of numerous genetic alterations and dysregulated signalling pathways that contribute to elucidate the molecular mechanisms underlying the process of hepatocarcinogenesis.It is also well known that the hepatic microenvironment plays a crucial role in the carcinogenic process, exerting a selective pressure that could explain the heterogeneity of HCC (Grivennikov et al., 2010).
Genomic instability is one of the major forces in tumorigenesis, and it can be classified in chromosomal instability (CIN) and microsatellite instability (MSI).As one of the most frequent CIN alterations (30%-80% of HCC cases), the loss of heterozygosity (LOH) in chromosome 8p leads to the inactivation of Deleted in Liver Cancer 1 (DLC-1) onco-suppressor and the development of early-stage HCC (Niu et al., 2016;HERATH et al., 2006;Lu, 2007;Chan et al., 2002;hang and yuan, 2005).It is also described that 8q chromosomic region is the second most frequently amplified in HCC (48-77%) (hang and yuan, 2005;Jia et al., 2014).There are also other types of CINs in HCC such as gain of function in chromosomes 1q, 5p, 6p, 7q, 8q, 17q and 20q and loose of function in chromosomes 1p, 4q, 6q, 9p, 13q, 14q, 16p-q, 17p, 21p-q and 22q (Marchio et al., 1997;Wong et al., 1999;Kusano et al., 1999;Guan et al., 2000;Chang et al., 2002).Many of these regions are associated with tumour-related genes, in the case of gaining they are related with the activation of proto-oncogenes and in the case of losses they are related with the inactivation of tumour suppressor genes (TSG) (Kan et al., 2013a;Wang et al., 2013;Homayounfar et al., 2013;Fernandez-Banet et al., 2014).
On the other hand, MSI is generally observed at low levels in HCC (Low Levels of Microsatellite Instability, 2017;zhong et al., 2006).Some studies have also highlighted the impact of single nucleotide polymorphisms (SNPs) in HCC development, introducing also the influence of immunoregulatory genes (Wang et al., 2016a;Eldafashi et al., 2021).Very little is known about this topic but it seems that SNPs might be used as predictors of risk and prognosis in HBV-related HCC (Liu et al., 2021).In late stages of HCC, somatic mutations in different TSGs (TP53, p16, AXIN1 and RB), oncogenes (c-MYC and CTNNB1-encoding β-catenin), telomerase reverse transcriptase-promoter (TERT) and other cancer-associated genes (e.g.genes encoding E-cadherin and cyclin D1) are crucial for the progression of the disease and have been found with abundant recurrence (Kan et al., 2013b;Guichard et al., 2012a;Schulze et al., 2015a).Recent studies have also reported somatic mutations in genes related to chromatin remodelling (such as ARID1A and ARID2), oxidative stress (NFE2L2 and KEAP1) and different signalling pathways like RAS/MAPK signalling (RPS6KA3) and JAK/STAT pathway (JAK1) (Kan et al., 2013c;Guichard et al., 2012b;Schulze et al., 2015b;Fujimoto et al., 2012).Also, because of the role of β-catenin and AXIN1, these two proteins have been related to the Wnt/β-catenin signalling pathway.Anyway, most of the last referred genes are mutated in less than 10% of the HCC cases.

Metabolic alterations in pNETs and HCC
Proliferating cancer cells often change their energy metabolism to facilitate their rapid growth and division.This is an identified hallmark of cancer known as energy metabolism reprogramming (Hanahan, 2022), which is closely linked to the malignant behaviour.
pNETs and HCC progression is influenced by aerobic glycolysis and abnormal lipid metabolism (Zhang et al., 2019b;Sun et al., 2021a;Sangineto et al., 2020a).Also, irregular AA metabolism contributes to provide energy for the fast growth and proliferation of malignant cells (Hayashi and Seino, 2018;Wei et al., 2021;Xu et al., 2020).
In certain populations, the occurrence of activating KRAS mutations and TP53 mutations in pNETs suggests that KRAS-driven metabolic changes may play a crucial role in pNETs biology.Oncogenic KRAS regulates glycolytic gene expression at both the transcriptional and posttranscriptional levels (Maharjan et al., 2021;Ying et al., 2012).KRAS signalling controls the transcription of both glucose transporters (GLUTs) and essential glycolysis genes (Yun et al., 2009).The requirement for glucose in the pNETs system appears to be related to the development of facilitated GLUTs and sodium-glucose transporters (SGLTs); pNETs displayed enhanced pyruvate carboxylation, and glucose oxidation via pyruvate dehydrogenase (Chu et al., 2017a).The enzyme PEPCK converts oxaloacetate to phosphoenolpyruvate (PEP).The cytosol and mitochondria contain two isozymes encoded by PCK1 and PCK2, respectively PEPCK-C and PEPCK-M (Chang and Lane, 1966;NORDLIE and LARDY, 1963).PEPCK-C has received increasing attention due to its critical involvement in gluconeogenesis (Yang et al., 2009).PEPCK-M has been shown to enhance, but not replace, PEPCK-C-mediated gluconeogenesis (Méndez-Lucas et al., 2013).PEPCK-M has also been found to be important in the survival strategy launched by stress during metabolism in cancer cells (Méndez-Lucas et al., 2014).PCK2 expression is increased in pNETs cells under low glucose circumstances, in which glutamine is utilised to produce PEP via the action of PEPCK-M, being employed as a biosynthetic intermediate for tumour cell growth (Chu et al., 2017a;Qian et al., 2017).Accordingly, in pNETs cells, the PCK2 silencing increased the expression of genes involved in glycolysis, namely monocarboxylate transporter 4 (MCT4), hexokinase 2 (HK2), lactate dehydrogenase A (LDHA) and glucose-6-phosphate dehydrogenase (G6PD) (Chu et al., 2017b).As a result, glucose deprivation may drive TCA cycle rewiring to enhance glucose-independent cell proliferation via PEPCK-M activation.So, PEPCK-M is overexpressed in pNETs cells and activated for cell proliferation in glucose-depleted circumstances (Vincent et al., 2015).In HCC, the opposite is observed, since the expression of both PCK1 and PCK2 genes are downregulated in primary tumours and low PCK expression was associated with poor prognosis in HCC patients (Liu et al., 2018a;Cai et al., 2020).Accordingly, overexpression of PCK1 or PCK2 in HCC cell lines induces increased levels of apoptosis upon glucose deprivation and suppresses liver tumorigenesis in mice, indicating loss of gluconeogenesis capacity is a feature of liver malignancy (Liu et al., 2018b).Hence, the switch from glycolysis to gluconeogenesis is a mean of supressing liver tumorigenesis (li et al., 2017) and putatively an indication for new therapies.However, data on the role of PEPCK-C and PEPCK-M in HCC must be clarified, since another study based on proteomics analysis identified PCK2 gene as being overexpressed in HCC specimens (KUHARA et al., 2021).
The activation of mutant KRAS causes pNETs progression, leading to an increase in glucose transporters, such as GLUT-1 (Berger and Zdzieblo, 2020;Heimberg et al., 1995), and the same was observed in HCC favouring tumourigenesis (Amann et al., 2009).SGLTs also have a functional role in facilitating glucose uptake.SGLT2 is overexpressed in HCC (Du et al., 2022), and the administration of SGLT1 inhibitors showed promising results in decreased HCC cell proliferation and tumour growth, concomitant with the improvement of liver function (Hendryx et al., 2022).Curiously, and because SGLT2 inhibitors were developed to treat diabetes (Hsia et al.) individuals who take SGLT2 inhibitors commonly exhibit elevated circulating levels of glucagon (Ferrannini et al., 2014;Merovci et al., 2014), often presented at higher glucose blood concentrations (Pedersen et al., 2016).
The increased import of glucose implies a coordination with glycolysis, as the HK2 that catalyses the conversion of glucose to glucose-6-phosphate (Wilson, 2003) has been described as overexpressed in HCC models (Sun et al., 2021b;DeWaal et al., 2018).The promoter region of HK2 contains functional glucose, insulin and glucagon responsive elements, presenting active binding sites for multiple oncogenic transcription factors such as NFκB, MYC, HIF-1α, and STAT3, in both distal and proximal regions (Yeung et al., 2008).Glucagon-induced upregulation of HK2 appears contradictory since it enables the synthesis of the enzyme in cancer cells regardless of the metabolic state of the individual.Nonetheless, this upregulation ultimately benefits cancer cells (Goel et al., 2003) reinforcing the action of glucagon directly on cancer cells metabolic remodelling, even in a subversive way.
Regarding the glycolytic flux of anabolic pathways in pNETs, the PPP serves as a branch of glycolysis that redirects glucose flux towards oxidation (Santana-Codina et al., 2018).The PPP also regulates NADP and nucleic acid synthesis, which facilitates FAs production and ensures cell survival under stressful conditions (Patra and Hay, 2014).When the body adapts to an acidic state, glucose levels increase, and glycolysis decreases.This shift towards PPP enables the redirection of glucose flux towards oxidation and ensures cell survival under such conditions (Patra and Hay, 2014;Bechard et al., 2018).In addition, oncogenic KRAS triggers the selective activation of the non-oxidative PPP, potentially by promoting the expression of genes that are involved in the non-oxidative arm, such as ribulose-5-phosphate isomerase (RPIA) (Santana-Codina et al., 2018).Accordingly, G6PD, the limiting enzyme in PPP, which deviates glucose-6-phospate from the glycolytic flow, is overexpressed in pNETs (Chu et al., 2017c) and HCC (Cao et al., 2021) being associated to more aggressive cell phenotypes (Lu et al., 2018) and to poor prognosis in HCC patients (Cao et al., 2021).
AAs metabolism is a bulk of pathways presenting adaptive alterations favouring cancer.pNETs may benefit by the rewiring of AAs, which contributes to the metabolic profile of pNETs by regulating cellular proliferation, invasion, and redox homeostasis (Smith et al., 2020b).In cellular homeostasis, glutamine plays a crucial role as a multifunctional AA, serving as a key energy source (Serpa, 2020;Cruzat et al., 2018).Glutamine has various biological functions, ranging from providing energy to stabilizing reducing agents, contributing to the biosynthesis of purines and pyrimidines, and its involvement in pNETs has been established (Smith et al., 2020b;Dean et al., 2017).To meet the increased metabolic demand, pNETs cells compensate either by increasing glutamine production or by enhancing glutamine uptake from the environment, resulting in decreased glutamine levels in blood serum (Dean et al., 2017).Just like glutamine, alanine also plays an important role in maintaining metabolic balance in pNETs (Müller et al., 1971) The uptake of alanine is facilitated by the activity of SNAT2 (encoded by SLC38A2 gene), although other transporters such as ASCT1 (encoded by SLC1A4 gene) have also been identified in α-cells (Dean et al., 2017).Alanine is known to be involved in regulating glucagon secretion (Galsgaard et al., 2020;PIPELEERS et al., 1985).The glutamine metabolic reliance, in HCC, is demonstrated in assays showing that glutamic-oxaloacetic transaminase 2 (GOT2) gene silencing improves proliferation, migration and invasion capacities and concomitantly increases the reliance on glutamine degradation by glutaminase (GLS1) (Li et al., 2022b).This is in agreement with evidence showing, in rat hepatocytes, that GOT activity enhances the TCA cycle anaplerotic role, taking advantage of oxaloacetate produced from glutamine-derived glutamate and ensuring the metabolism of lipids.In normal hepatocytes, toxicity due to lipids accumulation can occur through the generation of oxidative stress (Egnatchik et al., 2019), but it may be overcome in HCC by the improved capacity of cancer cells to take advantage of and cope with redox imbalance (Wang et al., 2016b).The metabolic adaptation to use cysteine is another aspect in HCC contributing to oxidative stress control, by using glutathione-dependent reactive oxygen species (ROS) scavenging as a protection (Bonifácio et al., 2021).In fact, HCC tumours present higher levels of glutathione comparing to normal adjacent liver tissue, and an increased reliance on glutathione antioxidant machinery is associated to increased risk of disease relapse (Hsiao et al., 2021).The dependence of HCC on cysteine is maintained by the increased expression of xCT (SLC11A1) transporter, which associates with patients' poor survival (Zhang et al., 2020).
Lipids metabolism is relevant to sustain pNETs and HCC.Little is known on lipids metabolism in pNETs cancer models, but in pNETs patients, a poor prognosis association was observed with increased levels of expression of acetyl-Co carboxylase 1 (ACC1) and FA synthase (FASN), moreover this study supported that systemic and tumour increased lipid bioavailability are associated with decreased progression-free survival (PFS) of advanced pNETs patients treated with everolimus, a mTOR inhibitor (Vernieri et al., 2019a).The inhibition of mTOR is an important therapeutical approach since abnormal PI3K-Akt/PKB-mTOR pathway signalling has been implicated in the pathogenesis of pNETs (Goldstein and Meyer, 2011).Interestingly, everomilus present as side effects hyperglycaemia, hyperglyceridaemia and hypercholesterolaemia (Vernieri et al., 2019b), indicating that PI3K pathway interferes with metabolism and is a core controller of pNETs' metabolic remodelling, in which glucagon may be a key activator.In HCC, the upregulation of FAs anabolism and catabolism, as well as the increased uptake is described as crucial in tumorigenesis and disease progression, as reviewed (Sangineto et al., 2020b).The FA metabolic usage is dependent on the TOME and the genetic profile of tumours, varying according to the clinical context of HCC.Upon oxygen and nutrients scarcity, β-catenin-activated HCC cells increase the uptake of FA from the TOME together with the β-oxidation rate (Iwamoto et al., 2018).Although β-oxidation suppression favoured HCC development in obesity and non-alcoholic steatohepatitis (NASH) contexts (Sangineto et al., 2020c).FA de novo synthesis also has a role in tumoral progression, since ACC1 enzyme has been associated with HCC cell survival under glucose scarcity (Wang et al., 2016c).Furthermore, the fundamental role of FASN in HCC has been demonstrated by genetic silencing and enzyme inhibition, in studies dedicated to HCC in vitro and in vivo models (Calvisi et al., 2011;HAO et al., 2014;Li et al., 2016).
Several evidences point out the relevance of glucagon-dependent signalling and metabolic pathways accounting for cancer metabolic remodelling (Fig. 2 B and C).Nevertheless, the mechanistic involvement of glucagon in pNETs and HCC biology and metabolic adaptation is ought to be determined.

Conclusion
The modulation of glucagon, a regulatory hormone with a direct impact on cell metabolism, holds promise as a new therapeutic approach for malignant neoplasms originating from organs involved in the control circuit of glucagon bioavailability, such as the pancreas, which is responsible for the production of glucagon, and the liver, the main responsible for the clearance of glucagon in peripheral blood.Obviously, it is urgent to confirm that glucagon exerts a role in controlling tumour growth, particularly in the context of pNETs and HCCs.There is already sufficient evidence to reinforce the importance of developing comprehensive studies that combine model assays, analysis of human tumours, and patients' biofluids metabolic monitoring to determine the mechanisms underlying the putative impact of glucagon in the progression of these diseases.
Once the real significance of glucagon in cancer is revealed, several therapeutic strategies approved for use in humans that can be readapted to the treatment of cancer.These therapeutic alternatives were developed mainly for the treatment of diabetes (Okamoto et al., 2015;Damond et al., 2016;Kaur et al., 2020;Lang et al., 2020) .Nevertheless, cancer is a systemic disease, and the main regulators of systemic metabolism can act on malignant cells that express specific receptors allowing them to benefit from the conditions of the tumour microenvironment.In this context, and as explained above, pNETs and HCCs are in a privileged location concerning the bioavailability of glucagon.On the other hand, the regulatory action of glucagon at the cellular level activates core signal transduction pathways that will control metabolic pathways crucial to sustain bioenergetics and biosynthesis essentially supporting cell survival in cancer.
The relevance of this integrative view is to bring to the stage the putative role of glucagon in cancer progression.Currently, the focus on cancer management evolution aims to be personalized and it gathers heterogeneity, specificity and rarity into the same equation.Therefore, the study of less prevalent/rare diseases, as pNETs and HCCs, can point out relevant features and mechanisms to be applied to some subsets of patients.Furthermore, we are facing a huge development on cancer metabolism scientific research, but integration of cancer cell and systemic metabolism components is still missing, in order to understand and manage cancer as a systemic disease.Moreover, the role of TOME is inseparable from cancer biology and its specificities and enriched players must be addressed.
More and more multidisciplinary approaches are needed to understand cancer biology as a whole network, from the molecule to the individual, paving a way of finding specificities amongst disease heterogeneity and defining clinical management towards personalized medicine.