Nuclear functional role of metabolic enzymes and related metabolites: Focus on gene expression regulation

Abstract

Background

Many biological processes from physiological development to different pathological conditions are closely linked to dynamic energetic metabolism and its dysregulations. Mounting evidence shows that metabolic rewiring allows cells to adapt to stress conditions, changes in extracellular cues, and nutritional fluctuations in a timely and precise manner by modulating gene expression. Recent studies reveal non-strictly metabolic functions of metabolic enzymes and related metabolites often confined to the nucleus. Indeed, beyond the diffusion of metabolites through nuclear pores, several metabolic enzymes translocate to the nucleus during cellular differentiation, macrophage activation, tumorigenesis, and so on.

Scope of review

This review aims to outline recent advances in the nuclear functions of metabolic enzymes, focusing on gene expression regulation through transcription factors and epigenetic mechanisms.

Major conclusions

The nuclear localization of metabolic enzymes and metabolites underlines the dual role of metabolism as both a driver and a controller of cellular processes by linking energy metabolism directly to gene expression and cellular reprogramming. The main involvement of respiratory enzymes in nuclear functions suggests a ready interplay between energy status and transcriptional regulation. We trust that these insights will contribute to a more extensive knowledge of the cellular and nuclear landscape and could inspire future investigations on metabolic-mediated gene regulation mechanisms with the aim of developing more effective therapies against diseases.

Highlights

• •Metabolic enzymes display non-canonical nuclear roles.
• •Dynamic nuclear localization occurs under various cellular conditions.
• •Metabolic enzymes control gene expression via transcription factors and epigenetics.
• •Metabolism is both a driver and a controller of gene expression reprogramming.
• •Cell energy is directly linked to transcription through nuclear metabolic enzymes.

1. Introduction

Cellular metabolism comprises a series of enzymatic reactions that convert nutrients into energy, biosynthetic precursors, and reducing equivalents. In normal and pathological conditions, cells rewire their metabolism by modulating gene expression through chromatin modifications and interaction with different transcriptional factors. In the last decade, an increasing body of evidence has demonstrated that many metabolic enzymes are specifically targeted to the nucleus, where they regulate gene expression in addition to their canonical functions [1,2]. By diffusing through nuclear pores or being produced locally inside the nucleus, metabolites can act as signaling molecules to modulate the expression of different genes [3]. Thus, both metabolic enzymes and metabolites play a crucial role in regulating and determining cell fate in normal and pathological conditions.

Metabolism and epigenome are strictly regulated and connected. Nutrient availability affects the activity of various epigenetic modifying enzymes, thus controlling the activity of the complex transcriptional network [4]. For example, glucose activates the transcription of genes involved in its own metabolism, in part by enhancing glucose-derived histone acetylation. High glucose promotes hyperacetylation of histones and consequently affects the expression of genes involved in glucose metabolism and modulation of cell growth [5,6].

It is known that chromatin can exist in different states and structures depending on post-translational modifications. Different histone/DNA modifications such as acetylation of lysine or N-terminal methylation of lysine or arginine, phosphorylation of serine, tyrosine and threonine, methylation/demethylation of DNA, and poly-ADP-ribosylation control spatiotemporal gene expression. Other chromatin modifications identified more recently as “fatty acylation” include propionylation, butyrylation, lactylation, succinylation, crotonylation, malonylation [7]. Altogether, histone post-translational modifications and post-replication modification of DNA, acting in a coordinated manner, affect gene expression, cell fate, tissue specification, and differentiation at different times relating to cell cycle activities, stimuli, and nutrients [8,9].

Non-canonical functions of metabolic enzymes may be also acquired by gene mutations such as mutations within isocitrate dehydrogenase (IDH1 and IDH2) which lead to the production of D-2-hydroxyglutarate (D-2HG) rather than α-ketoglutarate (αKG). D-2HG inhibits αKG-dependent dioxygenase, which is involved in histone and DNA demethylation and hypoxia-inducible factor (HIF) stability [10]. These acquisitions are particularly evident in cancer cells, which need to adapt their metabolism to the large demand of macromolecule precursors for proliferation and to maintain reactive homeostasis against oxidative stress during metastasis progression [11]. As such, some glycolytic enzymes are deregulated in cancer cells and are involved in tumorigenesis [2]. In cancer cells, glycolytic enzymes can be translocated to the nucleus where they participate in tumor progression independent of their canonical metabolic roles [2].

In this review, we describe the nuclear non-metabolic functions of metabolic enzymes and related metabolites focusing mainly on chromatin dynamics, transcription, cell cycle control, and apoptosis.

2. Glycolysis, neoglucogenesis and pentose phosphate shunt enzymes and related metabolites in the nucleus

Glycolysis is a fundamental and conserved metabolic pathway by which glucose is converted into pyruvate. Under aerobic conditions, pyruvate is transported to mitochondria where it is used to generate ATP via oxidative phosphorylation. In anaerobiosis, pyruvate is converted to lactate. In cancer cells, lactate is produced from pyruvate even in the presence of oxygen (Warburg effect). It is known that elevated glycolytic metabolism is a hallmark of cancer together with a global metabolic rewiring. All glycolytic enzymes are upregulated to support the increase of ATP demands and biomolecule precursors in cancer cells. Cell rewiring also involves the translocation of additional copies of glycolytic enzymes to the nucleus, where they carry out non-canonical functions such as gene expression regulation [12].

In cancer cells, elevated glycolytic activity facilitates an open chromatin state, whereas glycolytic inhibition results in chromatin condensation driven by histone hypoacetylation [13]. Glycolytic metabolism significantly impacts global chromatin structure through control of histone acetylation, one of the main histone modifications. For example, in growing cellular state, the acetyl-CoA level increases and triggers the acetylation of histones. Thus, nutritional state, metabolism, and transcription are strictly linked to the epigenetic control of gene expression [6]. In the next sections, we describe the nuclear function of glycolytic enzymes together with those involved in neoglucogenesis and pentose phosphate shunt.

2.1. Hexokinase (HK)/glucokinase (GK)

Hexokinase (HK) catalyzes the phosphorylation of glucose to glucose-6-phosphate, the precursor of nucleotide biosynthesis and NADPH. Isoforms HK 2–4 have been found in the nucleus [14]. HK2 is strongly overexpressed in many cancer cells. It is part of the AKT signaling pathway, and Akt inhibitor IV increases HK2 nuclear localization in HeLa cells [15]. The HK2 yeast homolog of human HKII was also found in the nucleus, where it can interact with Mig, a zinc finger protein, which subsequently recruits Cyc8 and Tup1 to form a protein complex resulting in the repression of carbohydrate metabolic gene expression [16]. Isoform HKI, purified from the rat brain, possesses protein kinase activity, including autophosphorylation and phosphorylation of histone H2B [17].

Hepatic glucokinase (GK), was also found in the nucleus. This enzyme changes its localization between the cytoplasm and the nucleus in response to metabolic requirements. In HeLa cells, GK moves toward the nucleus after interaction with GK regulatory protein (GKRP) via a piggyback mechanism and resides there until dissociation from GKRP [18,19]. GK-GKRP complex acts by modulating glucose utilization and storage in both liver cells and pancreatic β-cells in response to circulating glucose levels, making it a promising target for diabetes treatment [18].

These findings underscore the emerging role of nuclear localization of hexokinase isoforms, which may be particularly relevant in metabolic adaptation and disease conditions.

2.2. Phosphofructokinases (PFKs)

Phosphofructokinase 1 (PFK1) and phosphofructokinase 2 (PFK2) are the key enzymes of carbohydrate metabolism. PFK1 and PFK2 catalyze the phosphorylation of fructose-6-phosphate to fructose 1,6-bisphosphate, an intermediate of glycolysis, and to fructose 2,6-bisphosphate, the powerful allosteric regulator of glycolysis, respectively. In MCF cell cultures, PFK1 was found in the nucleus, where it interacts with transcriptional factors by modulating their expression or activity. Enzo et al. found that, in breast cancer cells, PFK1 modulates the activity of YAP/TAZ (yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ)), which are transcription coactivators required for cancer cell proliferation and progressiveness [20]. An isoform of PFK2, PFKFB3, has also been found in the nuclei of cells from breast cancer, cancerous cervical tumor, and colon cancer, where it takes part in the regulation of the cell cycle and cell proliferation. This was confirmed by the ectopic expression of PFKFB3, which led to increased expression of certain cell cycle proteins, such as cyclin-dependent kinase (Cdk)-1, Cdc25C, and cyclin D3, along with a reduction in the expression of p27, a universal inhibitor of Cdk-1 [21]. In cancer cells, the increase of fructose 2.6-bisphosphate levels might activate the phosphorylation of p27, likely through direct allosteric activation of Cdk1 and Cdk2.

2.3. Aldolase (ALDO)

ALDO catalyzes the conversion of fructose-1,6-bisphosphate to glyceraldehyde-3-phosphate and dihydroxyacetone phosphate (DHAP). Three human isoforms, all present in the nucleus, are expressed in a tissue-dependent manner: ALDOA in muscle, ALDOB in the liver, and ALDOC in neuronal tissue [22]. Isoform A is the most commonly expressed in the nucleus of tumor tissues [23]. Its nuclear translocation may be regulated by phosphorylation via Akt or Erk2 (extracellular-signal-regulated kinase 2). Within the nucleus, ALDOA is involved in the activation of some genes implicated in the S phase of the cell cycle, in the protection of DNA in response to damage, and in the stabilization of transcripts by binding to AT-rich DNA sequences [24,25]. This interaction may contribute to transcript stabilization and could play a role in sensitizing human cells to irradiation [26]. Interestingly, ALDOA expression was also increased in cells treated with Hoechst 33258 which binds to the AT-rich region of DNA [24]. Aldolase also binds to RNA polymerase III in yeast, muscle, and cancer cells [[27], [28], [29]].

2.4. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH)

GAPDH catalyzes the phosphorylation and oxidation of glyceraldehyde-3-phosphate to generate 1,3-biphosphoglycerate using NAD⁺ as the electron acceptor. Some non-canonical functions have been ascribed to GAPDH. Under a variety of stimuli, it accumulates in the nucleus across diverse cell types - such as in cultured neurons treated with cytosine arabinoside - and it works as an intracellular sensor of oxidative stress that takes part in the signaling cascade of cell death [30]. Following apoptotic stimulation, NO (nitric oxide), generated from inducible nitric oxide synthase (iNOS), nitrosylates GAPDH, abolishing its catalytic activity and allowing its binding to Siah1, an E3-ubiquitin-ligase with a nuclear localization signal (NLS). During apoptosis, the GAPDH–Siah1 protein complex moves into the nucleus and mediates proteasome-dependent degradation of specific nuclear proteins [31].

GAPDH acetylation also facilitates its transport from the cytoplasm to the nucleus. Sen et al. [32] demonstrated that acetyltransferase p300/CREB-binding protein (CBP) acetylates GAPDH through direct protein interaction. This modification enhances the enzymatic and acetyltransferase activity of p300/CBP, leading to the activation of downstream effectors, such as p53, and ultimately resulting in cell death. Inhibition of nuclear translocation by deprenyl, a monoamine oxidase inhibitor, prevents apoptosis in neuronal cells [33]. The binding of deprenyl to GAPDH was also found to exhibit a neuroprotective action and was associated with late-onset Alzheimer’s disease [34]. Interestingly, nuclear GAPDH plays a functional role in the maintenance and/or protection of telomeric DNA in response to ceramide and anticancer drugs [35]. Therefore, GAPDH - traditionally recognized for its metabolic role - also functions as a key nuclear regulator, especially in stress-activated pathways.

2.5. Triosephosphate isomerase (TPI)

Triosephosphate isomerase catalyzes the reversible interconversion of the triose phosphate isomers DHAP and d-glyceraldehyde 3-phosphate and its ectopic overexpression controls the nuclear acetate levels by conversion of acetate in 1-acetyl-DHAP. In the late G1 and S phases, Cdk2 phosphorylates TPI, which translocates to the nucleus, where it promotes histone acetylation by reducing nuclear DHAP and increasing the accumulation of nuclear acetate [36]. TPI has been found overexpressed in many tumors, including lung, breast, and gastric cancer, but its function remains elusive because it is associated with both increased cell proliferation and migration in lung adenocarcinoma, and decreased cell growth in hepatocellular carcinoma (HCC) [37]. Thus, TPI may play a role in controlling nuclear acetate levels during cell cycle progression and tumor development.

2.6. Enolase (ENO)

Enolase converts 2-phosphoglycerate to phosphoenolpyruvate (PEP). It has been identified in the nucleus as a short form of the full-length enolase that binds the c-Myc and inhibits the activating effect of transcription factors N1IC (Notch 1 intracellular active domain) and YY1 (Yin Yang 1) [38]. In humans, ENO was found to be mainly located in the nucleus of the zona fasciculata of the adrenal cortex. In particular, ENO increased the promoter activity of 3β-hydroxysteroid dehydrogenase type 2 (HSD3B2) [39], suggesting a possible functional role in the regulation of HSD3B2 expression.

2.7. Pyruvate kinase M2 (PKM2)

Pyruvate kinase catalyzes the conversion of PEP to pyruvate and the production of ATP. Pyruvate inhibits the activity of histone deacetylases (HDACs), leading to an increase in histone acetylation [40]. Four isoforms, differently expressed, have been identified: PKM1, PKM2, PKL, and PKR. However, the most prominent kinase found in the nucleus, in a dimeric form, is PKM2, the only one containing a NLS in its sequence. It is generated by alternative splicing [41] and is also highly expressed in different types of cancers [42]. Several non-canonical functions have been described for PKM2 [43]. Yang et al. showed that, upon epidermal growth factor receptor (EGFR) activation, PKM2 translocates to the nucleus of glioblastoma cancer cells, where it binds to c-Src-phosphorylated β-catenin Y333 (Figure 1). This interaction is required to address PKM2 to the β-catenin/TCF/LEF downstream genes, where it phosphorylates histone H3 at T11 using PEP as a phosphate donor [44]. This phosphorylation allows the dissociation of HDAC3 from CCND1 and MYC gene promoter, histone H3K9 acetylation, and cyclin D1 expression with subsequent G1/S phase transition. PKM2-mediated transactivation of β-catenin and the resulting upregulation of MYC expression promotes the expression of glycolytic genes, supporting the Warburg effect which promotes tumorigenesis and cell proliferation (Figure 1) [44,45]. H3T11 phosphorylation by PKM2 is also required for EGF-induced PD-L1 (programmed death-ligand-1) transcription in HCC [46]. By inhibiting glycolysis or enolase, the enzyme upstream PEP, the phosphorylation process is decreased [47,48].

Figure 1.

Nuclear gene-expression-related functions of PKM2 Pyruvate kinase M2 (PKM2) translocates to the nucleus in various cellular contexts, where it binds to multiple transcription factors and plays a significant role in chromatin modification and gene expression regulation. Abbreviations: PKM2: Pyruvate kinase M2; PIN1 – Peptidyl-prolyl cis/trans isomerase NIMA-interacting 1; P – Phosphate; STAT3 – Signal transducer and activator of transcription 3; HIF1-α – Hypoxia inducible factor 1 subunit alpha; Lac – Lactate group; TRX1 - Thioredoxin-1; NF-κB – Nuclear factor-κB; HCC – Hepatocellular carcinoma; PEP – Phosphoenolpyruvate; PDC – Pyruvate dehydrogenase complex; Amt – Aminomethyltransferase; AhR – Aryl hydrocarbon receptor; CYP1A1 – cytochrome P450 1A1; Ac – Acetyl group; CCND1 – Cyclin D1.

Matsuda et al. found that PKM2 forms a complex with the E2 subunit of pyruvate dehydrogenase complex (PDC) and histone acetyltransferase p300. This complex binds chromatin with aryl hydrocarbon receptor (AhR), a transcription factor associated with xenobiotic metabolism. Specifically, in HeLa cells, PKM2 contributes to the transcriptional activation of cytochrome P450 1A1 (CYP1A1), by promoting acetylation at lysine 9 of histone H3 at the CYP1A1 enhancer (Figure 1) [49]. By studying the impact of high glucose on HCC metastasis, Quian et al. showed that high glucose promotes PKM2 nuclear translocation by changing its lactylation modification. Nuclear PKM2 interacts with the redox protein TRX1 (Thioredoxin-1) and downregulates chemerin expression, thereby enhancing the immunosuppressive microenvironment to boost HCC metastasis (Figure 1) [50].

Lu and Hunter reported that, under EGFR and platelet-derived growth factor receptor (PDGFR) activation, serine 37 of PKM2 is phosphorylated. This modification allows the recruitment of PIN1 (peptidyl-prolyl cis–trans isomerase NIMA-interacting 1), which catalyzes the cis–trans isomerization of PKM2. This leads to the conversion from a tetramer assembly to a monomer unit, which binds to importin α5 for translocation into the nucleus [51]. The interaction with JMJD5, a Jumonji C domain-containing dioxygenase, or PKM2 k433 acetylation [52] prevents its binding to fructose 1-6-bisphosphate and promotes its nuclear translocation.

PKM2 is highly expressed in LPS-activated macrophages, where it regulates genes involved in glycolysis and inflammatory response.

Uptake of lactate into CD4⁺ T cells induces PKM2 to phosphorylate the signal transducer and activator of transcription 3 (STAT3), increases citrate level, resulting in the production of IL-17, fatty acid synthesis, and CD4⁺ T cells retention at the site of chronic inflammation involved in the onset of arthritis [53,54]. By blocking PKM2 nuclear translocation or deleting it in T cells, STAT3 cannot be phosphorylated resulting in the damage of Th17 and Th1 cells. In particular, PKM2 blockage impairs Th17 cell differentiation and ameliorates the symptoms of encephalomyelitis by decreasing Th17 cell-mediated inflammation and demyelination [55].

Other functions independent of catalytic activity can be ascribed to PKM2. The interaction between PKM2 and the HIF-1α subunit promotes transactivation of HIF-1 target genes by enhancing HIF-1 binding and p300 recruitment to hypoxia response elements (Figure 1) [56]. PKM2 also binds to the transcriptional factor Oct4 enhancing its transcriptional activity [57]. PKM2 is critical for LPS-induced dendritic cell (DC) activation. Upon DC activation, the c-Jun NH 2 -terminal kinase (JNK) signaling stimulates p300-mediated PKM2 k433 acetylation. Then PKM2, together with c-Rel, regulates gene expression in Th1 cell differentiation [58]. This outline highlights the non-metabolic role of PKM2 and sheds light on the complex function of an enzyme classically associated with cellular metabolism.

2.8. Lactate dehydrogenase (LDH)

Lactate is produced, under anaerobic glycolysis, through the reduction of pyruvate by lactate dehydrogenase A (LDHA). In the past, lactate was considered a metabolic waste product, but recent evidence suggests that this metabolite is involved in the regulation of histone modification. A high amount of lactate is produced in cancer cells affecting the metabolism of cancer and immune cells, thereby promoting carcinogenesis, angiogenesis, invasion, metastasis, and immune escape [59]. In human fibroblasts, phosphorylated LDH translocates to the nucleus where it binds to DNA and interacts with DNA polymerases α, δ, and ε, promoting DNA synthesis and repair post-UV irradiation, as demonstrated by in vitro experiments [60].

Zhang et al. reported that lactate acts as an epigenetic modifier resulting in lactylation of histones and non-histone proteins [61]. The authors showed that histone lactylation stimulates gene transcription from chromatin in response to hypoxia and pathogens. Twenty-eight lactylation sites were identified on core histones in M1 macrophages exposed to bacteria. In particular, H3K18 (H3K18la) lactylation is a marker for active CpG island-containing promoters of highly expressed genes and positively correlates with H3K27ac and H3K4me3 [62]. H3K18la marks active enhancers that are close to tissue-specific genes [62]. In addition, lactate itself is an inhibitor of HDCAs, giving rise to an increase of histone acetylation affecting gene expression [63]. Repression of LDHA decreases the lactylation process [64].

Recently Huang et al. reported a correlation between nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1), an enzyme of the NAD⁺-salvage pathway (see NAD⁺ section), and lactate in pancreatic adenocarcinoma cells [65]. Under glucose-deprived conditions, lactate promotes the lactylation at K128 of NMNAT1 inducing its nuclear translocation and activation of nuclear NMNAT1-dependent NAD + salvage pathway. Lactylated NMNAT1 plays a regulatory role in preventing NAD ⁺ depletion under glucose-deprived conditions and promoting cell survival [65]. Activation of NMNAT1 by lactate supplies NAD⁺ for Sirt1, one of the major consumers of NAD⁺ within the nucleus.

2.9. Fructose-1,6-bisphosphatase (FBP)

FBP1 is a key regulatory enzyme of gluconeogenesis. It catalyzes the hydrolysis of fructose-1,6-bisphosphate to generate fructose-6-phosphate and inorganic phosphate. Two isoforms, FBP1 (mainly in the liver and kidney) and FBP2 (ubiquitous) are present in mammals [66].

Nuclear FBP1 acts as a tumor suppressor through interaction with the enhancer of zeste homolog 2 (EZH2) in polycomb repressive complex 2 (PRC2) [67]. FBP1 inhibits clear renal cell carcinoma progression by suppressing glycolysis and blocking nuclear HIF function via direct interaction with the HIF inhibitory domain [68].

Another function carried out by FBP1 was reported by Wang et al. [69]. Upon glucose deprivation in hepatocytes, FBP1 is phosphorylated by protein kinase RNA-like ER kinase (PERK) leading to its conversion from a tetrameric to a monomeric structure promoting translocation to the nucleus, where it dephosphorylates H3T11 and suppresses the transcriptional factor PPARα (Peroxisome proliferator-activated receptor alpha) and related gene expression [69].

Nuclear FBP2 slows down mitochondrial biogenesis and respiration in a catalytic-activity-independent manner by inhibiting the expression of nuclear respiratory factor and mitochondrial transcription factor A (TFAM). Furthermore, FBP2 colocalizes with the c-Myc transcription factor at the TFAM locus and represses c-Myc-dependent TFAM expression [66].

2.10. Transketolase (TKT)

Transketolase belongs to the pentose phosphate pathway (PPP). TKT catalyzes a reversible reaction in which a two-carbon ketol unit is transferred from xylulose 5-phosphate to an aldose acceptor, such as ribose 5-phosphate, producing sedoheptulose 7-phosphate and glyceraldehyde 3-phosphate. Three genes have been identified in the human genome encoding TKT isoforms: TKT, TKTL1, and TKTL2 [70]. TKT makes up the majority of transketolase not only in human normal but also in most tumor cells. Qin et al. reported that TKT was associated with metastatic HCC [71]. A significant amount of TKT, distinct from other PPP cytosolic enzymes, has been found to be located in the nucleus of HCC cells and tissues. Although the role of TKT in HCC development is still unclear, mutant variants, proteomic analysis combined with mass spectrometry revealed that nuclear TKT interacted with kinases and transcriptional coregulators such as EGFR and mitogen-activated protein kinase 3 (MAPK3), which are associated with cell activation or stress response processes [71]. Interestingly, survival analysis using clinical data showed that a high nuclear TKT level was associated with impaired liver function and predicted a poor prognosis of HCC.

3. TCA cycle enzymes and related metabolites in the nucleus

In the metabolic scenario of cells, the mitochondrial tricarboxylic acid (TCA) cycle plays a central role in metabolism and energy production. In recent years this view has been overcome by observations that many metabolites, derived from TCA cycle, play an interesting role in nuclear rearrangement by contributing to the regulation of gene expression.

In addition, an increasing body of literature has demonstrated that TCA cycle enzymes are present within the nucleus opening a new view opposite to the dominant dogma that TCA cycle components are localized and act in the mitochondria for energy production.

These findings highlight the essentialness of nuclear TCA cycle enzymes and related metabolites for crucial nuclear functions, such as epigenetic modifications. In the next section, the involvement of specific TCA cycle components in nuclear activities will be discussed.

3.1. Pyruvate dehydrogenase complex (PDC)

It is known that the main function of the mitochondrial PDC is the production of acetyl-CoA for energetic purposes. PDC is a complex of three catalytic enzymes, pyruvate dehydrogenase (PDC-E1), dihydrolipoyl acetyltransferase (PDC-E2), and dihydrolipoyl dehydrogenase (PDC-E3) working with E3-binding protein (PDC-E3BP) subunit.

Since 2005 an extra-mitochondrial localization (juxta nuclear in spermatidis) for PDC-E2 was found [72]. The presence of PDC-E2 in the nucleus of BaF3, an interleukin-3 (IL-3)-dependent cell line, was also observed [73].

Sutendra et al. reported that all the components of the PDC are present in the nucleus of mammalian cells, potentially via chaperone-mediated transport [74]. Active PDC translocates from mitochondria to the nucleus during the cell cycle to generate acetyl units for histone acetylation.

Nuclear PDC was found activated by growth factors or mitochondrial inhibition during the S phase entry of the cell cycle. Interestingly, the mitochondrial pyruvate dehydrogenase kinase (PDK), which inhibits the activity of PDC by phosphorylation in the mitochondria, was not found in the nucleus, suggesting that nuclear PDC is always active [74].

During the cell cycle, translocation of active PDC complex to the nucleus is a fundamental process required for the S-phase entry (Figure 2). Cell cycle progression stimulated by serum or epidermal growth factor (EGF) increases acetylation of histone 3 (H3K9 and H3K18) promoting upregulation of E2F, cyclin A and Cdk2 expression, and phosphorylation of retinoblastoma protein [74]. Knockdown of one of the nuclear PDC units gives rise to a decrease of PDC activity, acetylation of Ac-H3 and G1-S phase progression. Wei Li et al. showed that the production of acetyl-CoA derived from nuclear PDC is important in regulating the epigenetics and pluripotency of stem cells through histone acetylation. In fact, this process is critical for maintaining an open chromatin structure for pluripotency [75]. The involvement of acetyl-CoA derived from nuclear PDC in the epigenetic control of gene expression was also found in mouse embryos during zygotic genome activation (Figure 2) at the 2-cell stage [76].

Figure 2.

Nuclear metabolic enzymes providing Acetyl-CoA Nuclear Acetyl-CoA - a critical metabolite for acetylation of both histones and transcriptional factors - can be generated by several enzymes localized to the nucleus. These include ATP-citrate lyase (ACLY), which converts citrate into acetyl-CoA and oxaloacetate; acetyl-CoA synthetase 2 (ACSS2), which synthesizes acetyl-CoA from acetate; the nuclear-localized pyruvate dehydrogenase complex (PDC2), which produces acetyl-CoA directly from pyruvate; and most likely also carnitine acetyltransferase (CRAT). The activity and localization of these metabolic enzymes can vary depending on cell type, metabolic state, or external stimuli, thereby linking cellular metabolism to epigenetic and transcriptional regulation.

PDC plays also a role in modulating gene expression by interaction with transcriptional factors. Two subunits (E1β and E2) of PDC interact with phosphorylated STAT5 in the promoter of STAT5 target genes, thus affecting STAT5-dependent gene expression [77]. Nuclear PDC-E1 alfa subunit controls the expression of sterol regulatory element-binding transcription factor (SREBF)-target genes through histone acetylation. This function is coordinated with that of mitochondrial PDC aimed to provide cytosolic citrate for lipid synthesis promoting cell proliferation and tumor growth [78]. E2 subunit of PDC forms a complex with PKM2 and p300 on the enhancer regions of arylhydrocarbon receptor (AhR)-target genes, a transcription factor involved in xenobiotic metabolism [49]. Thus, beyond its classical mitochondrial function, nuclear PDC links cellular metabolism to epigenetic and transcriptional control, with critical implications for different call functions.

3.2. Citrate synthase and citrate

The citrate pathway, consisting of citrate synthase, CIC (mitochondrial citrate carrier), and ACLY (ATP citrate lyase) is the major source of acetyl-CoA for histone acetylation in abundant nutrient conditions and for differentiation of 3T3-L1 preadipocytes into mature adipocytes in response to growth factors stimulation [5]. Mitochondria contribute to the acetyl pool through citrate synthesized by citrate synthase, the first step of the TCA cycle. Citrate is exported to the cytosol via CIC and migrates into the nucleus where it reacts with nuclear ACLY by leading to oxaloacetate and acetyl CoA [5].

Oxaloacetate is converted to malate via malate dehydrogenase 1 (MDH1) which moves to the nucleus under glucose deprivation conditions. Here, MDH1 stabilizes and transactivates p53 leading to regulation of p53-dependent cell-cycle arrest and apoptosis [79].

Acetyl CoA acts as a donor for the acetylation of histones by lysine acetyltransferases (KATs) [80]. ACLY was found to translocate to the nucleus in different conditions, for example, in inflammatory macrophages [81], in epithelial cell lines derived from glioblastoma [5], and during cell differentiation [5] (Figure 2). The nuclear translocation of ACLY is driven by the fine-tuned post-translational modification acetylation of ACLY as found in LPS-induced macrophages [81].

High glucose promotes hyperacetylation of histones by ACLY activation that interacts with carbohydrate response element-binding protein (ChREBP), leading to positive regulation of glycolytic and lipogenic gene expression in mice [5,6]. In summary, through the nuclear translocation of ACLY, the citrate pathway enables the gene regulation in response to metabolic and differentiation signals.

3.3. Aconitase (ACO2)

ACO2 catalyzes the isomerization of citrate to isocitrate via cis-aconitate, and it was found in mitochondria, cytosol, and nucleus in mouse embryo cultures [76] and in a yeast strain [82]. However, a specific nuclear function is not yet known. A byproduct of the TCA cycle derived from the decarboxylation of cis-aconitate is itaconate [83]. Upon LPS induction, it accumulates in large amounts in activated macrophages, where it regulates the inflammatory process [84]. Strelko et al. described itaconate as a potent inhibitor of the ten-eleven translocation (TET) DNA cytosine-oxidizing enzymes, α-KG-dependent dioxygenases [84]. It acts as an inhibitor of TET2 activity binding to the same site where αKG binds. Itaconate suppresses LPS-induced genes, including those regulated by NF-κB and STAT signaling pathways; it also decreases 5ʹ hydroxymethyl cytosine (5hmC) level, reduces LPS-induced acute pulmonary edema, lung and liver injury, and protects mice against lethal endotoxemia [84].

3.4. Isocitrate dehydrogenase (IDH)

IDH catalyzes the oxidative decarboxylation of isocitrate to αKG and CO₂. Mutations in isoforms IDH1/2 have been found in many tumors, such as gliomas and cholangiosarcomas [85]. The mutated version of the enzymes leads to the irreversible reduction of αKG to (D-2HG), a metabolite normally not present in physiological conditions, which acts as an inhibitor of αKG -dependent enzymes such as histone demethylases and the TET family of 5-methylcytosine (5 mC) hydroxylases [86].

3.5. α-Ketoglutarate dehydrogenase complex (α-KGDH)

Another post-translational modification of proteins is lysine succinylation, a process in which the metabolically derived succinyl-CoA modifies protein lysine groups [87]. As a consequence of succinylation, proteins are negatively charged affecting their functionality and interaction with other molecules, in both physiological and pathological processes [[88], [89], [90]]. Nuclear histone succinylation is an important step in chromatin-based regulatory epigenetic processes [87,91] and it promotes glycolysis, cell proliferation, migration, and invasion of PDAC cells with epithelial-to-mesenchymal transition [92].

Xie et al. identified histone lysine succinylation sites in several cell types [93]. Smestad et al. demonstrated that, in immortalized mouse embryonic fibroblast cell lines, defective TCA cycle metabolism perturbs the succinyl lysine distribution in chromatin, correlating with transcriptional responses [94] and the knockdown of α-KGDH leads to a decrease in succinylation [95].

The α-KGDH catalyzes the synthesis of succinyl-CoA from αKG. Through different approaches, Wang et al. showed that the nuclear α-KGDH binds to lysine acetyltransferase 2A (KAT2A), a member of the GCN5-related N-acetyltransferase (GNAT) superfamily [96]. Through this nuclear interaction, KATA2 can easily access succinyl-CoA generated in the conversion of αKG to succinyl-CoA by α-KGDH. KGDHC migrates to the nucleus because the KAT2A–α-KGDH nuclear complex increases the local concentration of succinyl-CoA, thereby facilitating histone succinylation. Interestingly, succinylation of lysine 79 on histone 3 was found with a higher frequency around the transcription start site of genes [96]. Inhibition of KAT2A or preventing entry of α-KGDH into the nucleus reduces gene expression, tumor cell proliferation, and overall tumor growth.

Remarkably, during adipogenesis, SLC25A10, the mitochondrial dicarboxylate carrier transporting succinate, was also found localized to the nucleus of adipocytes [97]. This effect could be related to the critical role of succinate levels in the nucleus to address cell fate.

It is known that the conversion of glutamine to glutamate can feed TCA cycle through αKG. Using ¹³C-glutamine as a tracer analysis in isolated HeLa cell nuclei Kafkia et al. observed an increase of TCA cycle metabolites downstream αKG (succinyl-CoA, succinate, fumarate), although enrichment was found at different times [98]. Furthermore, a nuclear location of ACO2 and α-KGDH in mouse embryonic stem cells was also demonstrated [99].

αKG plays a crucial role in epigenetic reprogramming by serving as a cofactor for many enzymes involved in the methylation/demethylation process of DNA and histones, thereby regulating chromatin structure. Removal of methyl groups from DNA is catalyzed by α-KG-dependent TET enzymes, which involve the formation of 5hmC, then oxidation to 5-formylcytosine (5 fC) and 5-carboxycytosine (5caC) and finally conversion to cytosine [100]. Based on this mechanism, histone lysines are demethylated by FAD-dependent amine oxidases and αKG -dependent dioxygenases.

Different histones αKG -dependent demethylases are actively involved in the acquisition and maintenance of embryonic stem (ES) cell pluripotency [101]. Naive ES cells use glucose and glutamine to ensure a high level of αKG, which promotes the demethylation of repressive histone marks and DNA [102]. These repressive marks accumulate under glutamine-deprived conditions [102]. Recently, Dhat et al. reported that αKG might be used as an epigenetic drug. Aberrant 5 mC and 5hmC marks were found in the genome of the diabetic heart [103]. The methylation pattern can be ameliorated by αKG supplementation: a treatment that enhances the demethylation process by increasing the binding activity of TET enzymes.

3.6. Succinate dehydrogenase (SDH) and fumarase (FH)

SDH catalyzes the oxidation of succinate to fumarate with the reduction of ubiquinone to ubiquinol. Mutations in its sequence lead to epigenome alterations. SDH, together with fumarase (FH), is mutated in many cancers affecting genome-wide histone and DNA methylation as a consequence of succinate accumulation [104]. High levels of these dicarboxylic acids can competitively inhibit α-KG-dependent dioxygenases, including histone demethylases as well as the TET family of (5 mC) hydroxylases [104].

FH catalyzes the reversible conversion of fumarate to malate. Beyond the mitochondrial role, cytosolic FH also metabolizes fumarate generated as a by-product of the urea cycle and amino acid metabolism. Upon DNA damage, FH translocates to the nucleus where it promotes the repair of DNA double-strand break, a process, not yet completely known, that requires chromatin and nucleosome remodeling [105]. Jiang et al. demonstrated that FH and related fumarate, are involved in this process [106]. DNA damage by ionization radiation triggers FH phosphorylation by DNA-dependent protein kinase (PK) promoting the binding of FH to H2A.Z, a histone-variant involved in DNA repair [107], at DNA double-strand break regions. This association induces the production of fumarate at the level of double-strand region. Accumulation of fumarate leads to inhibition of KDM2B histone demethylase activity, resulting in enhanced demethylation of K36 in histone H, which promotes accumulation of DNA–PK complex at DNA double-strand break regions for non-homologous end joining DNA repair and cell survival [106].

Another interesting matter regards FH function as a tumor suppressor by fumarate. In fact, a significant reduction of fumarate concentration as a consequence of loss of fumarase leads to inhibition of αKG -dependent prolyl hydroxylase enzymes (PHD 1,2, and 3). This inhibition stabilizes α subunit of HIF, that enhances angiogenesis, glucose metabolism and promotes tumorigenesis [[108], [109], [110], [111]]. Nevertheless, while fumarate can be a positive regulator of genome stability [106,112], its accumulation may lead to angiogenesis and proliferation. Thus, FH on one end promotes genome integrity, while on the other end can contribute to cancer development, a double and contradictory function not yet understood.

4. Fatty acid enzymes and related metabolites in the nucleus

Although glucose is the main source of acetyl-CoA to be used in the histone acetylation process in higher organisms, other evidence shows that acetyl units can be derived from sources other than glucose [113]. Other sources include fatty acids, which are sensed and integrated into the epigenome. Activation of lipid-specific gene expression triggers histone acetylation, suggesting a relationship between fatty acid metabolism and histone acetylation.

4.1. Acyl-CoA synthetase short chain family member 2 (ACSS2)

Another source of acetyl-CoA is the biochemical reaction catalyzed by ACSS2, which generates nuclear acetyl-CoA from acetate. Takahashi et al. reported that nuclear acetyl-CoA synthesis by ACSS2 is rate-limiting for histone acetylation [114]. In ACLY-deleted mouse embryonic fibroblasts, ACSS2 is upregulated and catalyzes the production of acetyl-CoA from exogenous acetate, supporting both lipogenesis and histone acetylation [115]. Under metabolic stress, prostatic adenocarcinoma and breast cancer cells utilize acetate as a nutritional source and for histone acetylation in an ACSS2-dependent manner [116].

Under metabolic stress, the cytosolic form of ACSS2 is phosphorylated and translocates to the nucleus where it binds to the transcriptional factor EB (TFEB), a regulator of lysosomal and autophagy genes. Then, it catalyzes the conversion of acetate derived from the deacetylation of histones to acetyl-CoA, which binds to promoters of TFEB target genes, promoting lysosomal biogenesis, autophagy, cell survival, and brain tumorigenesis (Figure 2) [117]. ACSS2 can enhance HIF-2α acetylation, promoting the formation of HIF-2α-CBP complex. Interaction between ACSS2 and HIF-2α is also required for tumor cell proliferation in hypoxic and low glucose conditions in fibrosarcoma cell lines [118]. However, it is likely that the involvement of ACSS2 as acetyl-CoA donor depends on the type of cell signaling and composition of ACSS2 protein complexes.

4.2. Carnitine acetyltransferase (CRAT)

Acetyl CoA, derived from mitochondrial oxidative metabolism, can be transformed in acetyl-l-carnitine by carnitine acetyltransferase (CRAT) localized in the mitochondrial lumen [119]. Mitochondrial acetyl-l-carnitine is transported into the cytosol by carnitine/acylcarnitine translocase (CAC), then moves to the nucleus, where it is converted to acetyl-CoA to supply acetyl groups for histone acetylation.

Madiraju et al. reported that CRAT is present in liver and kidney nuclear extracts, as well as in the nucleus of cancer cells, such as prostate cancer cells [120], where it may promote histone acetylation in a CRAT-dependent manner [121] (Figure 2). Indeed, CAC deficiency strongly reduced the acetyl-l-carnitine-dependent nuclear histone acetylation [122]. Conversely, Izzo et al. have recently demonstrated that CRAT plays an “acetyl-CoA buffering function” by fostering the transfer of two-carbon units from mitochondria to cytosol in order to prevent PDH inhibition. Cytosolic acetyl-carnitine may provide nuclear acetyl-CoA, likely by carnitine octanoyltransferase (CrOT) [123]. In light of the above-described findings, the mitochondrial-nuclear acetyl-l-carnitine-acetyl-CoA pathway needs further investigation.

4.3. Novel acylations

Thanks to the use of mass spectrometry and other very sensitive tools, more recently, other histone modifications derived from lipid metabolism have been identified. Generically named “acylations”, they are derived from short-chain fatty acids. Similarly, to acetylation, these modifications accumulate on lysine residues and correlate with gene activity. Acylations include: propionylation, butyrylation, crotonylation, malonylation and 2-hydroxyisobutyrylation [124]. By using quantitative subcellular metabolomic measurements, Trefely et al. revealed a nuclear acyl-CoA profile distinct from that of the cytosol with notable nuclear enrichment of propionyl-CoA [125].

4.3.1. Propionyl-CoA/butyryl-CoA

Propionyl-CoA and butyryl-CoA are intermediates in fatty acid catabolism [126]. Propionyl-CoA is derived from the oxidation of odd-chain fatty acids and catabolism of branched-chain amino acids. Using isotope tracing, Trefely et al. identified isoleucine as a major metabolic source of nuclear propionyl-CoA and histone propionylation [125]. By using site-specific antibodies, Kebede et al. found many H3K14 propionylation and butyrylation sites at the promoter of the most highly expressed genes together with acetylation sites [127]. The genome-wide profile of H3 acylation was reshaped in response to alterations in the metabolic state. In fact, the depletion of two enzymes that degrade propionyl-CoA and butyryl-CoA leads to increased levels of propionyl-CoA and butyryl-CoA and protein propionylation and butyrylation [128].

4.3.2. Malonyl-CoA and acetyl-CoA carboxylase

Malonyl-CoA may affect lysine histone and other proteins malonylation within the nucleus and other cellular compartments (cytosol and mitochondria) [129]. Zhang et al. found that deacetylase SIRT5 reduces histone malonylation [130]. The biological function of malonylation is not completely clear. Interestingly, the malonyl-CoA-producing enzyme, acetyl-CoA carboxylase (ACC1), was found in the nucleus and nucleoli. The amount of global lysine malonylation and ACC1 expression was higher in elder mouse brains than in younger ones [130]. Nuclear localization of ACC1 is also reported in human protein atlas (HPA) databases.

4.3.3. Hydroxybutyrate (BHB)

BHB is generated in the ketone body synthesis. Acetyl-CoA and acetoacetyl CoA condense to form 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) through HMG-CoA synthase 2, which releases acetoacetate by HMG-CoA lyase. Acetoacetate is converted to β-hydroxybutyrate by β-hydroxybutyrate dehydrogenase 1 coupled to NAD⁺/NADH. It is known that BHB is used as an energy metabolite but recent evidence shows that it can act as a signal transduction molecule and epigenetic modifier [131]. Indeed, various studies have shown that BHB can influence multiple epigenetic processes—including lysine methylation (Kme), lysine acetylation (Kac), lysine β-hydroxybutyrylation (Kbhb), and microRNA expression—ultimately affecting gene transcription levels [131,132]. For example, it upregulates the expression of JMJD3, a demethylase specific for H3K27me3, and reduces H3K27me3 occupancy at the promoter region of the brain-derived neurotrophic factor (BDNF) gene; it is an endogenous and specific inhibitor of class I HDAC [133].

4.3.4. Crotonyl-CoA

Crotonyl-CoA is generated from butyryl-CoA or glutaryl-CoA. Tan et al. reported the identification of lysine histone crotonylation [134]. The authors observed that histone crotonylation is associated with active promoter regions and putative enhancers in both human somatic cells and murine male germ cells. In male germinal cells immediately following meiosis, lysine crotonylation is enriched on sex chromosomes and specifically marks testis-specific genes, including a significant proportion of X-linked genes that escape sex chromosome inactivation in haploid cells [134].

4.3.5. Glutaryl-CoA

Another PTM involves the histone lysine glutarylation. Tan et al. reported that lysine glutarylation is regulated by deacetylase, sirtuin 5 [135]. Bao et al. found that glutarylation of lysine 91 destabilizes nucleosome in vitro. Mutagenesis on lysine K91 of histone H4, performed on Saccharomyces cerevisiae, affects chromatin structure leading to global upregulation of transcription and defects in cell-cycle progression, DNA damage repair, and telomere silencing [136].

5. Other enzymes and metabolites located in the nucleus

5.1. Methionine and methionine adenosyltransferase (MAT)

DNA and histone methylation are processes catalyzed by methyltransferase (MT) that uses S-adenosyl methionine (SAM) as principal methyl group donor. SAM is synthesized from methionine (MET) and ATP by methionine adenosyltransferase (MAT) which presents two isoforms: the ubiquitous isoform MAT2A and MAT1A found in hepatocytes.

MET is synthesized from homocysteine (HCY) by methionine synthase (MS) that uses 5-methyl-tetrahydrofolate (5-MTHF), from the folate cycle, as a methyl donor (Figure 3).

Figure 3.

The role of the methionine cycle, glutaminase pathway, and NAD⁺ salvage pathway in regulating nuclear metabolic fluxes and epigenomic landscape The nuclear biosynthesis of key metabolites - including S-adenosylmethionine (SAM), the principal methyl group donor for DNA and histone methylation; alpha-ketoglutarate (αKG), a critical cofactor for TET enzymes involved in DNA demethylation; and nicotinamide adenine dinucleotide (NAD⁺), the essential cofactor for sirtuin-mediated histone deacetylation - directly affects epigenetic dynamics. The new insights into the nuclear localization of the enzymes involved in these biochemical reactions highlight a tight interplay between metabolism and gene expression. Abbreviations: HDAC – Histone deacetylases; Ac – Acetyl group; TET3 – Ten Eleven Translocation 3; Me – Methyl group; MT – Methyltransferase; SAH – S-Adenosylhomocysteine; AHCY – Adenosylhomocysteinase; HCY – Homocysteine; SAM- S-Adenosylmethionine; MAT – Methionine adenosyltransferase; MET – Methionine; ATP – Adenosine triphosphate; MS – methionine synthase; THF – Tetrahydrofolate; SER – Serine; GLY – Glycine; 5,10-CH₂-THF – 5,10 methylenetetrahydrofolate; 5-CH₃-THF – 5-methyltetrahydrofolate; MTHFR - Methylenetetrahydrofolate reductase; GLN – Glutamine; GLS – Glutaminase; L-GLU – Glutamate; GDH – Glutamate dehydrogenase; OXA – Oxaloacetate; CIT – Citrate; α-KG – α-ketoglutarate; NAD⁺ – Nicotinamide adenine dinucleotide; NAM – Nicotinamide; NMN – Nicotinamide mononucleotide; NAMPT – Nicotinamide phosphoribosyltransferase; NMNAT – Nicotinamide mononucleotide adenylyltransferase.

SAM is the substrate of MTs for nuclear methylation. Thus, every factor that perturbs methionine metabolism may affect histone, DNA, and RNA methylation levels.

MAT2A has a nuclear as well as cytoplasmatic localization. In the nucleus, it interacts with the MAF BZIP transcription factor (MafK) involved in heme oxygenase-1 gene (HO1) expression [137]. Nuclear production of SAM by MAT2A provides methyl groups for H3K4 and H3K9 methylation resulting in repression of HO1. Furthermore, MAT2A represses the expression of cyclooxygenase 2 (COX-2) by interacting with histone H3K9 methyltransferase SETDB1, thereby promoting the trimethylation of H3K9 at the COX-2 locus [138].

The competitive inhibitor of MT, S-adenosylhomocysteine (SAH)—a byproduct of SAM—is converted into HCY by S-adenosylhomocysteinase (AHCY), which has also been detected in the nucleus (Figure 3) [139].

Nutrient availability controls SAM and SAH levels, thereby affecting H3K4 methylation and gene transcription [140,141]. Dietary methionine depletion reduces the H3K4 methylation catalyzed by SET-domain containing 2 (SET2) leading to apoptosis and cell cycle arrest [142].

These insights suggest a nuclear methionine cycle linking metabolism to epigenetic regulation through SAM- and SAH-dependent methylation.

5.2. Glutamine, glutaminase, and glutamate dehydrogenase

Glutamine is one of the most abundant amino acids in the organism, and participates in many pathways, such as de novo nucleotide biosynthesis, non-essential amino acid synthesis, TCA cycle, urea cycle, and redox balance.

Glutamine metabolism is considered an important adaptative mechanism in the metabolic reprogramming of tumor cells. In this process, glutamine is the major source of α-KG in the TCA cycle through the glutaminolysis pathway, consisting of glutaminase and glutamate dehydrogenase (GDH) [143].

In mammals, four isoenzymes of glutaminase have been characterized, originating from two distinct genes, GLS and GLS2, located on chromosome 2 and chromosome 12, respectively. GLS gene encodes for two isoforms, kidney-type glutaminase (KGA, long transcript isoform) and glutaminase C (GAC, short transcript isoform), while GLS2 gene encodes for two liver-type isoforms, LGA, also named GLS2, and GAB isoforms [144].

The expression level of glutaminase is increased in many tumors and is related to malignancy [145]. GAC is the mitochondrial GLS isoform overexpressed in tumors such as breast cancer, HCC, and colon adenocarcinoma [146,147] playing a role in cellular proliferation, invasion and migration, on energetic metabolism, redox homeostasis and in many other processes involved in tumor progression.

GLS2 was found overexpressed in many kinds of cancer, such as colon, lung and breast cancers and in neuroblastoma [148] where it shows a pro-oncogenic function, by sustaining the viability and replenishing the TCA cycle.

Interestingly, GLS2 was found located in both mitochondria and nuclei in HCC cells in which it shows a tumor suppression activity by inducing an antiproliferative response and underlying non-glutaminolysis functions of GLS2 in this kind of cancer. The precise mechanism is yet unknown, but a correlation between the nuclear targeting of GLS2 and its tumor suppression function was suggested [149]. This function was demonstrated by GLS2 overexpression in non-small-cell lung carcinoma cells and in glioblastoma cells where a phenotype reversion of cancer cells was observed [150,151].

GLS2 does not contain a NLS, opening the hypothesis of either a post-translational modification of GLS2 or an unidentified mechanism of import, as an interaction with other proteins [152].

The next step of glutaminolysis is catalyzed by GDH, which produces α-KG. In humans, there are two isoforms, GDH1 and GDH2, both localized in the mitochondrial matrix, but it was proposed a nuclear localization of GDH with different kinetic constants for the substrates, and different activation by inorganic phosphate, leaving us to speculate a specific function of this enzyme in the nucleus [153,154]. A possible function of neuronal GDH in the nucleus is to regulate TET3 activity by replenishing nuclear αKG to avoid changes in its cellular distribution [155].

5.3. NAD ⁺ salvage pathway

Acetyl units are removed from histone lysine residues by HDAC. Among the four classes of HDAC, only class III, also known as sirtuins, catalyzes the NAD⁺-dependent deacetylations by the ADP-ribosylation of acetyl-lysine of histones, producing nicotinamide (NAM) and O-acetyl ADP ribose (O-acetyl ADPR). Thus, histone deacetylation may depend on the availability of NAD ⁺ derived from de novo biosynthesis, NAD⁺ salvage pathway, but also from glycolysis, TCA cycle, and fatty acids β-oxidation through regulation of NAD⁺/NADH ratio [156]. In cancer cells, aerobic glycolysis lowers NAD⁺/NADH ratio, inhibiting sirtuins and fostering histone hyperacetylation, chromatin de-condensation, and gene dysregulation [157,158]. Additionally, the availability of nutrients affects sirtuin activity. Under restricted diet, starvation or intermittent fasting, the NAD⁺/NADH ratio increases, activating sirtuins. It is interesting to note that in this condition, SIRT1, SIRT3, and SIRT6 suppress several age-related diseases and promote healthspan [159]. On the contrary, a high-fat diet reduces both NAD⁺ levels and sirtuin activity.

Once sirtuins have acted, NAD⁺ is recovered by NAD⁺ salvage pathway that consists of two steps: the first step in which NAM is converted into nicotinamide mononucleotide (NMN) by nicotinamide phosphoribosyltransferase (NAMPT), and the second step in which NMN is converted in NAD⁺ by nicotinamide mononucleotide adenylyltransferases (NMNATs) [160].

NAMPT is the rate-limiting enzyme of the NAD⁺ salvage pathway and exists in two isoforms: an extracellular isoform (eNAMPT) and an intracellular isoform (iNAMPT). The latter was found localized in the cytoplasm, mitochondria, and nucleus [161].

NMNAT exists in three isoforms with different subcellular localization: NMNAT1 is localized in the nucleus, NMNAT2 in the Golgi complex, and NMNAT3 in the mitochondria [162].

Targeting NAD⁺ synthesis via NAMPT is essential for pancreatic cancer cell survival and proliferation [163]. Since NAMPT is the rate-limiting enzyme in the NAD⁺ salvage pathway and is upregulated in many cancers such as colorectal cancer [164], it is an attractive target for cancer therapies.

The NAD⁺ salvage pathway is crucial for maintaining NAD⁺ homeostasis, indispensable for regulating the activity of nuclear deacetylases. For this reason, the existence of this pathway, localized in different subcellular compartments, allows NAD⁺ regeneration in response to many metabolic conditions.

5.4. Sensors of nutrient and metabolic signals

Nutrient and metabolic sensors play a central role in maintaining cellular homeostasis by regulating a wide range of biological processes. The best-characterized include AMP-activated protein kinase (AMPK), mechanistic target of rapamycin (mTOR), and liver kinase B1 (LKB1). Interestingly, these three key regulators are closely interconnected, working together to integrate signals related to energy status, nutrient availability, and cellular stress, thereby regulating critical cellular functions. Dysregulation of nutrient and metabolic sensors has been linked to a broad spectrum of diseases and metabolic dysfunctions [165,166].

AMPK is a heterotrimeric complex composed of a catalytic α subunit (α1 and α2), a scaffolding β subunit (β1 and β2), and a regulatory γ subunit (γ1, γ2, and γ3). During cellular energy stress, AMPK functions as a master energy sensor that is activated by an increased AMP/ADP ratio through phosphorylation at Thr172 by upstream kinases like LKB1 [167]. Once activated, AMPK inhibits anabolic pathways while promoting catabolic processes such as fatty acid oxidation and autophagy [168]. Importantly, AMPK not only works in the cytoplasm but also translocates to the nucleus, where it drives transcriptional and epigenetic changes triggered by metabolic stress. The nuclear translocation of AMPK is strongly dependent on the phosphorylation of its α subunits, particularly the α2 subunit, which exhibits nuclear localization [169], as well as on the calcium-dependent nuclear localization of the α1 subunit [170]. Nuclear AMPK phosphorylates key regulators such as HDACs and TET2, thereby affecting chromatin remodeling, gene expression, and oxidative stress responses [171]. Nuclear accumulation of AMPK-α1 promotes neurodegeneration in Huntington’s disease [167]. In addition, AMPK nuclear translocation supports cell migration and invasive behavior through β-catenin signaling in thyroid cancer [172].

Similarly, LKB1 - a serine/threonine kinase and the main upstream kinase of AMPK - although is primarily localized in the cytoplasm, it contains a nuclear localization signal and can undergo nucleocytoplasmic shuttling. Nuclear LKB1 is mostly inactive unless exported, but it may play regulatory roles in chromatin dynamics and transcriptional control [173]. Therefore, the subcellular localization of these nutrient and energy sensors – particularly their nucleocytoplasmic shuttling - adds another layer of regulation, that affects gene expression reprogramming and, consequently, cell phenotype. Thus, understanding the spatial dynamics of these regulators is essential to decipher their roles in health and disease.

6. Concluding remarks

Research investigations in the past decade have significantly expanded our understanding of metabolic enzymes, revealing their roles beyond the traditional view of catalyzing reactions within a specific metabolic pathway.

Intracellular and extracellular signals may spatially and temporally regulate metabolic enzymes, conferring distinct functions, sometimes together with a different subcellular localization (Figure 4). In this context, in addition to their well-known metabolic role, metabolic enzymes exert non-canonical functions within the nucleus. However, while the role of some chromatin-modifying enzymes within the nucleus is quite clear (i.e. ACLY, PKM2, or ACSS2), for others metabolic enzymes, a specific function is not yet known. Thus, some questions arise for further studies and investigations. A first question to address is understanding the non-canonical role of each metabolic enzyme within the nucleus and how this role is affected in normal and pathological conditions such as cancer, inflammation, or metabolic diseases.

Figure 4.

Nuclear localization of metabolic enzymes Schematic representation of the major energy metabolism pathways highlighting key enzymes with nuclear localization. Metabolic enzymes exhibiting nuclear localization are shown in red. The star indicates that fructose 2,6-bisphosphate (F2,6BP) acts as an activator of PFK1. Abbreviation: HK, HK2 and GK: hexokinase, isoform 2 and glucokinase; G6P: glucose-6-phosphate, F6P: fructose-6-phosphate; PFK1: Phosphofructokinase 1; PFKFB3: isoform of phosphofructokinase 2; F2,6BP: fructose 2,6-bisphosphate; F1,6BP: fructose 1,6-bisphosphate; FBP and FBP1: fructose bisphosphatase and isoform 1; ALDO and ALDOA: aldolase and isoform A; G3p: Glyceraldehyde- 3-phosphate; DHAP: dihydroxyacetone phosphate; GAPDH: Glyceraldehyde- 3-phosphate dehydrogenase; TPI: triose phosphate isomerase; 1,3PPG: 1,3-biphosphoglycerate; 3 PG: 3-phosphoglycerate; 2 PG: 2-phosphoglycerate; ENO: enolase; PEP: phosphoenolpyruvate; PK and PKM2: pyruvate kinase and isoform M2; LDH and LDHA: lactate dehydrogenase and isoform A; PDC: pyruvate dehydrogenase complex; CRAT: carnitine acetyltransferase; CrOT: carnitine octanoyltransferase; CS: citrate synthase; ACO2: aconitase 2; IDH2: isocitrate dehydrogenase 2; αKGDHC: αketoglutarate dehydrogenase complex; SDH: succinate dehydrogenase; FH: fumarase; MDH2: malate dehydrogenase 2; CAC: carnitine/acyl carnitine carrier; CIC: citrate/isocitrate carrier; ACLY: ATP citrate lyase; OAA: oxaloacetate; MDH1: malate dehydrogenase 1; ACSS1 and ACSS2: acetyl CoA synthetase and acyl-CoA synthetase short chain family member 1 and member 2. MOM and MIM refer to the outer and inner mitochondrial membranes, respectively.

Transcriptomic and proteomics analysis have improved the understanding of functional remodeling during cellular processes. Recently, it was demonstrated that about 20% of proteins change localization during differentiation from preadipocytes to adipocytes [97].

The timeline of metabolic pathways seems to be finely tuned by continuous metabolic rewiring in which metabolites and enzymes ensure cell directions such as cell growth, repair, starvation, or homeostasis conditions. In this context, citrate and succinate accumulate and can act as signal molecules activating pro-inflammatory gene expression in M1-activated macrophages. At the same time, ACLY translocates to the nucleus where it fosters the acetylation of both histones and transcription factors such as p65 inducing a reprogramming of gene expression and driving the macrophage phenotype [81,[174], [175], [176]]. Kafkia et al. identified the conversion of glutamine to fumarate, citrate to succinate, and glutamine to aspartate in the nucleus of HeLa cells [98].

Understanding the role of metabolic enzymes and related metabolites in the nucleus is crucial to elucidate the relationship between metabolism and epigenetic regulation. This clarification is also important given that metabolic pathways and epigenetic landscape are different between normal and cancer cells. Most metabolic enzymes that regulate epigenetic modifications are up-regulated in cancer cells. Thus, a path to investigate would be the potential future drugs to be used for different pathological conditions. At present, several potential molecular drugs (i.e. inhibitors of ACLY, ACSS2, glutaminase, and glycolytic enzymes) are under development or investigation. Pharmacological inhibition of ACLY through SB-204990 reduced cisplatin resistance and tumor growth in ovarian cancer cell lines, as well as in xenograft models, likely via modulation of the AMPK-ROS and PI3K-AKT pathways [177]. At the same time, bempedoic acid (Nexletol®) - already FDA-approved for hypercholesterolemia - is being repurposed for oncology based on preclinical efficacy in colorectal and hepatocellular carcinoma models [80]. Moreover, MTB-9655, the first oral ACSS2 inhibitor, is currently in a Phase I clinical trial assessing its safety in advanced solid tumors [178].

Additionally, combining chemotherapeutic agents with glutaminase inhibitors has been shown to reduce cancer cell proliferation in ovarian cancer [179]. Recently, IACS-6274 entered Phase I clinical trials for the treatment of solid tumors [180]. Meanwhile, Telaglenastat (CB-839), combined with nivolumab - a fully human antibody targeting programmed death-1 (PD-1) - has progressed to Phase I/II trials for melanoma, advanced/metastatic clear cell renal cell carcinoma, and non-small-cell lung cancer [181].

Furthermore, 2-deoxy-d-glucose (2-DG), a glycolytic inhibitor, has entered Phase I/II trials for several cancers, including breast and hepatocellular cancer, frequently combined with conventional therapeutic protocols [182]. These examples underscore the growing interest in targeting metabolic enzymes with nuclear functions. Ongoing efforts in drug development, structural studies on the nuclear complexes, and patient stratification will be essential for translating these findings into clinically effective therapies.

One key question to address is why many metabolic enzymes found in the nucleus lack a canonical NLS. How do these enzymes enter the nucleus? Some mechanisms to explain this change of localization were hypothesized, with many information concerning mitochondrial enzymes, such as PDC, for which different hypotheses have been proposed. The existence of a translocation process from mitochondria to the nucleus is supported by the evidence that cells treated with cycloheximide, do not induce any change in the (trans)location of the PDC. It has also been found that an intact functional PDC complex translocates to the nucleus by chaperone proteins [74,183].

By bioinformatic analysis, the presence of NLS in the sequences of TCA cycle enzymes for every enzyme except CS was identified, confirming the existence of a “nuclear” TCA cycle [98].

It was hypothesized the existence of communication domains between mitochondria and nucleus, in which there are protein complexes, such as those formed by the translocator protein (TSPO) [184]. It can be assumed the presence of a fusion of mitochondria or cytosol with nuclei driven by signal molecules, as retrograde signals. These retrograde signals are first demonstrated in yeast [[185], [186], [187]].

We can speculate that this retrograde signaling pathway leads to activating the fusion process in a dynamic way in which the metabolic reprogramming is also accompanied by a different subcellular localization of metabolic pathways to ensure the activation of processes for cell response depending to cellular conditions.

Therefore, despite significant advances in recent years, several questions remain unresolved and represent important challenges for future research. For instance, the extent to which these nuclear-localized enzymes influence transcriptional regulation under different physiological and pathological conditions is still poorly understood. Additionally, the mechanisms by which many metabolic enzymes translocate to the nucleus, in the absence of a canonical nuclear localization signal, remain to be fully elucidated. Future investigations could explore the dynamic and context-specific interactions between metabolic enzymes and chromatin-associated factors. It will also be important to define whether nuclear-localized metabolic enzymes operate independently or as part of macromolecular complexes, and how these complexes are regulated. Addressing these questions may lead to the identification of novel regulatory layers in gene expression and create innovative prospects for therapeutic targeting in diseases marked by metabolic and epigenetic dysregulation.

In conclusion, the nuclear localization of many metabolic enzymes (Figure 4) and related metabolites firstly provides a broader perspective on cellular metabolism, strengthening the concept that metabolism is both a driver and a controller of many cellular processes.

Through the nuclear translocation of many metabolic enzymes, energetic metabolism has a direct impact on gene expression and, consequently, on cellular reprogramming. Moreover, the predominant involvement of cellular respiration enzymes in nuclear functions enables a timely link between gene expression and cellular energy levels.

Therefore, we believe it is essential to acknowledge these new findings to gain a broader understanding of metabolism-related cellular functions, which can be applied to future and, hopefully, more effective therapies.

CRediT authorship contribution statement

Simona Todisco: Writing – review & editing, Writing – original draft, Conceptualization. Dominga Iacobazzi: Writing – review & editing, Writing – original draft. Anna Santarsiero: Writing – review & editing, Writing – original draft. Paolo Convertini: Writing – review & editing, Writing – original draft. Vittoria Infantino: Writing – review & editing, Writing – original draft, Supervision, Funding acquisition, Conceptualization.

Funding

This research was supported by FSC European Funds (Grant number C37G22000400001).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Data availability

No data was used for the research described in the article.