Currently, the only cellular process confirmed to be regulated by PRMT3 in animals and plants is ribosome-mediated protein biogenesis, because the ribosomal protein rpS2 has been shown to be a methylation substrate of PRMT3 [
11,
28]. Although PRMT3 has been shown to enhance hepatic lipogenesis, this effect is methylation-independent and is mediated by direct interaction between PRMT3 and liver X receptor-α, a nuclear receptor that controls the transcription of lipogenic enzymes like fatty acid synthase and acetyl-coenzyme A carboxylase [
29]. In this study, we provide the first evidence that PRMT3 directly methylates GAPDH to promote glycolysis and mitochondrial respiration. The intermediates in the glycolytic pathway and tricarboxylic acid cycle are all increased in PRMT3-overexpressing cells. In addition, these cells exhibit increased ECAR and OCR, which can be reversed by ectopic expression of methylation-deficient R248K mutant GAPDH, confirming the importance of GAPDH in the regulation of cellular metabolism by PRMT3.
Posttranslational modifications (PTM) such as
S-nitrosylation, acetylation, phosphorylation, and
O-linked
N-acetyl glucosamine modification of GAPDH have been demonstrated previously [
30,
31]. However, little is known about arginine methylation of this glycolytic enzyme. When our study was undergoing, two studies reported that PRMT1 and PRMT4 could methylate GAPDH in cells [
32,
33]. Cho et al. demonstrated that PRMT1 induces arginine methylation of GAPDH, resulting in the inhibition of GAPDH
S-nitrosylation and nuclear localization [
32]. However, no methylation site was identified in the study. Zhong et al. showed that PRMT4 methylates GAPDH at R234 and suppresses its catalytic activity to suppress glycolysis and proliferation of liver cancer cells [
33]. Our results indicate that R248 is the major residue methylated by PRMT3 in vivo, and R248 methylation enhances metabolic reprogramming and cellular proliferation of pancreatic cancer cells. R248 is located at the dimer interface, which plays a critical role in the formation of active tetramer [
34]. It is possible that methylation at this residue may promote tetramer assembly or stabilize active tetramer. This hypothesis is supported by our finding that mutation of R248 significantly decreases tetramer formation (Fig.
3e) and dramatically reduces GAPDH activity (Fig.
3b, c). Another important issue to be considered is the synergy or antagonism between different PTMs adjacent to R248. The Cys247 (C247) residue of GAPDH has been shown to be modified by
S-nitrosylation, and this PTM is stimulated by oxidized low-density lipoprotein and interferon-γ [
35]. Phosphorylation of Thr246 (T246) induced by protein kinase C δ under the stress of cardiac ischemia and reperfusion increases the association of GAPDH with mitochondria and inhibits GAPDH-triggered mitophagy [
36]. Functional interplay between phosphorylation and arginine methylation was firstly demonstrated in the transcription factor C/EBPβ [
37]. Methylation of R3 in the N-terminal transactivation domain of C/EBPβ by PRMT4 regulates the interaction of C/EBPβ with the SWI/SNF chromatin remodeling complex and alters the transcription of target genes. Interestingly, phosphorylation of T220 of C/EBPβ by mitogen-activated kinase attenuates PRMT4-mediated R3 methylation. These data suggest that phosphorylation may antagonize the effect of arginine methylation in the regulation of transcription factor activity. Whether the
S-nitrosylation, phosphorylation, and arginine methylation at the 246–248 residues of GAPDH may occur independently, or simultaneously or consequently under various physiological or pathological circumstances, and whether the crosstalk between these PTMs may fine-tune GAPDH function to adapt extracellular alterations are important issues for further characterization.
Metabolic reprogramming is an important process for cancer cells to fit the high demand of energy requirement and supplementation of biosynthetic building blocks. Glycolysis is the metabolic pathway that converts one molecule of glucose to two molecules of pyruvate and generates two molecules of ATP and NADH per reaction. Although the efficiency of ATP production is low, the intermediates generated during the reactions could be used for synthesis of amino acids, lipids, and nucleotides to support rapid tumor growth. Therefore, many cancers switch their cellular metabolism to glycolysis under oxygen-rich conditions and the inhibition of the glycolytic pathway is considered to be a novel strategy for cancer therapy [
38,
39]. However, recent studies point out that mitochondrial respiration also plays a critical role in the survival and metastasis of cancer cells [
40]. In pancreatic cancer, inhibition of KRAS signaling induces extensive cancer cell death. However, a minor population of cancer cells with stemness properties may survive after oncogene ablation and those cells are highly dependent on mitochondrial respiration for survival and regrowth [
41]. Similarly, chronic myeloid leukemia stem cells left after target therapy rely on mitochondrial metabolism for survival [
42]. In addition, breast cancer cells may increase their invasive ability by upregulating peroxisome proliferator-activated receptor γ coactivator 1α-mediated mitochondrial biogenesis and oxidative phosphorylation [
43]. An important finding of this study is the simultaneous increase of glycolysis and mitochondrial respiration in PRMT3-reprogrammed cells. This unique feature provides a molecular basis for the double blockade of these two metabolic pathways in attempts to kill PRMT3-overexpressing cancer cells. Indeed, the combination of oligomycin with heptelidic acid induces a synergistic antitumor effect in vitro and in vivo.