Direct stimulation of de novo nucleotide synthesis by O-GlcNAcylation

Chen, Lulu; Zhou, Qi; Zhang, Pingfeng; Tan, Wei; Li, Yingge; Xu, Ziwen; Ma, Junfeng; Kupfer, Gary M.; Pei, Yanxin; Song, Qibin; Pei, Huadong

doi:10.1038/s41589-023-01354-x

Download PDF

Article
Open access
Published: 12 June 2023

Direct stimulation of de novo nucleotide synthesis by O-GlcNAcylation

Lulu Chen^1,2,
Qi Zhou^1,2,
Pingfeng Zhang ORCID: orcid.org/0000-0001-5985-4914¹,
Wei Tan³,
Yingge Li^1,2,
Ziwen Xu²,
Junfeng Ma ORCID: orcid.org/0000-0002-5183-5425²,
Gary M. Kupfer²,
Yanxin Pei⁴,
Qibin Song ORCID: orcid.org/0000-0002-4350-0916¹ &
…
Huadong Pei²

Nature Chemical Biology volume 20, pages 19–29 (2024)Cite this article

7830 Accesses
2 Citations
73 Altmetric
Metrics details

Subjects

Abstract

O-linked β-N-acetyl glucosamine (O-GlcNAc) is at the crossroads of cellular metabolism, including glucose and glutamine; its dysregulation leads to molecular and pathological alterations that cause diseases. Here we report that O-GlcNAc directly regulates de novo nucleotide synthesis and nicotinamide adenine dinucleotide (NAD) production upon abnormal metabolic states. Phosphoribosyl pyrophosphate synthetase 1 (PRPS1), the key enzyme of the de novo nucleotide synthesis pathway, is O-GlcNAcylated by O-GlcNAc transferase (OGT), which triggers PRPS1 hexamer formation and relieves nucleotide product-mediated feedback inhibition, thereby boosting PRPS1 activity. PRPS1 O-GlcNAcylation blocked AMPK binding and inhibited AMPK-mediated PRPS1 phosphorylation. OGT still regulates PRPS1 activity in AMPK-deficient cells. Elevated PRPS1 O-GlcNAcylation promotes tumorigenesis and confers resistance to chemoradiotherapy in lung cancer. Furthermore, Arts-syndrome-associated PRPS1 R196W mutant exhibits decreased PRPS1 O-GlcNAcylation and activity. Together, our findings establish a direct connection among O-GlcNAc signals, de novo nucleotide synthesis and human diseases, including cancer and Arts syndrome.

O-GlcNAcylation promotes pancreatic tumor growth by regulating malate dehydrogenase 1

Article 25 July 2022

Salvage of ribose from uridine or RNA supports glycolysis in nutrient-limited conditions

Article Open access 17 May 2023

O-GlcNAcylation of PGK1 coordinates glycolysis and TCA cycle to promote tumor growth

Article Open access 07 January 2020

Main

As a distinguishing hallmark of cancer, metabolic reprogramming enhances the production of glycolytic intermediates to synthesize cellular building substances, such as nucleotides, lipids and amino acids^1,2. A supply of nucleotides is essential for rapid cancer cell proliferation^3,4. Phosphoribosyl pyrophosphate synthetase 1 (PRPS1) acts as the first rate-limiting enzyme to produce nucleic acid precursors by converting ribose 5-phosphate (R5P) into phosphoribosyl pyrophosphate (PRPP)^5,6 (Fig. 1a). PRPP is a crucial intermediate in multiple cellular metabolic pathways, including the synthesis of purine and pyrimidine nucleotides, histidine, tryptophan, nicotinamide adenine dinucleotide (NAD) and nicotinamide adenine dinucleotide phosphate (NADP)⁷ (Fig. 1a).

**Fig. 1: OGT regulates nucleotide synthesis and NAD production.**

In Homo sapiens, three distinct isoforms of PRPS, which share a high degree of similarity in their sequences, have been identified. PRPS1 and PRPS2 are expressed across a broad array of tissues, whereas PRPS3 expression appears exclusive to testis^8,9. PRPS1 exists in dynamic monomer, dimer and hexamer forms, and the hexamers serve as active units¹⁰. PRPS1 uses ATP as an energy source to facilitate the transformation of R5P into PRPP, simultaneously generating AMP as a byproduct¹⁰. On the other hand, the PRPS reaction is subject to feedback inhibition by products of nucleotide biosynthesis, such as ADP, AMP and GDP^8,9. Nutritional flux affects nucleotide synthesis directly or indirectly. For example, the energy sensor AMPK directly regulates de novo nucleotide synthesis under metabolic stress conditions. Mechanistically, glucose deprivation leads to PRPS1 serine 180 (S180) and PRPS2 serine 183 (S183) phosphorylation mediated by AMPK, inhibiting PRPS1 and PRPS2 activity as well as subsequent nucleic acid synthesis¹¹.

PRPS1 dysregulation is associated with multiple diseases. Aberrant activation of PRPS1 has been observed in colorectal cancer^12,13, hepatocellular carcinoma¹⁴, breast cancer¹⁵ and T cell acute lymphoblastic leukemia (ALL)¹⁶. PRPS1 uninhibitable activation also results in thiopurine resistance in relapsed childhood ALL¹⁷. Decreased PRPS1 activity is also linked to disorders such as Arts syndrome and retinal dystrophy¹⁸, whereas its superactivity is accountable for neurosensory deficits, hyperuricemia and gouty arthritis^19,20.

O-linked β-N-acetylglucosamine (O-GlcNAc) represents a glycosylation event where a sugar molecule is attached to the hydroxyl groups of serine or threonine residues on specific target proteins. In this procedure, referred to as O-GlcNAcylation, UDP-GlcNAc serves as the sugar donor, originating from glucose through the hexosamine biosynthetic pathway (HBP)^21,22,23. Apart from being a sensor of glucose, UDP-GlcNAc is also modulated by metabolisms of other cellular nutrients, including fatty acids (acetyl-CoA) and amino acids (glutamine), thereby signifying the energy state of cells. O-GlcNAcylation is catalyzed by O-GlcNAc transferase (OGT), whereas its reversal, de-O-GlcNAcylation, is facilitated by β-N-acetylglucosaminidase (OGA)^21,24,25,26. O-GlcNAcylation plays important roles in physiologic and pathophysiologic events, such as transcription, proliferation, differentiation, metabolic homeostasis, inflammation, immunity, development and tumorigenesis^{27,28,29,30,31,32,33,34}. However, the details of the functions and underlying mechanisms of O-GlcNAc in cancers remain mostly unclear.

In the current research, we revealed that O-GlcNAc directly regulates de novo nucleotide synthesis. Mechanistically, PRPS1 serine 83 (S83) and threonine 166 (T166) O-GlcNAcylation mediated by OGT leads to PRPS1 hexamer formation and reduction of feedback inhibition of PRPS1 activity, thereby promoting PRPS1 activation, nucleotide synthesis and NAD production. Moreover, PRPS1 O-GlcNAcylation promoted tumorigenesis and affected cancer cell response to chemoradiotherapy.

Results

OGT regulates nucleotide synthesis and NAD production

Cancer cells exhibit abnormal glucose metabolism and aberrant O-GlcNAcylation level, implying that the flux through the HBP may be altered^21,23,26,35. On the other hand, cancer cells have reprogrammed glucose metabolism to accelerate de novo nucleotide biosynthesis^4,36. To explore the direct regulatory role of O-GlcNAcylation in de novo nucleotide synthesis, we used siRNAs to silence OGT in H1299 cells. Subsequently, we monitored de novo nucleotide synthesis in cells incubated with either ¹⁴C-glucose or ¹³C₆-glucose. As illustrated in Fig. 1b,c, OGT knockdown substantially decreased ¹⁴C-labeled RNA and DNA levels derived from glucose. Both ¹³C-labeled purine intermediate (IMP, AMP and GMP) and pyrimidine intermediate (UMP and CMP) levels were reduced in OGT-deficient H1299 cells and A549 cells (Fig. 1d and Extended Data Fig. 1a). However, silencing OGT did not affect PPP-derived R5P production (Fig. 1e). PRPP levels decreased in OGT-deficient cells (Fig. 1f and Extended Data Fig. 1b). These findings illustrate that OGT regulates de novo nucleotide synthesis by controlling the conversion of R5P to PRPP.

In addition to its role in nucleotide synthesis, PRPP also serves as an intermediate in NAD production⁷. OGT knockdown decreased ¹³C-labeled NAD production (Fig. 1g), indicating that OGT also plays a role in regulating NAD biosynthesis.

The conversion of R5P to PRPP, catalyzed by PRPS, is a process that limits the rate of nucleotide synthesis. Two isoforms (PRPS1 and PRPS2) in lung cancer cells have a similar enzymatic function. PRPS1 mRNA level and protein level were higher than those of PRPS2 in H1299 cells and A549 cells (Fig.1h,i). We further knocked down PRPS1, PRPS2 or PRPS1 plus PRPS2 in these two cells, respectively, and checked cellular PRPP production. As shown in Fig. 1j and Extended Data Fig. 1c, both PRPS1 and PRPS2 promoted PRPP production, and PRPS1 contributed more than PRPS2. In line with the findings above, PRPS1 activity decreased or increased in OGT knockdown or OGT overexpressed cells, respectively (Fig. 1k and Extended Data Fig. 1d–f). However, protein levels did not change considerably (Extended Data Fig. 1g). Consistently, glucose starvation or OGT inhibitor treatment also inhibited PRPS1 activity (Extended Data Fig. 1h,i). These findings suggest that OGT promotes PRPS1 activity, leading to increased nucleotide synthesis.

OGT interacts with and O-GlcNAcylates PRPS1

Based on these results, we hypothesized that OGT O-GlcNAcylates PRPS1 and enhances its activity. Consistent with this hypothesis, OGT interacted with PRPS1 in cells and in vitro (Fig. 2a,b). The C-terminal catalytic domain of OGT interacted with PRPS1, whereas the C-terminal of PRPS1 (amino acids (aa) 148–318) bound to OGT (Extended Data Fig. 2a,b). Furthermore, knockdown of OGT, glucose starvation or OGT inhibitor treatment considerably decreased PRPS1 O-GlcNAcylation (Fig. 2c,d and Extended Data Fig. 2c). Overexpressed OGT led to increased O-GlcNAcylation of PRPS1, whereas decreased O-GlcNAcylation of PRPS1 was observed upon overexpression of OGA. (Fig. 2e,f). Similar results were observed in A549 cells (Extended Data Fig. 2d–h).

To verify the O-GlcNAcylation of PRPS1 in cells, we conducted a chemoenzymatic labeling experiment as previously reported³⁷. Immunoblotting analysis demonstrated detectable signals in the experimental group but not in the control groups, providing evidence that PRPS1 was indeed O-GlcNAcylated (Fig. 2g). As anticipated, knockdown or overexpression of OGT decreased or increased immunoblotting signals considerably (Fig. 2h and Extended Data Fig. 2i), indicating that OGT O-GlcNAcylated PRPS1 in cells. Furthermore, OGT directly O-GlcNAcylated PRPS1 but not PRPS2 in vitro (Fig. 2i,j), indicating the specificity of this modification.

To identify O-GlcNAcylation sites on PRPS1, we purified PRPS1 from HEK293T cells and identified the sites of O-GlcNAcylation on the protein through liquid chromatography with tandem mass spectrometry (LC–MS/MS) analysis. As illustrated in Extended Data Fig. 2j, S83 and T166 were potential O-GlcNAcylation sites on PRPS1. These two sites are conserved in mice, rats and other species (Extended Data Fig. 2k). S83 and T166 are located at the inner side and the outer side of PRPS1 hexamer, respectively (Extended Data Fig. 2l). Furthermore, we found that O-GlcNAcylation-deficient mutations of both S83 and T166, but not only one residue, markedly reduced O-GlcNAcylation signals, suggesting that S83 and T166 are the major O-GlcNAcylation sites (Fig. 2k,l).

O-GlcNAcylation boosts PRPS1 activity

In vitro enzymatic experiments demonstrated that both T166 O-GlcNAcylation and S83 O-GlcNAcylation promoted PRPS1 enzyme activity, and T166 played a more important role than S83 (Fig. 3a). PRPS1 S83A/T166A mutant (2A mutant) showed about 60% less enzymatic activity than PRPS1 wild-type (WT) (Fig. 3a). Thus, S83/T166 O-GlcNAcylation activates PRPS1. Further supporting these results, PRPS1 2A mutant cells showed decreased de novo synthesis of purine intermediate (IMP/AMP/GMP) as well as pyrimidine intermediate (UMP/CMP) levels (Fig. 3b).

PRPS1 exists in monomer or hexamer forms in cells, and the latter show much stronger activity. Structural analyses showed that T166 residue is situated in proximity to the interface of PRPS1 hexamer subunits (Fig. 3c). Based on these observations, we hypothesized that OGT-mediated O-GlcNAcylation of PRPS1 may regulate the hexamer formation of the protein. To investigate this hypothesis, we performed size-exclusion chromatography analyses as previously described¹¹ as well as oligomerization assays using the crosslinking reagent. Consistent with the previous reports^8,11, PRPS1 was present in both monomer and hexamer forms in cells (Fig. 3d). Knockdown of OGT promoted PRPS1 conversion from hexamers to monomers (Extended Data Fig. 3a). OGT overexpression promoted PRPS1 conversion from monomers to hexamers (Extended Data Fig. 3b). Glucose starvation or OGT inhibitor treatment showed similar effects (Extended Data Fig. 3c,d). Furthermore, the PRPS1 2A mutant exhibited much less hexamer formation (Fig. 3d,e). PRPS1 monomer showed low O-GlcNAcylation and high S180 phosphorylation. PRPS1 hexamer showed high O-GlcNAcylation and low S180 phosphorylation (Fig. 3d and Extended Data Fig. 3a). These findings reveal that OGT-mediated O-GlcNAcylation of PRPS1 promotes PRPS1 hexamer formation.

ADP and GDP inhibit PRPS1 activity through a feedback mechanism^8,9,17. To assess the impact of byproduct feedback inhibition on PRPS1 activity, we conducted an enzymatic assay with increasing amounts of ADP or GDP. Knockdown or overexpression of OGT promoted or relieved ADP/GDP inhibition on PRPS1 activity, respectively (Fig. 3f,g and Extended Data Fig. 3e,f). The PRPS1 2A mutation also showed a much stronger feedback inhibitory effect (Fig. 3h,i), suggesting that reduced feedback inhibition may be the mechanism behind the activation of PRPS1 via O-GlcNAcylation. Structure analysis showed that T166 is close to ADP/GDP binding sites, and S83 is not proximal to ADP/GDP binding sites (Fig. 3j). T166 is located in the catalytic domain, adjacent to the outer surface of the hexamer, and is not directly exposed in the available structures. Insertion of O-GlcNAcylation at this site induces both local and global conformational changes. Given that it is connected to the helix that links the allosteric domain, which is further connected to the β-strand that binds ADP/GDP, we speculated that the opening of ADP/GDP binding site would relieve ADP/GDP-mediated feedback inhibition (Fig. 3j). Structure analysis also revealed that S83 is located in the allosteric domain and positioned at the inner circle of the hexamer, without any interactions with other protomers or ligands. However, O-GlcNAcylation at this site functions as a molecular adhesive, contributing to the stabilization of the hexamer (Extended Data Fig. 2l). Perhaps OGT-mediated T166 and/or S83 O-GlcNAcylation induced PRPS1 local and/or global conformational change, respectively, weakening or blocking ADP/GDP binding and relieving ADP/GDP-mediated feedback inhibition.

Similarly to OGT knockdown cells (Fig. 1g), PRPS1 2A mutant cells also showed lower production of ¹³C-labeled NAD (Fig. 3k). These findings show that O-GlcNAcylation of PRPS1 mediated by OGT enhances NAD biosynthesis. Both NAD and nucleotides are important for cancer cell proliferation. To distinguish whether the phenotype stemmed from the lack of nucleotides or the lack of NAD, we replenished NAD, nucleotides or NAD plus nucleotides to the medium of PRPS1 O-GlcNAcylation-deficient cells and examined cell proliferation. As shown in Fig. 3l, replenishment of NAD or nucleotides could partially rescue cancer cell proliferation, and replenishment of both NAD and nucleotides almost fully rescued cancer cell proliferation. These results indicate that both NAD and nucleotides are essential for the phenotype of PRPS1.

PRPS1 O-GlcNAcylation inhibits its phosphorylation by AMPK

PRPS1 is directly phosphorylated at S180 by the energy sensor AMPK, resulting in the PRPS1 transformation from hexamers to monomers as well as subsequent PRPS1 activity inhibition¹¹. Furthermore, PRPS1 monomers showed low O-GlcNAcylation and high S180 phosphorylation. PRPS1 hexamers showed high O-GlcNAcylation and low S180 phosphorylation (Fig. 3d and Extended Data Fig. 3a). Given these observations, we sought to explore the possibility of crosstalk between OGT-mediated PRPS1 O-GlcNAcylation and AMPK-dependent phosphorylation. Knockdown of OGT or OGT inhibitor treatment substantially increased PRPS1 S180 phosphorylation but had no effect on AMPK activation (AMPK T172 phosphorylation) (Fig. 4a and Extended Data Fig. 4a), and overexpression of OGT decreased PRPS1 S180 phosphorylation (Fig. 4b) in H1299 cells. These findings demonstrated that OGT-mediated O-GlcNAcylation of PRPS1 blocked AMPK-mediated phosphorylation of PRPS1. Consistently, the PRPS1 O-GlcNAcylation-deficient mutants (S83A, T166A and 2A) boosted AMPK binding and showed higher phosphorylation levels in HEK293T cells (Fig. 4c,d). Moreover, we produced O-GlcNAcylated PRPS1 protein in vitro and confirmed that O-GlcNAcylated PRPS1 bound less AMPK and O-GlcNAcylation blocked AMPK-mediated PRPS1 phosphorylation in vitro (Fig. 4e,f). We also synthesized O-GlcNAcylated or unmodified peptides of PRPS1 from aa 160–185. By employing AMPK in vitro kinase assays and MS, we further confirmed that T166 O-GlcNAcylation blocked AMPK-mediated S180 phosphorylation (Fig. 4g).

On the other hand, AICAR (an AMPK activator) treatment considerably inhibited PRPS1 O-GlcNAcylation in Ampkα1/α2 WT mouse embryonic fibroblasts (MEFs) (Fig. 4h) . This depends (at least partially) on AMPK, because AICAR treatment had no considerable effect on PRPS1 O-GlcNAcylation in Ampkα1/α2-depleted MEFs (Fig. 4h). The PRPS1 S180 phosphorylation-deficient mutant (S180A) showed increased O-GlcNAcylation level (Fig. 4i). We also generated S83A/S180A and T166A/S180A double mutants and found that S180A increased both S83 O-GlcNAcylation (T166A/S180A mutant) and T166 O-GlcNAcylation (S83A/S180A mutant) (Fig. 4i). Moreover, PRPS1 S180 phosphorylation induced by AICAR or S180D mutant markedly decreased PRPS1 binding to OGT (Fig. 4j,k), resulting in decreased PRPS1 O-GlcNAcylation at S83 and T166. PRPS1 binds OGT through its C-terminal region (aa 148–318) (Extended Data Fig. 2b). Perhaps AMPK-mediated phosphorylation of PRPS1 at S180 induced PRPS1 conformational change, which blocked PRPS1–OGT interaction. All these findings show that OGT-mediated O-GlcNAcylation of PRPS1 and AMPK-mediated PRPS1 S180 phosphorylation inhibited each other, perhaps through inducing global conformational change and blocking protein binding. We also found that AMPK is not essential for O-GlcNAcylation on PRPS1, because OGT inhibitor treatment still promoted PRPS1 conversion from hexamers to monomers (Fig. 4l,m) and inhibited PRPS1 activity in Ampkα1/α2-depleted MEFs (Fig. 4n).

PRPS1 O-GlcNAcylation promotes tumor growth

To assess physiological significance about PRPS1 O-GlcNAcylation, we investigated its impacts on cancer cell proliferation and growth. PRPS1 knockout in H1299 cells reduced EdU incorporation rate and cell proliferation, which were restored by the introduction of WT PRPS1 but not by the O-GlcNAcylation-deficient mutants (Fig. 5a,b). Both PRPS1 S83 O-GlcNAcylation and T166 O-GlcNAcylation promoted EdU incorporation and cell proliferation, and T166 played a more important role than S83 (Fig. 5a,b).

To examine the applicability of these findings in vivo, we performed animal studies by implanting H1299 cells into nude mice. As shown in Extended Data Fig. 4b–e, overexpressed OGT substantially promoted tumor growth in vivo, with elevated PRPS1 O-GlcNAcylation and activity. Moreover, PRPS1 knockdown substantially reduced tumor growth (Fig. 5c). The introduction of PRPS1 WT rescued tumor growth, whereas the PRPS1 2A mutant failed to produce the same effect. (Fig. 5c), indicating that PRPS1 O-GlcNAcylation promotes tumor growth. We also analyzed the nucleotide levels in the tumor tissues. As shown in Fig. 5d, PRPS1 O-GlcNAcylation promotes nucleotide synthesis in xenografts.

Elevated nucleotide synthesis confers cancer cell resistance to chemoradiotherapy^38,39. Similarly to PRPS1 knockdown, the PRPS1 2A mutant also sensitized H1299 cells to etoposide and irradiation (Fig. 5e and Extended Data Fig. 4f). The PRPS1 2A mutant affected DNA repair kinetics and genomic stability (Extended Data Fig. 4g,h), and DNA damage induced PRPS1 O-GlcNAcylation (Extended Data Fig. 4i). Consistently, PRPS1 2A mutant tumors were more sensitive to etoposide than PRPS1 WT tumors, just as in the PRPS1-deficient group (Fig. 5c,f), indicating that PRPS1 O-GlcNAcylation confers lung cancer cell resistance to chemoradiotherapy.

Next, we generated etoposide-resistant lung cancer cells using procedures described previously⁴⁰. Cell viability analyses revealed that the half-maximal inhibitory concentration (IC₅₀) of etoposide for etoposide-resistant (Eto-R) H1299 cells increased by 30-fold (16.31 μM to 0.57 μM) (Fig. 5g). Both PRPS1 O-GlcNAcylation and activity were increased in Eto-R cells (Fig. 5h,i). However, OGT, OGA and PRPS1 protein levels did not change considerably (Extended Data Fig. 4j). Interactions between PRPS1 and OGT substantially increased in Eto-R cells, partially explaining the increased PRPS1 O-GlcNAcylation and activity in those cells (Fig. 5j). Furthermore, OGT or PRPS1 knockdown resensitized Eto-R cells to etoposide (Fig. 5g), indicating that elevated PRPS1 activity is at least one cause of etoposide resistance. H1299 Eto-R cells also produced more NAD than H1299 parental cells (Fig. 5k), which would facilitate therapy resistance of cancer cells as well.

To assess the clinical significance of OGT-mediated PRPS1 O-GlcNAcylation, we conducted a data analysis using the ULCAN platform to determine OGT levels in lung tumor tissue⁴¹. As shown in Extended Data Fig. 5a–c, we observed enhanced mRNA level of OGT in cases of lung cancer and its positive correlation with both malignancy and metastasis. We also measured OGT and PRPS1 O-GlcNAcylation levels in fresh frozen samples of lung cancer tissues obtained during surgery. Consistently, OGT mRNA level was higher in lung tumor tissue than that in normal lung tissue (Extended Data Fig. 5d). PRPS1/2 mRNA levels did not change considerably (Extended Data Fig. 5d). Notably, PRPS activity and PRPS1 O-GlcNAcylation were much higher in tumor tissues (Extended Data Fig. 5e,f), implying that PRPS1 O-GlcNAcylation is involved in lung cancer tumorigenesis.

R196W decreases PRPS1 O-GlcNAcylation and activity

PRPS1 gene mutations have been implicated in several diseases, such as Arts syndrome, an X-chromosome-linked genetic disorder distinguished by cognitive impairment, delayed motor development and hearing impairment^{19,20,42,43,44}. The presence of PRPS1 loss-of-function mutations in patients with this disease is thought to contribute to deficient purine biosynthesis proved by the absence of detectable urine hypoxanthine and decreased levels of uric acid in serum^18,20,44. We generated Arts-syndrome-associated PRPS1 mutations. The R196W mutant showed low PRPS1 O-GlcNAcylation modification and less activity (Fig. 6a,b). We further examined R196W–OGT interaction and found that R196W mutant considerably decreased PRPS1–OGT interaction (Fig. 6a). Perhaps the weak binding between R196W and OGT could not result in enough O-GlcNAcylation, and activation of PRPS1, even overexpressed of OGT, had an insignificant impact on O-GlcNAcylation and activation of R196W (Fig. 6c,d). Interestingly, R196W mutant also decreased AMPK binding and S180 phosphorylation (Extended Data Fig. 4k). We speculate that R196 is important for PRPS1 structure. R196W induces PRPS1 conformational change and weakens PRPS1 binding to other proteins, such as OGT and AMPK, resulting in decreased both O-GlcNAcylation and phosphorylation. This result also further supports that O-GlcNAcylation is essential for PRPS1 enzyme activity.

R196W is an evolutionarily conserved germline mutation in Arts syndrome that was found in the PRPS1 pedigree¹⁸ (Fig. 6e). As expected, the R196W mutant converted PRPS1 from a hexamer to a monomer structure (Fig. 6f). Furthermore, structural analyses revealed that the PRPS1 R196 residue is situated in close proximity to the subunit interface of the hexamers (Fig. 3c). The R196W mutant also inhibited cell proliferation (Fig. 6g) and tumor cell growth in vivo (Fig. 6h) and had much lower PRPS1 activity (Fig. 6i). We also generated an S83A/T166A/R196W triple mutant. As shown in Fig. 6g,i,j, the triple mutant showed functions similar to the 2A mutant or the R196W mutant, implying that the R196W mutant acts, at least in part, through PRPS1 O-GlcNAcylation.

Discussion

The HBP serves as a crucial metabolic switch in cells that regulates glucose and lipid metabolism in reaction to changes of intracellular energy levels and nutrient availability^30,45. Cancer cells usually grow quickly and require more glucose than normal cells to reprogram their metabolism and, thus, support rapid cell proliferation. Here we demonstrated that glucose-induced PRPS1 O-GlcNAcylation promotes de novo nucleotide synthesis and NAD production to support tumor growth. PRPS1 O-GlcNAcylation converted PRPS1 monomers to hexamers, relieved nucleotide synthesis products (including ADP and GDP) mediated feedback inhibition, and promoted activity of PRPS1 and synthesis of nucleotides and NAD (Extended Data Fig. 6).

To sustain homeostasis, dividing cells need to synthesize enough nucleotides to catch up with the cell division rate⁴. Thus, de novo nucleotide synthesis is strictly regulated at different levels in cancer. PRPS1 serves as a crucial enzyme in de novo nucleotide synthesis networks. The energy sensor AMPK phosphorylates PRPS1 at S180 to inhibit nucleotide biosynthesis¹¹. Here we show that OGT-mediated S83 and T166 O-GlcNAcylation promoted PRPS1 activation, de novo nucleotide synthesis and NAD production. Both NAD and nucleotides are essential for PRPS1 functions in cancer cell proliferation (Fig. 3n). The two different post-translational modifications antagonized each other. Mechanistically, AMPK-mediated phosphorylation of PRPS1 at S180 induced PRPS1 conformational change, which blocked PRPS1–OGT interaction, resulting in decreased PRPS1 O-GlcNAcylation. Vice versa, OGT-mediated PRPS1 O-GlcNAcylation also induced PRPS1 structural alteration, which resulted in decreased AMPK binding and phosphorylation. The AMPK and O-GlcNAc pathways represent the principal pathways responding to nutrients that are regulated by cellular energy and nutrient availability^26,46. These pathways are highly interconnected, although they integrate nutrient information in distinct ways. AMPK monitors the ratio of AMP/ATP, whereas glucose, acetyl-coenzyme A and glutamine are involved in HBP activation and protein O-GlcNAcylation^47,48. Thus, crosstalk between OGT and AMPK in de novo nucleotide synthesis allows cells to sense a variety of environmental cues to control de novo nucleotide synthesis. However, OGT still regulates PRPS1 activity and de novo nucleotide synthesis in AMPK-deficient MEFs (Fig. 4m–o). That said, inhibition of AMPK-mediated PRPS1 phosphorylation is not essential for PRPS1 activation by OGT, indicating the complexity and importance of OGT-PRPS1 cascade to sense environmental cues and regulate de novo nucleotide synthesis.

PRPS1 mutations have been linked to enhanced DNA damage response and chemoradiotherapy resistance^38,49. Consistent with these findings, we report here that elevated O-GlcNAcylation caused activation of PRPS1, which is one cause of resistance to etoposide and irradiation treatments in lung cancer. Thus, reducing PRPS1 activity by silencing OGT could resensitize lung cancer cells to chemoradiotherapy. Our findings suggest that PRPS1 O-GlcNAcylation could be used as a biomarker to predict etoposide or radiation resistance and as a treatment target in patients with lung cancer. It is expected that the development of T166 and S83 site-specific O-GlcNAcylation antibodies will open up new avenues of PRPS1 O-GlcNAcylation in the prediction of clinical outcome.

PRPS1 mutations have been identified as the causative factor of Arts syndrome, which is characterized by cognitive impairment, hearing impairment, optic atrophy, early-onset hypotonia, ataxia and delayed motor development^{18,19,20,42,43,50}. Despite these correlations, the exact mechanisms by which these mutations lead to the clinical phenotype are not yet fully understood. We found that Arts-syndrome-associated PRPS1 R196W mutation decreased OGT binding, PRPS1 O-GlcNAcylation and activity. R196W also markedly decreased AMPK binding and PRPS1 S180 phosphorylation. Perhaps R196W mutation induced PRPS1 conformational change and weakened PRPS1 binding to other proteins, such as OGT and AMPK, resulting in both decreased O-GlcNAcylation and phosphorylation. The R196W mutant partially functions through PRPS1 O-GlcNAcylation, which results in impaired nucleotide biosynthesis. Previous reports described undetectable levels of urine hypoxanthine and decreased serum uric acid in patients with Arts syndrome^18,20,44. Interestingly, a clinical trial showed that replenishing purines and S-adenosylmethionine administration had curative effects against Arts syndrome⁴⁴. Given these findings, it is possible that OGT-mediated PRPS1 O-GlcNAcylation may be involved in the etiology of PRPS1 R196W mutation and Arts syndrome.

Collectively, our results highlight a central role for glucose sensor O-GlcNAcylation in de novo nucleotide synthesis. Rescuing the homeostasis of PRPS1 activity by controlling OGT-mediated O-GlcNAcylation or AMPK-mediated phosphorylation is a potential clinical prevention strategy for PRPS1 dysregulation-associated diseases, such as Arts syndrome, or for tumorigenesis and resistance to chemoradiotherapy in lung cancer.

Methods

Mice

Experiments involving animals adhered to a protocol with approval from Institutional Animal Care and Use Committee (IACUC) at Georgetown University. The animals were accommodated in well-ventilated cages with regulated temperature between 20 °C and 26 °C, with humidity levels within the range of 30–70%, with a 12-h light/dark cycle and having unrestricted access to food and water. No more than five mice lived in one cage. Subcutaneous injections of cells were performed in 6–8-week-old male nude mice. This investigation adhered to all applicable ethical guidelines concerning the use of animals in research.

Cell lines

H1299 cells (American Type Culture Collection (ATCC), CRL-5803) (isolated from the lung of a 43-year-old male patient with carcinoma); A549 cells (ATCC, CCL-185) (isolated from the lung tissue of a 58-year-old male patient with lung cancer); and HEK293T cells (ATCC, CRL-11268) (isolated from human embryo kidney tissue) were purchased from ATCC. Ampk^+/+ and Ampk^−/− MEF cells were kindly provided by Eduardo Chini (Mayo Clinic). These cells were cultured in DMEM (Cytiva, SH30243.01) supplemented with 10% FBS (HyClone, SH30910.03) and 1% penicillin–streptomycin (Corning, 30-002-CI) unless otherwise indicated. All cells were maintained at 37 °C and 5% CO₂.

Microbe strains

The strain of DH5α Escherichia coli (New England Biolabs, C2527) was cultured at 37 °C in Luria–Bertani (LB) broth (Fisher BioReagents, BP1427) to amplify plasmids. The strain of BL21 E. coli (New England Biolabs, C2987) was cultured at 16 °C in LB broth (Fisher BioReagents, BP1427) with 0.2–0.5 µM isopropyl β-d-1-thiogalactopyranoside (IPTG) (Sigma-Aldrich, I6758) for protein purification.

Human specimens

Lung tumor samples and adjacent tissues from the same individuals were collected from surgically resected lung tissues at Renmin Hospital of Wuhan University (Supplementary Table 1). The specimens were diagnosed by pathologists. The study’s protocol was approved by the Ethics Committee of Renmin Hospital of Wuhan University. All participants provided informed consent for the study.

Plasmids, shRNA, siRNA and transfection or infection

PRPS1 WT, O-GlcNAcylation-deficient mutants (S83/Thr166 → Ala: S83A, T166A and 2A), S180 phosphorylation-deficient mutant (S180 → Ala: S180A), S180 phosphorylation mimic mutant (S180 → Asp: S180D) or disease-associated mutants (Leu152 → Pro: L152P; Arg196 → Trp: R196W; Gln133 → Pro: Q133P) were cloned into pEFF-FLAG or pCDH-FLAG. Cell transfection was performed using polyethyleneimine (PEI) (Polysciences, 23966) or Lipofectamine 3000 (Thermo Fisher Scientific, L3000) in accordance with the manufacturersʼ instructions.

The shRNA oligonucleotides targeting human OGT (Supplementary Table 2) were cloned into a pLKO.1 vector (Addgene, 10878). Lentivirus particles were produced to transduce cells to express shRNA or PRPS1 cDNA (pCDH-FLAG plasmids). The indicated plasmids, PEI (Polysciences, 23966) and packaging vector pMD2.G (Addgene, 12259) and psPAX2 (Addgene, 12260) were used to transfect HEK293T cells (ATCC, CRL11268) to produce lentivirus particles. The medium containing virus was collected and filtered at 48 h and 72 h after transfection. Cells were infected by the virus with 8 µg ml⁻¹ polybrene (Millipore, TR1003G). Medium containing 2 µg ml⁻¹ puromycin (Sigma-Aldrich, 9620) was used for selection.

Lipofectamine RNAiMAX (Thermo Fisher Scientific, 13778) was used to deliver siRNA oligonucleotides into cells based on the manufacturer’s protocol. The siRNA sequences targeting human OGT and PRPS1 are summarized in Supplementary Table 2.

Gene deletion by CRISPR–Cas9

The sgRNA oligonucleotide targeting human PRPS1 (Supplementary Table 2) was cloned into a lentiCRISPRv2 vector according to Feng Zhang’s protocol^51,52. The sgRNA oligonucleotide targeting human PRPS2 (Addgene, 78005)⁵³ was obtained from Addgene. Lentiviruses were produced as above to infect cells. Puromycin selection was conducted, and single clone cells were picked and amplified to test knockout efficiency by immunoblotting. For the PRPS1 expression rescue study, PRPS1 cDNAs carrying knockout-resistant synonymous mutations (CATGGTACTAGTAGGCGACG) against sgPRPS1 were introduced into PRPS1 knockout cells.

Real-time PCR analysis

Total RNA was extracted using a Total RNA Miniprep Super Kit (BIO BASIC, BS784). RNA was used to do reverse transcription with the iScript Reverse Transcription Supermix (Bio-Rad, 1708840) in preparation for real-time PCR analysis. SYBR Green Supermix (Bio-Rad, 1725270) was used to do real-time PCR with the Bio-Rad CFX96 device. Reactions were run in triplicate using primers targeting human genes (Supplementary Table 2).

Protein purification, immunoprecipitation and immunoblotting

Cells were harvested and resuspended in NETN buffer supplemented with protease inhibitor cocktail (Roche, 4693159001) or phosphatase inhibitor cocktail (Roche, 4693159001). The cells were lysed by rotation at 4 °C for 10–20 min and then centrifuged at 13,000g for 15 min at 4 °C after short sonication. The supernatant was collected, and protein concentrations were quantified with a BCA Protein Assay Kit (Tiangen, PA115).

For immunoprecipitation, the lysate was incubated with indicated primary antibodies and protein A/G agarose beads (Santa Cruz Biotechnology, sc-2003) or with anti-FLAG beads (Sigma-Aldrich, A2220) at 4 °C overnight. Normal IgG (Cell Signaling Technology, 2729) was used as control in endogenous immunoprecipitation. The beads were washed five times with NETN buffer, and the precipitated proteins were subjected to immunoblot analyses or other functional assays.

For immunoblotting, equal amounts of protein were mixed with loading buffer (Bio-Rad, 1610747) and boiled for 5–10 min. The protein samples were loaded and separated in an SDS-PAGE gel, followed by transfer onto a PVDF membrane (Thermo Fisher Scientific, 88518). The membranes were incubated in indicated primary antibody solutions at 4 °C for 12–16 h after being blocked in 5% milk TBST solution. The following primary antibodies were used for immunoblotting: PRPS1 (Proteintech, 15549-1-AP); phospho-PRPS1 (S180) (Thermo Fisher Scientific, PA5-106230); PRPS2 (Novus, NBP1-31435); OGT (Cell Signaling Technology, 24083); OGA (Proteintech, 14711-1-AP); AMPKα (Cell Signaling Technology, 2532); phospho-AMPKα (Thr172) (Cell Signaling Technology, 2535); ACC (Cell Signaling Technology, 3662); phospho-ACC (S79) (Cell Signaling Technology, 3661); β-tubulin (Proteintech, 66240-1-Ig); β-actin (Proteintech, 60008-1-Ig); γ-H2AX (S139) (Cell Signaling Technology, 9718); histone H3 (Millipore, 06-755); O-GlcNAc (RL2) (Abcam, ab2739); O-GlcNAc (CTD110.6) (Cell Signaling Technology, 9875); O-GlcNAc (18B10.C7) (Thermo Fisher Scientific, MA1-038); FLAG (Sigma-Aldrich, F3165); HA (BioLegend, MMS-101P, clone 16B12); HA (Cell Signaling Technology, 3724); and Myc (Santa Cruz Biotechnology, sc-40). The membranes were washed with TBST buffer before secondary antibody (Jackson ImmunoResearch, 115-035-003 and 111-035-003) incubation. The secondary antibodies were washed off, and blots were developed with chemiluminescent substrate (Thermo Fisher Scientific, 34580) by an AI600 imager (GE Healthcare).

Click-iT O-GlcNAc enzymatic labeling and stoichiometric analysis of O-GlcNAcylated PRPS1

O-GlcNAcylation of PRPS1 using the enzymatic labeling method was carried out as described previously^37,54,55,56. In brief, 400 μg of total cell or tissue lysate was labeled according to the Invitrogen Click-iT O-GlcNAc enzymatic labeling protocol (Thermo Fisher Scientific, C33368). Enzymatic labeled proteins were conjugated with an alkyne-biotin compound based on Invitrogen Click-iT protein analysis detection kit protocol (Thermo Fisher Scientific, C33372). The biotinylated proteins were precipitated by streptavidin resin (Thermo Fisher Scientific, 20353) and then eluted in a loading buffer containing 20 mM biotin (Sigma-Aldrich, B4501) by boiling. The immunoblot intensities of PRPS1 in elution and in input were measured by ImageJ (National Institutes of Health, https://imagej.nih.gov/ij/) to quantify the PRPS1 O-GlcNAcylation level for each sample.

O-GlcNAcylation in vitro assay

The O-GlcNAcylation in vitro assay was reported previously^30,33. In brief, about 2 μg of GST-PRPS1 or GST-PRPS2 protein was incubated with 1 μg of enzymatic His-OGT domain (aa 323–1,041) in 100 μl of 50 mM Tris-HCl (pH 7.5) containing 12.5 mM MgCl₂, 2 mM UDP-GlcNAc and 1 mM dithiothreitol at 37 °C for 2–5 h. The proteins were subjected to immunoblot analyses or other functional assays.

Cell growth and cell colony formation assays

For cell growth assays, cells were seeded into a 96-well plate. A reagent (Dojindo, CK04-11) was added into each well to measure cell numbers by reading optical density at 450 nm (OD₄₅₀) values using a plate reader based on the protocol from the manufacturer. For cell colony-forming assays, 500–800 cells were seeded in tissue culture plates and grown for 1 week. Cell colonies were fixed in 4% formaldehyde (Sigma-Aldrich, 47606) and then stained with 1% crystal violet (Sigma-Aldrich, C6158) for counting.

For cell proliferation rate analysis after replenishment of IAGUC and NAD, inosine (I4125), adenosine (A4036), guanosine (G6264), uridine (U3003), cytidine (C4654) and β-nicotinamide adenine dinucleotide (N7004) were obtained from Sigma-Aldrich. Cells were seeded in a 96-well plate (3,000 cells per well) and incubated in the medium with or without 0.5 mM inosine, adenosine, guanosine, uridine, cytidine and β-nicotinamide adenine dinucleotide from Sigma-Aldrich for 72 h from day 1 to day 4. The OD₄₅₀ was measured, and the proliferation rate was calculated by dividing the OD₄₅₀ on day 4 by that on day 1 for each group.

GST pulldown assays

PRPS1, PRPS1 truncated mutants and OGT were cloned into a pGEX-4T-2 vector. GST, GST-OGT, GST-PRPS1 and GST-PRPS1 truncated mutant fusion proteins were expressed and purified from E. coli BL21 cells (New England Biolabs, C2527) and captured by glutathione sepharose gels (Cytiva, 17075601) and then rotated with HEK293T cell lysate at 4 °C overnight. The recombinant O-GlcNAcylated and non-O-GlcNAcylated GST-PRPS1 was obtained from the in vitro O-GlcNAcylation assays as previously described, and, after being washed three times, the recombinant GST-PRPS1 was then incubated with the purified AMPK (Upstate Biotechnology, 14-840) at 4 °C overnight. The beads were washed five times and boiled in a loading buffer (Bio-Rad, 1610747). The samples were subjected to immunoblot analyses and Coomassie blue (Sigma-Aldrich, B0149) staining.

Immunostaining experiments

Cells were cultured on glass coverslips in 3.5-cm dishes. After indicated treatments, cells were washed in PBS, fixed in 4% formaldehyde for 15 min, permeabilized in 0.5% Triton X-100 for 5 min and blocked in 5% BSA for 1 h at room temperature. The coverslips were incubated in a diluted γ-H2AX primary antibody (Cell Signaling Technology, 9718) solution for 1 h at room temperature. After being rinsed twice in PBS, the cells were incubated in a diluted secondary antibody (Jackson ImmunoResearch, 111-585-045) solution for 30 min at room temperature. The coverslips (Globe, 1404) were then washed in PBS. A mounting medium with DAPI (Thermo Fisher Scientific, 00-4959-52) was used to mount the coverslips on a glass slide (Globe, 1354). Images were captured by Leica SP8 microscopy, and γ-H2AX focus-positive (≥10 foci per cell) cell ratios were counted and analyzed.

EdU incorporation assays

A Click-iT EdU Cell Proliferation Kit for Imaging (Thermo Fisher Scientific, C10337) was used to perform EdU incorporation assays according to the manufacturer’s instructions. In brief, cells were seeded on coverslips (Globe, 1404) and labeled in 10 µM EdU medium for 30 min. After fixation and permeabilization, cells were treated with the reaction cocktail mixture for about 30 min in the dark at room temperature. The coverslips were washed twice in PBS and mounted with a mounting medium with DAPI (Thermo Fisher Scientific, 00-4959-52) on a glass slide (Globe, 1354). Images were captured by Leica SP8 microscopy, and EdU-positive cells were counted and analyzed.

Size-exclusion chromatography

Size-exclusion chromatography was performed using a Superdex 200 Increase 10/300 GL column (GE Healthcare, 28-9909-44) and a GE ÄKTA pure machine. Then, 500 µl of cell lysate was loaded onto the column and fractionated at a flow rate of 0.5 ml min⁻¹. A protein standard mix of 15–600 kDa (Supelco, 69385) was used as a calibration marker. The sample fractions were subjected to immunoblot analyses.

PRPS1 oligomerization assays

PRPS1 oligomerization assays were conducted as previously reported⁵⁷. Then, 20 mM HEPES (pH 7.5) was used to lyse cells, and cells were then centrifuged at 13,000g for 10 min at 4 °C. The crosslinking reagent 0.025% glutaraldehyde (Sigma-Aldrich, G5882) was added to the supernatant, and the mixture was incubated at 37 °C for 5 min. The reaction was ended by adding Tris-HCl (pH 8.0, 100 mM). The samples were mixed with loading buffer (Bio-Rad, 1610747) and subjected to immunoblot analyses.

PRPS1 structure modeling

The cryo-electron microscopy (cryo-EM) structures of human PRPS1 (Protein Data Bank (PDB) ID: 2H06 (ref. ⁸) and PDB ID: 8DBE (ref. ⁵⁸), which contains ADP in both allosteric and catalytic sites) were used to show PRPS1 hexamer, dimer and the residue locations. Pymol (Schrödinger, https://pymol.org/2/) was applied to perform structural analysis and generate the structural figures.

Enzymatic activity measurement

PRPS1 activity was measured as previously described¹¹. Endogenous or recombinant PRPS1 was purified from cells and incubated in 100 µl of freshly prepared reaction buffer containing 50 mM Tris-HCl (pH 7.5), 4.7 mM ribose 5-phosphate (Sigma-Aldrich, 83875), 0.4 mM NADH (Sigma-Aldrich, N8129), 3.2 mM ATP (Sigma-Aldrich, A9187), 1.8 mM phosphoenolpyruvate (Sigma-Aldrich, 860077), 6 mM MgCl₂ (Thermo Fisher Scientific, 021315), 31 mM NaHCO₃ (Thermo Fisher Scientific, S5761), 7 U pyruvate kinase (Roche, 56105724), 10 U lactic dehydrogenase (Sigma-Aldrich, L1006) and 10 U myokinase (Sigma-Aldrich, M3003) in a 96-well plate (CELLTREAT Scientific Products, 229196) at 25 °C. Absorbance at 340 nm was read for 15 min by a microplate reader (Molecular Devices, SpectraMax iD3 Multi-Mode Microplate Reader). The changes in the OD value and the intensities of PRPS1 immunoblots were used to quantify the PRPS1 activity.

PRPS activity from human tissues was detected using a PRPP-S Assay Kit (NovoCIB, K0709-04-2) following the protocol from the manufacturer.

ADP and GDP feedback inhibition on PRPS1 activity was detected as previously described¹⁷ with some modifications. In brief, recombinant PRPS1 was purified from HEK293T cells and incubated in 10 µl of 50 mM Tris-HCl (pH 7.5) buffer containing 2 mM ATP (Sigma-Aldrich, A9187), 2 mM ribose 5-phosphate (Sigma-Aldrich, 83875), 10 mM MgCl₂ (Thermo Fisher Scientific, 021315) and 1 mM DTT (Roche, 10197777001) and indicated concentrations of ADP (Sigma-Aldrich, 01905) or GDP (Sigma-Aldrich, G7127) for 1 h at 37 °C in a 384-well white board plate (Greiner Bio-One, 784075). ATP was consumed to different degrees during the reaction based on the PRPS1 activity. Reactions were ended with 10 µl of Kinase-Glo reagent (Promega, V6072) being added to each well to terminate the reaction. Luminescence representing the residual ATP amount was detected by a microplate reader (Molecular Devices, SpectraMax iD3 Multi-Mode Microplate Reader). The luminescence signal values were used to calculate the inhibition rate of PRPS1 activity.

Influx of glucose into RNA and DNA

H1299 cells were cultured with ¹⁴C-glucose (PerkinElmer, NEC042V) (1 μCi) for about 30 s. Qiagen RNA and DNA kits (74104 and 69504) were used to purify RNA and DNA. The de novo synthesized ¹⁴C-RNA or ¹⁴C-DNA was measured using a scintillation counter (Beckman Coulter, LS 6500 Multi-Purpose Scintillation Counter).

AMPK in vitro kinase assays

The in vitro kinase assays were performed as previously described^34,59. The PRPS1 O-GlcNAcyated peptide (SEWRNCT(O-GlcNAc) IVSPDAGGAKRVTSIADRL) and the corresponding non-O-GlcNAcylated peptide were ordered from Chinese Peptide Company. The recombinant O-GlcNAcylated and non-O-GlcNAcylated GST-PRPS1 proteins were obtained from the in vitro O-GlcNAcylation assays as previously described. The purified AMPK (Upstate Biotechnology, 14-840) was incubated with the peptides or the GST-PRPS1 proteins in the kinase reaction buffer (15 mM HEPES (pH 7.0), 450 µM DTT (Roche, 10197777001), 18.75 mM MgCl₂ (Thermo Fisher Scientific, 021315), 6.25 mM β-glycerophosphate (Sigma-Aldrich, G9422), 1.25 mM EGTA (Sigma-Aldrich, 324626) and 125 µM ATP (Sigma-Aldrich, A9187) with or without 150 µM AMP (Sigma-Aldrich, 01930)) at 30 °C for 15 min. For PRPS1 peptides, S180 phosphorylation was analyzed by MS. For GST-PRPS1 proteins, S180 phosphorylation was analyzed by immunoblotting.

Metabolite analysis by LC–MS/MS

For cell-based ¹³C₆ metabolite labeling analysis, around 90% confluent H1299 cells were washed by glucose-free culture medium (Gibco, 11966-025) and cultured in medium containing ¹³C₆-glucose (10 mM) (Cambridge Isotope Laboratories, CLM-1396) for 30 min. High-resolution LC–MS/MS was used for analyzing the levels of ¹³C₆-labeled intracellular metabolites¹¹.

For label-free xenograft metabolite analysis, the xenograft tissue samples were homogenized in 150 μl of extraction solution (half methanol and half water) with 250 ng ml⁻¹ 4-nitrobenzoic acid and debrisoquine as internal standard for negative and positive modes, respectively, followed by adding 150 μl of acetonitrile. After incubation for 30 min at −20 °C, the samples were centrifuged at a speed of 16,000g for 30 min at 4 °C. The standards—namely inosine 5′-monophosphate disodium salt (IMP, I4625), adenosine 5′-monophosphate disodium salt (AMP, 01930), guanosine 5′-monophosphate disodium salt (GMP, G8377), uridine 5′-monophosphate disodium salt (UMP, U6375) and cytidine 5′-monophosphate disodium salt (CMP, C1131)—were obtained from Sigma-Aldrich. The samples were transferred to MS vials for LC–MS analysis by QTRAP 7500 System (SCIEX). Data were acquired and normalized to the internal standards and then analyzed with SCIEX OS software.

MS analysis of PRPS1 O-GlcNAcylation sites

The FLAG-PRPS1 protein isolated from HEK293T cells was digested with sequencing-grade trypsin (Promega, VA9000) at a 50:1 protein-to-trypsin ratio. The incubation process was carried out at 37 °C for a duration of 16 h. Subsequently, the samples were passed through Zip-Tips (Millipore, ZTC18). The LC–MS/MS was conducted using the Easy nLC-1000 system in combination with the Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Fisher Scientific)^30,33.

Animal experiments

Experiments involving animals adhered to the rules and standards established by the Georgetown University IACUC. The indicated 5 × 10⁶ H1299 cells were introduced into the flank area of male nude mice (Jackson Laboratory, 002019) by subcutaneous injection. Tumor volume was calculated by measuring the width and length of tumors based on the following formula: \({vo}{lume}=[{{length}{\rm{\times }}({width})}^{2}]/2\).

Next, 20 mg kg⁻¹ etoposide (Cayman Chemical Company, 12092) was administered to indicated groups intraperitoneally in 200 µl of PBS twice a week for 3 weeks (PBS for controls).

Statistics and reproducibility

Experiments were repeated independently by indicated biological replicates in the figure legends. Statistical analyses were performed using Prism 8 (GraphPad Software, https://www.graphpad.com). The statistical methods are indicated in corresponding figure legends. Two-tailed Student’s t-tests were applied to analyze data from two unpaired groups. Two-tailed Wilcoxon tests were applied to analyze data for two paired groups. The clinical data of human lung cancer subtypes were analyzed by Welch’s t-test. P values for comparisons are indicated in the figures. The sample size was estimated based on the variations and mean values without employing pre-established statistical techniques.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The source data files related to this study are supplied with this paper. The protein structures of PRPS1 were obtained from the Protein Data Bank (PDB) database (https://www.rcsb.org) with PBD IDs 2H06 and 8DBE. The clinical data of human lung cancer subtypes are accessible from the ULCAN platform (http://ualcan.path.uab.edu/) with OGT gene symbol and lung cancer pathological classification. Additional data or details about the paper can be obtained from the lead contact upon reasonable request. Materials produced in this research can also be accessed by reaching out to the lead contact. Source data are provided with this paper.

Code availability

This study did not generate any original code.

References

Pavlova, N. N., Zhu, J. & Thompson, C. B. The hallmarks of cancer metabolism: still emerging. Cell Metab. 34, 355–377 (2022).
Article CAS PubMed PubMed Central Google Scholar
Hanahan, D. Hallmarks of cancer: new dimensions. Cancer Discov. 12, 31–46 (2022).
Article CAS PubMed Google Scholar
Zhu, J. & Thompson, C. B. Metabolic regulation of cell growth and proliferation. Nat. Rev. Mol. Cell Biol. 20, 436–450 (2019).
Article CAS PubMed PubMed Central Google Scholar
Chen, J. et al. De novo nucleotide biosynthetic pathway and cancer. Genes Dis. https://doi.org/10.1016/j.gendis.2022.04.018 (2022).
Khorana, H. G., Fernandes, J. F. & Kornberg, A. Pyrophosphorylation of ribose 5-phosphate in the enzymatic synthesis of 5-phosphorylribose 1-pyrophosphate. J. Biol. Chem. 230, 941–948 (1958).
Article CAS PubMed Google Scholar
Kornberg, A., Lieberman, I. & Simms, E. S. Enzymatic synthesis and properties of 5-phosphoribosylpyrophosphate. J. Biol. Chem. 215, 389–402 (1955).
Article CAS PubMed Google Scholar
Hove-Jensen, B. et al. Phosphoribosyl diphosphate (PRPP): biosynthesis, enzymology, utilization, and metabolic significance. Microbiol. Mol. Biol. Rev. 81, e00040–16 (2017).
Article CAS PubMed Google Scholar
Li, S., Lu, Y., Peng, B. & Ding, J. Crystal structure of human phosphoribosylpyrophosphate synthetase 1 reveals a novel allosteric site. Biochem. J. 401, 39–47 (2007).
Article CAS PubMed Google Scholar
Nosal, J. M., Switzer, R. L. & Becker, M. A. Overexpression, purification, and characterization of recombinant human 5-phosphoribosyl-1-pyrophosphate synthetase isozymes I and II. J. Biol. Chem. 268, 10168–10175 (1993).
Article CAS PubMed Google Scholar
Chen, P. et al. Crystal and EM structures of human phosphoribosyl pyrophosphate synthase I (PRS1) provide novel insights into the disease-associated mutations. PLoS ONE 10, e0120304 (2015).
Article PubMed PubMed Central Google Scholar
Qian, X. et al. Conversion of PRPS hexamer to monomer by AMPK-mediated phosphorylation inhibits nucleotide synthesis in response to energy stress. Cancer Discov. 8, 94–107 (2018).
Article CAS PubMed Google Scholar
Jing, X. et al. Cell-cycle-dependent phosphorylation of PRPS1 fuels nucleotide synthesis and promotes tumorigenesis. Cancer Res. 79, 4650–4664 (2019).
Article CAS PubMed Google Scholar
Qiu, Z. et al. MicroRNA-124 reduces the pentose phosphate pathway and proliferation by targeting PRPS1 and RPIA mRNAs in human colorectal cancer cells. Gastroenterology 149, e1511 (2015).
Article Google Scholar
Li, X. et al. A splicing switch from ketohexokinase-C to ketohexokinase-A drives hepatocellular carcinoma formation. Nat. Cell Biol. 18, 561–571 (2016).
Article CAS PubMed PubMed Central Google Scholar
He, M., Chao, L. & You, Y. P. PRPS1 silencing reverses cisplatin resistance in human breast cancer cells. Biochem. Cell Biol. 95, 385–393 (2017).
Article CAS PubMed Google Scholar
Srivastava, S. et al. NOTCH1-driven UBR7 stimulates nucleotide biosynthesis to promote T cell acute lymphoblastic leukemia. Sci. Adv. 7, eabc9781 (2021).
Article CAS PubMed PubMed Central Google Scholar
Li, B. et al. Negative feedback-defective PRPS1 mutants drive thiopurine resistance in relapsed childhood ALL. Nat. Med. 21, 563–571 (2015).
Article CAS PubMed PubMed Central Google Scholar
Al-Maawali, A. et al. Prenatal growth restriction, retinal dystrophy, diabetes insipidus and white matter disease: expanding the spectrum of PRPS1-related disorders. Eur. J. Hum. Genet. 23, 310–316 (2015).
Article CAS PubMed Google Scholar
Delos Santos, K., Kwon, E. & Moon, N. S. PRPS-associated disorders and the Drosophila model of Arts syndrome. Int. J. Mol. Sci. 21, 4824 (2020).
Article PubMed PubMed Central Google Scholar
de Brouwer, A. P. et al. PRPS1 mutations: four distinct syndromes and potential treatment. Am. J. Hum. Genet. 86, 506–518 (2010).
Article PubMed PubMed Central Google Scholar
Hart, G. W. Nutrient regulation of signaling and transcription. J. Biol. Chem. 294, 2211–2231 (2019).
Article CAS PubMed PubMed Central Google Scholar
Hanover, J. A., Krause, M. W. & Love, D. C. Bittersweet memories: linking metabolism to epigenetics through O-GlcNAcylation. Nat. Rev. Mol. Cell Biol. 13, 312–321 (2012).
Article CAS PubMed Google Scholar
Slawson, C. & Hart, G. W. O-GlcNAc signalling: implications for cancer cell biology. Nat. Rev. Cancer 11, 678–684 (2011).
Article CAS PubMed PubMed Central Google Scholar
Ge, Y. et al. Target protein deglycosylation in living cells by a nanobody-fused split O-GlcNAcase. Nat. Chem. Biol. 17, 593–600 (2021).
Article CAS PubMed PubMed Central Google Scholar
Yang, X. & Qian, K. Protein O-GlcNAcylation: emerging mechanisms and functions. Nat. Rev. Mol. Cell Biol. 18, 452–465 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hardivillé, S. & Hart, G. W. Nutrient regulation of signaling, transcription, and cell physiology by O-GlcNAcylation. Cell Metab. 20, 208–213 (2014).
Article PubMed PubMed Central Google Scholar
Zhao, M. et al. Epithelial STAT6 O-GlcNAcylation drives a concerted anti-helminth alarmin response dependent on tuft cell hyperplasia and gasdermin C. Immunity 55, e625 (2022).
Article Google Scholar
Wang, Q. et al. Ventromedial hypothalamic OGT drives adipose tissue lipolysis and curbs obesity. Sci. Adv. 8, eabn8092 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zhu, Y. & Hart, G. W. Nutrient regulation of the flow of genetic information by O-GlcNAcylation. Biochem Soc. Trans. 49, 867–880 (2021).
Article CAS PubMed Google Scholar
Tan, W. et al. Posttranscriptional regulation of de novo lipogenesis by glucose-induced O-GlcNAcylation. Mol. Cell 81, e1897 (2021).
Article Google Scholar
Chatham, J. C., Zhang, J. & Wende, A. R. Role of O-linked N-acetylglucosamine protein modification in cellular (patho)physiology. Physiol. Rev. 101, 427–493 (2021).
Article CAS PubMed Google Scholar
Reily, C., Stewart, T. J., Renfrow, M. B. & Novak, J. Glycosylation in health and disease. Nat. Rev. Nephrol. 15, 346–366 (2019).
Article PubMed PubMed Central Google Scholar
Peng, C. et al. Regulation of the Hippo-YAP pathway by glucose sensor O-GlcNAcylation. Mol. Cell 68, e595 (2017).
Article Google Scholar
Xu, Q. et al. AMPK regulates histone H2B O-GlcNAcylation. Nucleic Acids Res 42, 5594–5604 (2014).
Article CAS PubMed PubMed Central Google Scholar
Zhu, Y. & Hart, G. W. Targeting O-GlcNAcylation to develop novel therapeutics. Mol. Aspects Med. 79, 100885 (2021).
Article CAS PubMed Google Scholar
Hay, N. Reprogramming glucose metabolism in cancer: can it be exploited for cancer therapy? Nat. Rev. Cancer 16, 635–649 (2016).
Article CAS PubMed PubMed Central Google Scholar
Yi, W. et al. Phosphofructokinase 1 glycosylation regulates cell growth and metabolism. Science 337, 975–980 (2012).
Article CAS PubMed PubMed Central Google Scholar
Liu, R. et al. Innate immune response orchestrates phosphoribosyl pyrophosphate synthetases to support DNA repair. Cell Metab. 33, e2079 (2021).
Article Google Scholar
Navas, L. E. & Carnero, A. NAD⁺ metabolism, stemness, the immune response, and cancer. Signal Transduct. Target. Ther. 6, 2 (2021).
Article CAS PubMed PubMed Central Google Scholar
Minato, K. et al. Characterization of an etoposide-resistant human small-cell lung cancer cell line. Cancer Chemother. Pharm. 26, 313–317 (1990).
Article CAS Google Scholar
Chandrashekar, D. S. et al. UALCAN: an update to the integrated cancer data analysis platform. Neoplasia 25, 18–27 (2022).
Article CAS PubMed PubMed Central Google Scholar
Puusepp, S. et al. Atypical presentation of Arts syndrome due to a novel hemizygous loss-of-function variant in the PRPS1 gene. Mol. Genet. Metab. Rep. 25, 100677 (2020).
CAS PubMed PubMed Central Google Scholar
Mittal, R. et al. Association of PRPS1 mutations with disease phenotypes. Dis. Markers 2015, 127013 (2015).
Article PubMed PubMed Central Google Scholar
de Brouwer, A. P. et al. Arts syndrome is caused by loss-of-function mutations in PRPS1. Am. J. Hum. Genet. 81, 507–518 (2007).
Article PubMed PubMed Central Google Scholar
Akella, N. M., Ciraku, L. & Reginato, M. J. Fueling the fire: emerging role of the hexosamine biosynthetic pathway in cancer. BMC Biol. 17, 52 (2019).
Article CAS PubMed PubMed Central Google Scholar
Dibble, C. C. & Manning, B. D. Signal integration by mTORC1 coordinates nutrient input with biosynthetic output. Nat. Cell Biol. 15, 555–564 (2013).
Article CAS PubMed PubMed Central Google Scholar
Bar-Peled, L. & Sabatini, D. M. Regulation of mTORC1 by amino acids. Trends Cell Biol. 24, 400–406 (2014).
Article CAS PubMed PubMed Central Google Scholar
Gabriel, S. S. & Kallies, A. Glucose- and glutamine-fueled stabilization of C-Myc is required for T-cell proliferation and malignant transformation. Cell Death Discov. 2, 16047 (2016).
Article CAS PubMed PubMed Central Google Scholar
Shi, D. D. et al. De novo pyrimidine synthesis is a targetable vulnerability in IDH mutant glioma. Cancer Cell 40, e916 (2022).
Article Google Scholar
Mercati, O. et al. PRPS1 loss-of-function variants, from isolated hearing loss to severe congenital encephalopathy: new cases and literature review. Eur. J. Med. Genet. 63, 104033 (2020).
Article PubMed Google Scholar
Ran, F. A. et al. Genome engineering using the CRISPR–Cas9 system. Nat. Protoc. 8, 2281–2308 (2013).
Article CAS PubMed PubMed Central Google Scholar
Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).
Article CAS PubMed PubMed Central Google Scholar
Doench, J. G. et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR–Cas9. Nat. Biotechnol. 34, 184–191 (2016).
Article CAS PubMed PubMed Central Google Scholar
Zhu, Q. et al. O-GlcNAcylation promotes pancreatic tumor growth by regulating malate dehydrogenase 1. Nat. Chem. Biol. 18, 1087–1095 (2022).
Article CAS PubMed Google Scholar
Nie, H. et al. O-GlcNAcylation of PGK1 coordinates glycolysis and TCA cycle to promote tumor growth. Nat. Commun. 11, 36 (2020).
Article CAS PubMed PubMed Central Google Scholar
Rao, X. et al. O-GlcNAcylation of G6PD promotes the pentose phosphate pathway and tumor growth. Nat. Commun. 6, 8468 (2015).
Article CAS PubMed Google Scholar
Wang, Y. et al. O-GlcNAcylation destabilizes the active tetrameric PKM2 to promote the Warburg effect. Proc. Natl Acad. Sci. USA 114, 13732–13737 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hvorecny, K. L., Hargett, K., Quispe, J. D. & Kollman, J. M. Human PRPS1 filaments stabilize allosteric sites to regulate activity. Nat. Struct. Mol. Biol. 30, 391–402 (2023).
Article CAS PubMed PubMed Central Google Scholar
Deng, M. et al. Deubiquitination and activation of AMPK by USP10. Mol. Cell 61, 614–624 (2016).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank A. Cheema and S. Bansal from Metabolomics Shared Resource of Georgetown University Medical Center (partially funded by National Institutes of Health (NIH)/National Cancer Institute (NCI)/Cancer Center Support Grant (CCSG) P30-CA051008) for their contributions to the mass spectrometry assays and analyses. G.M.K. was supported by NIH grant RO1-CA68135. H.P. was supported by NIH grant RO1-NS121243.

Author information

Authors and Affiliations

Cancer Center, Renmin Hospital of Wuhan University, Wuhan, China
Lulu Chen, Qi Zhou, Pingfeng Zhang, Yingge Li & Qibin Song
Department of Oncology, Georgetown Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC, USA
Lulu Chen, Qi Zhou, Yingge Li, Ziwen Xu, Junfeng Ma, Gary M. Kupfer & Huadong Pei
Department of Lymphoma and Myeloma, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Wei Tan
Center for Cancer and Immunology, Brain Tumor Institute, Children’s National Health System, Washington, DC, USA
Yanxin Pei

Authors

Lulu Chen
View author publications
You can also search for this author in PubMed Google Scholar
Qi Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Pingfeng Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Wei Tan
View author publications
You can also search for this author in PubMed Google Scholar
Yingge Li
View author publications
You can also search for this author in PubMed Google Scholar
Ziwen Xu
View author publications
You can also search for this author in PubMed Google Scholar
Junfeng Ma
View author publications
You can also search for this author in PubMed Google Scholar
Gary M. Kupfer
View author publications
You can also search for this author in PubMed Google Scholar
Yanxin Pei
View author publications
You can also search for this author in PubMed Google Scholar
Qibin Song
View author publications
You can also search for this author in PubMed Google Scholar
Huadong Pei
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

L.C. performed most of the experiments and statistical analysis and wrote the manuscript. Q.Z. helped with human tissue sample collection and analysis. P.Z. contributed to protein structure modeling and helped with size-exclusion chromatography. Z.X. contributed to the preparation of reagents. W.T. and Y.L. contributed to the experimental design. G.M.K., J.M. and Y.P. read the manuscript and made substantive revisions. Q.S. helped with human tissue sample collection and analysis and with experimental design. H.P. conceived, designed and supervised the study, analyzed data and wrote and revised the manuscript. All authors reviewed the results and approved the final version of the manuscript.

Corresponding authors

Correspondence to Qibin Song or Huadong Pei.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Chen, L., Zhou, Q., Zhang, P. et al. Direct stimulation of de novo nucleotide synthesis by O-GlcNAcylation. Nat Chem Biol 20, 19–29 (2024). https://doi.org/10.1038/s41589-023-01354-x

Download citation

Received: 15 November 2022
Accepted: 03 May 2023
Published: 12 June 2023
Issue Date: January 2024
DOI: https://doi.org/10.1038/s41589-023-01354-x

This article is cited by

Role of O-GlcNAcylation in Central Nervous System Development and Injuries: A Systematic Review
- Li Zhang
- Wanshan Bai
- Mi Tian
Molecular Neurobiology (2024)
O-GlcNAcylation: A Crucial Regulator in Cancer-Associated Biological Events
- Zhihong Ran
- Lei Zhang
- Qibin Song
Cell Biochemistry and Biophysics (2023)