NEAT1 drives SARS-CoV-2 N protein–induced inflammation, metabolic reprogramming, and mitochondria–ER stress crosstalk

Abstract

SARS-CoV-2 remains a global health concern. Although its nucleocapsid (N) protein supports viral replication and evades host immunity, its role in host metabolic reprogramming and organelle homeostasis is not fully understood. This study aims to elucidate whether the N protein modulates glycolysis and mitochondrial-ER stress crosstalk via the long non-coding RNA NEAT1. We established human bronchial epithelial (HBE) cells stably expressing the N protein. Inflammatory and glycolytic gene expression was analyzed by qRT–PCR and Western blot. ROS levels were measured by flow cytometry, while mitochondrial membrane potential, Ca²⁺ overload, and mitochondria-ER contact sites (MAMs) were assessed by confocal microscopy. NEAT1 knockdown and HK2–VDAC1 interaction studies were performed to explore underlying mechanisms. The N protein induced inflammatory responses, enhanced LPS sensitivity, and triggered mitochondrial dysfunction, ER stress, and MAM formation. It promoted glycolytic reprogramming by upregulating key enzymes (GLUT1, HK2, PKM2). NEAT1 was essential for these effects—N protein increased NEAT1 expression, and NEAT1 knockdown attenuated inflammation, glycolysis, and mitochondrial damage. Mechanistically, NEAT1 silencing restored HK2–VDAC1 association and suppressed VDAC1 oligomerization. The SARS-CoV-2 N protein exacerbates inflammation through a NEAT1-dependent mechanism that drives glycolytic reprogramming and disrupts mitochondrial-ER homeostasis.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-40957-x.

Introduction

The emergence of Coronavirus Disease 2019 (COVID-19) has significantly impacted global public health¹. SARS-CoV-2 exhibits rapid transmissibility and triggers severe inflammatory responses that may progress to acute respiratory distress syndrome (ARDS) and multi-organ failure^2–4. The viral nucleocapsid (N) protein, a structural core component, coordinates viral replication and subverts host immunity. Critically, the N protein directly interacts with the NLRP3 inflammasome⁵, activating IL-1β maturation (p17) and Caspase-1 cleavage (p20)⁶, thereby amplifying the inflammatory cascade characteristic of SARS-CoV-2 infection.

During viral replication, infected cells undergo metabolic reprogramming characterized by a ‘Warburg effect’—shifting from oxidative phosphorylation to high-throughput glycolysis to rapidly generate ATP and biosynthetic precursors⁷. SARS-CoV-2 infection elevates lactate production and glycolytic flux⁸, fueling viral replication and inflammatory mediator synthesis. However, whether the N protein directly mediates this metabolic shift remains unexplored.

Viral infections induce oxidative stress, disrupting mitochondrial calcium (Ca²⁺) homeostasis and increasing reactive oxygen species (ROS) production, ultimately causing mitochondrial damage^9,10. Notably, the SARS-CoV-2 ORF3 non-structural protein modulates endoplasmic reticulum (ER) stress via autophagy-related receptor pathways¹¹. Mitochondria-associated membranes (MAMs) serve as platforms for Ca²⁺/ROS crosstalk, regulating immune-metabolic homeostasis¹². Interactome studies reveal enrichment of multiple SARS-CoV-2 proteins in MAMs, disrupting Ca²⁺ flux and exacerbating ROS generation, thereby activating ER stress and downstream inflammatory amplification.

Long non-coding RNA (lncRNA) NEAT1 is a structural molecule that assembles paraspeckles in the nucleus¹³. Recent studies have demonstrated that NEAT1 is capable of upregulating key glycolytic enzymes, including HK2 and GLUT1, through binding to microRNAs or altering chromatin accessibility^14–16. This process contributes to the aerobic glycolysis process, which in turn promotes tumour cell proliferation and migration. Furthermore, NEAT1 has been demonstrated to activate the NLRP3 inflammasome, thereby promoting the secretion of inflammatory factors such as IL-6 and TNF-α, thus playing a significant role in infections and immune-related diseases¹⁷. However, there is currently a paucity of evidence regarding the potential involvement of NEAT1 in the metabolism-inflammation coupling associated with SARS-CoV-2 infection. Recent studies have indicated that the inhibition of NEAT1 can restore mitochondrial function and improve ROS imbalance through the PINK1-Parkin pathway, thereby alleviating mitochondrial damage¹⁸.

Based on this evidence, we hypothesize that the N protein amplifies inflammation by upregulating NEAT1, which orchestrates glycolytic reprogramming and mitochondrial-ER stress crosstalk. This study elucidates the molecular mechanism underlying N protein–driven inflammation via NEAT1-mediated metabolic and organelle dysregulation, revealing novel therapeutic targets for severe COVID-19.

Materials and methods

Cell culture

Human bronchial epithelial (HBE) cells (kindly provided by Prof. Ying Ying, First Affiliated Hospital, Jiangxi Medical College, Nanchang University, Nanchang, China) were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) at 37 °C under 5% CO₂. Prior to stimulation, cells were serum-starved for 4 h in medium containing 0.5% FBS, followed by treatment with 20 µg/mL lipopolysaccharide (LPS) for 12 h. Cells were then harvested for downstream analyses. All experiments were performed in triplicate.

Lentiviral transfection and cell grouping

Lentiviruses encoding SARS-CoV-2 N protein (LV-N) and control vector (LV-CT) were obtained from Anhui General Biotechnology Co., Ltd. siRNA targeting NEAT1 (siNEAT1) and negative control siRNA (siNC) were synthesized by Jima Gene (Suzhou) Co., Ltd. (sequences in Supplementary Table 1). For lentiviral transduction, 1 × 10⁵ HBE cells/well were seeded in 6-well plates. At 30–50% confluence, cells were transfected with lentivirus (MOI = 30, titer: ~1 × 10⁸ TU/mL) and 8 µg/mL Polybrene (TR-1003, Sigma-Aldrich) for 24 h. Stable pools were selected using 5 µg/mL puromycin (A1113803, Thermo Fisher) for 48 h post-transfection.

Reactive oxygen species (ROS) detection

Intracellular ROS levels were measured using 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA; Beyotime Biotech). After 12 h LPS (20 µg/mL) treatment, cells were incubated with 10 µM DCFH-DA at 37 °C for 30 min in the dark. Fluorescence intensity was quantified by flow cytometry (BD FACSCalibur).

Quantitative real-time PCR (qRT-PCR)

Total RNA was extracted with TRIzol™ (Invitrogen, 15596026CN) and reverse-transcribed using Hifair^® II 1st Strand cDNA Synthesis SuperMix (Yeasen, 11120ES). qPCR was performed with SYBR Green Master Mix (Yeasen, 11199ES) under cycling conditions: 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. β-Actin (ACTB) served as the endogenous control. Primer sequences are provided in Supplementary Table 2. All reactions were run in triplicate.

Western blotting

Cells were lysed in RIPA buffer (Solarbio) at 4 °C and centrifuged at 12,000 ×g for 15 min. Protein concentrations were determined by BCA assay. Proteins were separated by SDS-PAGE, transferred to PVDF membranes, and blocked with 5% non-fat milk for 1 h at room temperature. Primary antibodies included: SARS-CoV-2 N (#33336, 1:1000, CST), GRP78 (11587-1-AP, 1:1000, Proteintech), CHOP (81482-1-RR, 1:1000, Proteintech), GLUT1 (21829-1-AP, 1:1000, Proteintech), HK2 (66974-1-Ig, 1:1500, Proteintech), PKM2 (15822-1-AP, 1:1500, Proteintech), β-actin (20536-1-AP, 1:1000, Proteintech). Membranes were incubated with HRP-conjugated anti-rabbit (SA00001-2, 1:5000) or anti-mouse (SA00001-1, 1:5000) secondary antibodies (Proteintech) for 1 h. Signals were detected using an ECL imaging system.

Mitochondrial calcium and transmembrane potential analysis

Assessment of mitochondrial calcium levels and mitochondrial transmembrane potential was performed using the fluorescent probes Rhod-2 (Yeasen) and JC-1 (Beyotime Biotech), respectively, with fluorescence visualization carried out on a Leica confocal microscope. Specifically, Rhod-2 was applied at a working concentration of 5 µmol/L and JC-1 at 2 µmol/L. Following loading with the probes, the cells were incubated at 37 °C in the dark for 30 min. Thereafter, they were washed with PBS in order to remove any extracellular probes. The cells were then placed in confocal-specific dishes containing culture medium for observation. For Rhod-2, the excitation wavelength was set at 552 nm and the emission at 581 nm. For JC-1, the monomeric form was excited at 514 nm with emission at 529 nm, while the aggregated form emitted at 590 nm. The images were acquired using a Leica TCS SP8 confocal microscope equipped with a 63× oil immersion objective, and the fluorescence signals were quantified with LAS X software.

ER and mitochondria contact analysis

The labelling of cells was conducted using MitoTracker™ Green FM and ER-Tracker™ Red (both from Invitrogen, Thermo Fisher Scientific Inc.). Both probes were added to the cell culture medium at working concentrations of 100nM and 1µM, respectively, followed by co-incubation at 37 °C in the dark for 30 min. Following the incubation period, the cells were washed thrice with pre-warmed phosphate-buffered saline (PBS) in order to remove any unbound probes. Thereafter, the cells were transferred to phenol-free imaging medium. Mitochondrial and endoplasmic reticulum fluorescence signals were acquired using a Leica TCS SP8 confocal microscope, with excitation/emission wavelengths of 490/516 nm for MitoTracker™ Green FM and 587/615 nm for ER-Tracker™ Red. The images were captured using a 63× oil immersion objective, and multi-channel synchronous acquisition and subsequent analysis were performed with LAS X software.

Extracellular acidification rate (ECAR)

The extracellular acidification rate (ECAR) was measured using a commercial assay kit (Elabscience, Cat# E-BC-F069) in accordance with the manufacturer’s instructions. In summary, following a 12-hour LPS treatment, cells were seeded at a density of 1 × 10⁵ cells per well into a 96-well black-walled clear-bottom plate and allowed to adhere under conditions of 37 °C and 5% CO₂. The employment of blank wells containing solely the probe working solution was undertaken for the purpose of background correction. Prior to detection, the culture medium was removed, and 200 µL of probe working solution was added to each well. The fluorescence microplate reader was preheated to 37 °C. Fluorescence intensity was recorded at intervals of 2–3 min over a period of 100–120 min, employing excitation and emission wavelengths of 490 nm and 535 nm, respectively. The ECAR was calculated based on the linear phase of the fluorescence kinetic curve.

Statistical analysis

Data are presented as mean ± SD from ≥ 3 independent experiments. Statistical significance was determined using Student’s t-test (two groups) or one-way ANOVA (multiple groups) in GraphPad Prism 9.5. p < 0.05 was considered significant.

Result

N protein potentiates inflammatory responses

The stable expression of the SARS-CoV-2 N protein in HBE cells was established via lentiviral transduction, and this was validated by Western blotting and qRT-PCR (Fig. 1.A-B). Overexpression of the N protein alone led to a significant increase in the mRNA levels of the pro-inflammatory cytokines TNF-α and IL-6 when compared with vector-transduced controls. A subsequent analysis of anti-inflammatory cytokines revealed a slight upregulation in IL-1RA mRNA, whereas IL-10 expression remained unchanged (see Fig S1.B-C).In accordance with the results of the transcriptional analysis, ELISA evaluation of cell culture supernants demonstrated a corresponding augmentation in the secretion of TNF‑α and IL‑6 in N‑protein‑expressing cells. Furthermore, upon stimulation with LPS (20 µg/mL, 12 h), N-protein-overexpressing cells exhibited a further synergistic enhancement of the inflammatory response, which was also confirmed at the protein level by ELISA (see Fig. 1.C-D and S1.A).Collectively, these results indicate that the N protein possesses intrinsic pro-inflammatory activity and also primes epithelial cells for an augmented response to exogenous inflammatory stimuli, thereby amplifying downstream inflammatory cascades.

Fig. 1.

Overexpression of SARS-CoV-2 N protein enhances inflammatory response in HBE cells. (A–B), N protein expression was validated at the mRNA and protein levels in HBE cells transduced with lentiviral vector carrying the SARS‑CoV‑2 N gene (OE‑N), compared with wild‑type (WT) and empty‑vector control cells. (C–D), Secretion levels of IL‑6 and TNF‑α detected by ELISA in cell culture supernatants from control (Con), N‑overexpressing (OE‑N), LPS‑treated, and OE‑N + LPS‑treated groups. Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.

N protein induces MAMs formation and exacerbates ER-mitochondrial stress

High-resolution confocal imaging revealed a marked increase in mitochondrial-endoplasmic reticulum contact sites (MAMs) within HBE cells overexpressing N protein, a phenomenon that was further amplified following LPS treatment (Fig. 2A). This finding indicates that N protein promotes structural coupling between these two organelles. Concurrently, both N protein overexpression and combined LPS treatment led to a significant elevation in intracellular reactive oxygen species levels (Fig. 2B). Furthermore, immunofluorescence analysis revealed a substantial accumulation of mitochondrial calcium overload in N protein-overexpressing cells (Fig. 2C), accompanied by a marked reduction in mitochondrial membrane potential (ΔΨm) (Fig. 2.D). Collectively, these results indicate that N protein induces significant mitochondrial dysfunction.

Fig. 2.

Overexpression of N protein promotes MAMs formation and induces mitochondrial and endoplasmic reticulum stress. (A), Representative immunofluorescence images showing mitochondria–endoplasmic reticulum contact sites (MAMs) in control, OE‑N, LPS‑treated and OE‑N + LPS‑treated cells. MAMs formation was quantified using Pearson’s correlation coefficient. (B), Flow cytometry analysis of mitochondrial ROS levels in the indicated groups. (C), Mitochondrial Ca²⁺ load detected by Rhod‑2 immunofluorescence and quantified as mean fluorescence intensity. (D), Mitochondrial membrane potential measured by JC‑1 staining (red/green fluorescence ratio). (E-F), Expression levels of ER stress markers GRP78 and CHOP were analyzed by Western blot and quantified by densitometry. Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.

Concomitantly, N protein expression induced endoplasmic reticulum (ER) stress, as evidenced by robust upregulation of the ER stress chaperone GRP78 and the pro-apoptotic transcription factor CHOP (Fig. 2.E-F). Under LPS stimulation, both mitochondrial injury and ER stress were further exacerbated. Taken together, these data demonstrate that the SARS-CoV-2 N protein promotes MAM formation, disrupts mitochondrial homeostasis, and triggers ER stress, thereby intensifying intracellular stress signaling.

N protein promotes glycolytic reprogramming

Given that organelle contact remodeling and mitochondrial dysfunction are frequently coupled with metabolic adaptation, we next investigated whether N protein expression influences cellular metabolism. Results showed that N protein‑overexpressing HBE cells exhibited a significant increase in extracellular lactate accumulation, indicating enhanced glycolytic flux (Fig. 3A). Extracellular acidification rate (ECAR) measurement further confirmed that N protein overexpression combined with LPS stimulation markedly elevated ECAR levels, suggesting heightened glycolytic activity (Fig. 3B). In addition, the expression of key glycolytic regulators—GLUT1, HK2, and PKM2—was upregulated by the N protein at both mRNA and protein levels, and this effect was further potentiated upon LPS treatment (Fig. 3C–E). Together, these findings indicate that the N protein drives glycolytic reprogramming in airway epithelial cells, likely as part of a coordinated stress‑metabolic response.

Fig. 3.

N protein promotes glycolytic reprogramming in airway epithelial cells. (A), Extracellular lactate levels in control, OE‑N, LPS‑treated, and OE‑N + LPS‑treated cells. (B), ECAR measured in the indicated groups. (C), mRNA expression levels of GLUT1, HK2, and PKM2. (D–E), Protein expression levels of GLUT1, HK2, and PKM2 analyzed by Western blot and quantified by densitometry. Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.

Silencing NEAT1 attenuates N protein-driven inflammation

In order to elucidate the upstream molecular hubs that mediate cellular stress and metabolic reprogramming, a screening of long non-coding RNAs (lncRNAs) was conducted, with the focus being on those previously implicated in SARS-CoV-2 infection and inflammation^19–23. qPCR analysis revealed differential expression of multiple lncRNAs, including MALAT1, NEAT1, DANCR, and CAS5, in HBE cells overexpressing the N protein. Among these, NEAT1 showed the most pronounced upregulation (Fig. 4A). NEAT1, a core component of paraspeckles, has been extensively documented as playing a pivotal role in cellular stress responses and inflammatory regulation^24–26. To further investigate this finding, we performed RNA interference to silence NEAT1. The results demonstrated that NEAT1 knockdown significantly suppressed the N protein-induced inflammatory response, as evidenced by reduced mRNA levels of inflammatory cytokines (Fig. 4B,C). In line with these findings, ELISA analysis of cell culture supernants confirmed that NEAT1 silencing also inhibited the secretion of inflammatory factors triggered by the N protein (Fig. 4D,E). Collectively, these results indicate that NEAT1 likely functions as a pivotal downstream mediator of the N protein, orchestrating subsequent cellular phenotypic alterations.

Fig. 4.

NEAT1 knockdown attenuates N protein‑induced inflammatory response in HBE cells. (A), Differential expression of lncRNAs (MALAT1, NEAT1, DANCR, CAS5) in control and N‑overexpressing (OE‑N) cells. (B), NEAT1 mRNA expression after siRNA‑mediated knockdown. (C), mRNA levels of IL‑6 and TNF‑α in control, OE‑N, and NEAT1‑knockdown (si‑NEAT1) groups. (D–E), Secretion levels of IL‑6 and TNF‑α in cell culture supernatants measured by ELISA.Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Knocking down NEAT1 reverses N-protein-induced glycolysis and organelle stress

The depletion of NEAT1 has been demonstrated to attenuate N protein-induced inflammatory signalling, whilst concomitantly reversing the associated metabolic phenotype. Specifically, NEAT1 knockdown resulted in a significant reduction in extracellular lactate accumulation and suppression of the N protein-driven elevation in extracellular acidification rate (ECAR) (Fig. 5.A-B), indicating a reversal of N protein-mediated glycolytic reprogramming. Concurrently, protein expression of key glycolytic enzymes HK2, GLUT1, and PKM2 was downregulated upon NEAT1 silencing (Fig. 5.C), a trend that was further confirmed by quantitative analysis (Fig. 5.E). Furthermore, NEAT1 knockdown has been shown to alleviate endoplasmic reticulum stress, as evidenced by a decrease in the levels of the ER stress markers GRP78 and CHOP (Fig. 5.D). The quantitative data corresponding to these findings is presented in Fig. 5F. At the level of organelle interaction, NEAT1 depletion significantly reduced the formation of mitochondria-endoplasmic reticulum contact sites (MAMs) (Fig. 5.G) and partially restored mitochondrial function. Intracellular ROS accumulation was diminished (Fig. 5.H), mitochondrial Ca²⁺ overload was mitigated (Fig. 5I), and the loss of mitochondrial membrane potential was attenuated (Fig. 5.J). Collectively, these findings identify NEAT1 as a critical mediator of N protein-triggered metabolic dysregulation, ER stress, and mitochondrial dysfunction.

Fig. 5.

NEAT1 knockdown rescues N protein‑induced metabolic and organelle stress responses. (A), Extracellular lactate levels in control, OE‑N, and NEAT1‑knockdown (si‑NEAT1) groups. (B), Extracellular acidification rate (ECAR) in the indicated groups. (C), Protein expression of glycolytic enzymes HK2, GLUT1, and PKM2 analyzed by Western blot. (D), Protein expression of ER stress markers GRP78 and CHOP. (E,F), Quantitative analysis of protein levels. (G), Representative immunofluorescence images showing mitochondria–endoplasmic reticulum contact sites (MAMs) and their quantification using Pearson’s correlation coefficient. (H), Intracellular ROS levels measured by flow cytometry. (I), Mitochondrial Ca²⁺ load detected by Rhod‑2 fluorescence and quantified as mean fluorescence intensity. (J), Mitochondrial membrane potential assessed by JC‑1 staining (red/green fluorescence ratio). Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.

NEAT1 regulates HK2-VDAC1 interaction and VDAC1 oligomerisation

In order to further delineate the manner in which NEAT1 links metabolic regulation to mitochondrial integrity, an investigation was conducted into the interaction between hexokinase 2 (HK2) and the mitochondrial outer membrane channel protein VDAC1. The latter is a complex known to couple glycolysis to mitochondrial function. Immunofluorescence analysis revealed that NEAT1 knockdown markedly enhanced the co-localisation of HK2 and VDAC1 on mitochondria (Fig. 6A). As demonstrated in Fig. 6B, co-immunoprecipitation assays consistently confirmed increased binding between HK2 and VDAC1 upon NEAT1 silencing. This finding suggests that there is a reduction in the dissociation of HK2 from the mitochondrial surface. Furthermore, NEAT1 knockdown effectively suppressed VDAC1 oligomerization (Fig. 6C). It is evident that VDAC1 oligomerization represents a pivotal step in the opening of the mitochondrial permeability transition pore (mPTP), a process that is closely associated with mitochondrial dysfunction and apoptosis. These findings establish a mechanistic link through which NEAT1 promotes N protein–induced mitochondrial injury. Consequently, NEAT1 appears to facilitate mitochondrial stress and metabolic imbalance, at least in part, by destabilising the HK2–VDAC1 interaction and promoting VDAC1 oligomerisation.

Fig. 6.

NEAT1 regulates mitochondrial HK2–VDAC1 interaction and VDAC1 oligomerization. (A), Representative immunofluorescence images and quantification of HK2 and VDAC1 co‑localization on mitochondria in control (Empty vector + siNC), OE‑N+siNC, and NEAT1‑knockdown (si‑NEAT1) cells. Co‑localization was analyzed using Pearson’s correlation coefficient. (B), Co‑immunoprecipitation assay showing the interaction between HK2 and VDAC1 in the indicated groups. (C), Detection of VDAC1 oligomerization by non‑reducing Western blot. Data are presented as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

This study establishes that SARS-CoV-2 N protein exacerbates inflammation through NEAT1-mediated glycolytic reprogramming and organelle stress crosstalk (Fig. 7). These findings reveal how viruses hijack host metabolic and organellar networks to amplify pathology, suggesting NEAT1 as a potential therapeutic target for COVID-19 hyperinflammation.

Fig. 7.

Diagram of the inflammatory mechanism promoted by N protein. This diagram summarises the core pathogenic pathway of the N protein. By upregulating the long non-coding RNA NEAT1, the N protein simultaneously promotes the expression of key glycolytic enzymes (GLUT1, HK2 and PKM2), thereby influencing cellular metabolic levels. It also facilitates the formation of MAM structures, weakening the HK2-VDAC1 interaction and promoting VDAC1 oligomerisation. This triggers mitochondrial and endoplasmic reticulum stress. Ultimately, this induces a robust inflammatory response.

Notably, current research on the SARS-CoV-2 N protein has primarily focused on its structural characteristics, viral replication mechanisms, and direct regulation of the host immune system. The N protein can competitively bind to STAT1/STAT2, blocking their interaction with JAK1/TYK2, thereby inhibiting the phosphorylation and nuclear translocation of STAT1/STAT2. This suppresses the expression of interferon-stimulated genes (ISGs), weakens the host’s antiviral immune response, and significantly enhances viral replication²⁷. Furthermore, the N protein can bind to the helicase domain of the host pattern recognition receptor RIG-I, reducing RIG-I’s ability to recognise viral RNA, decreasing IFN-β production, and further promoting viral immune escape²⁸. Additionally, studies have shown that the N protein binds to the host stress granule core protein G3BP1 with high affinity, stripping it from the stress granules, preventing stress granule assembly, and relieving the inhibitory effect of stress granules on viral mRNA translation, thereby promoting viral protein translation and viral replication²⁹. In contrast, its regulation of host metabolic and organelle dynamics remains poorly characterized. Our work addresses this gap by demonstrating that N protein upregulates NEAT1 to synchronize inflammatory, metabolic, and organellar dysregulation, thereby expanding its known pathogenic mechanisms.

Beyond immune evasion, viruses commonly reprogram host metabolism. The balance of cellular metabolism is maintained by several key regulatory factors, including hypoxia-inducible factor 1α (HIF-1α) and p53, which influence energy metabolism, oxidative stress, and amino acid metabolism regulation by altering the balance between glycolysis and OXPHOS³⁰. Previous studies have reported that NSP5, as the main viral protease, can inhibit the transcriptional activity of p53 and thereby regulate cellular glucose metabolism³¹. In addition, glucose uptake by Vero E6 cells infected with SARS-CoV-2 increased, and the levels of glucose transporters GLUT1, GLUT3, and GLUT4 significantly increased³². Similar to Vero cells, SARS-CoV-2 replication in Calu-3 cells depends on aerobic glycolysis and glutamine degradation, and inhibition of these pathways limits viral production³³. In this study, we report that the SARS-CoV-2 N protein can enhance cellular aerobic glycolysis by upregulating the expression of key enzymes such as HK2/PKM2/GLUT1. Additionally, Su et al. reported that the long non-coding RNA NEAT1 can upregulate pyruvate dehydrogenase kinase (PDK1) through the WNT/β-catenin signalling pathway in cervical cancer cells, thereby enhancing glycolysis and promoting cervical cancer metastasis³⁴. Our study preliminarily explored the role of NEAT1 in promoting glycolysis in cells overexpressing the N protein, although the underlying mechanisms remain unclear.

Concurrently, organelle stress emerges as critical in COVID-19 pathogenesis. Studies have shown that SARS-CoV-2 infection causes mitochondrial dynamics imbalance and regulates the binding of mitochondrial autophagy-related proteins (such as p62 and LC3 proteins) and Spike protein expression, leading to damage to the host cell’s mitochondrial autophagy function. This promotes sustained viral replication, increased IL-18 expression, and SARS-CoV-2 infection-related pulmonary inflammatory responses^35,36. Furthermore, Youn et al. found that interfering with SARS-CoV-2-induced mitochondrial oxidative stress significantly reduces endothelial dysfunction³⁷. Additionally, Shin et al. observed that viral replication decreases when Grp78 is inhibited, indicating the importance of ER stress in the pathogenesis of SARS-CoV-2³⁸, which is consistent with our findings. Notably, analysis of blood samples collected post-COVID-19 revealed that SARS-CoV-2 infection significantly downregulates or alters the expression of a series of miRNAs, such as miR-355-5p, which are associated with ER stress activation or UPR sensors³⁹. In summary, the study highlights that the replication and pathogenic mechanisms of the novel coronavirus after invading the host are closely associated with the imbalance of organelle homeostasis. However, organelles are not isolated from one another; we found that the novel coronavirus N protein promotes increased formation of MAMs, facilitating the transfer of calcium ions from the endoplasmic reticulum to mitochondria, thereby causing mitochondrial homeostasis imbalance and dysfunction.

A significant extension of these findings is the observation that NEAT1 knockdown restores the interaction between HK2 and VDAC1 on the mitochondrial outer membrane and inhibits VDAC1 oligomerisation. HK2 acts as both a rate-limiting glycolytic enzyme and a structural regulator of mitochondrial integrity upon binding to VDAC1, thereby contributing to the stabilisation of mitochondrial function. Conversely, existing research confirms that HK2 is extensively localised at the mitochondrial-endoplasmic reticulum membrane (MAMs). Dissociation of HK2 from VDAC1 triggers substantial calcium ion transfer between the endoplasmic reticulum and mitochondria by affecting the IP3R-GRP75-VDAC1 complex, thereby inducing mitochondrial calcium overload^40,41. Furthermore, dissociation of HK2 from VDAC1 promotes VDAC1 oligomerisation. This leads to the excessive opening of the mitochondrial permeability transition pore (mPTP), which exacerbates ROS-induced cellular damage and promotes cell death-related signalling pathways⁴². Our data suggest that N protein-induced NEAT1 disrupts this protective HK2–VDAC1 coupling, favouring a state in which glycolysis is increased but the mitochondrial membrane becomes more vulnerable. This provides a unifying model in which NEAT1 increases glycolytic flux and ‘uncouples’ glycolysis from mitochondrial protection, permitting metabolic activation and mitochondrial distress to occur simultaneously.

Three limitations merit consideration. First, validation in animal models and clinical specimens is essential. Second, the mechanism underlying N protein-induced NEAT1 upregulation requires clarification. Third, NEAT1’s crosstalk with other inflammatory pathways remains unexplored. Addressing these questions may yield therapies targeting viral-metabolic-inflammation networks.

Conclusion

This study elucidates a novel mechanism wherein SARS-CoV-2 N protein exacerbates inflammation through NEAT1-mediated coordination of glycolytic reprogramming and organelle stress crosstalk. These findings advance our understanding of COVID-19 pathogenesis and identify NEAT1 as a promising therapeutic target for anti-inflammatory interventions.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Cheng Qing* & Huaigang Chen*: Writing – review & editing, Writing – original draft; Shuying Huang: Data curation, Conceptualization; Jianguo Zhang: Data curation, Conceptualization; Chaoqi Zhou: Investigation, Formal analysis; Shichao Zhang: Project administration, Methodology; Kaihang Luo: Project administration, Methodology; Cheng Wang: Software, Resources; Zhiguo Hu: Supervision, Software, Resources; Yuting Yang: Visualization, Validation; Jia Zhou: Supervision, Methodology; Zhenguo Zeng*: Supervision, Validation, Funding acquisition.

Funding

This work was supported by National Natural Science Foundation of China (82060361, 82360373, 81460015, 82060360) and Jiangxi Provincial Key Laboratory of Critical Care Medicine (1210570003) and Natural Science Foundation of Jiangxi Province (20224ACB206020) and Key Research and Development Project of Jiangxi Provincial Natural Science Foundation (20202BBG73026) and Shandong Province Medical and Health Science and Technology Project (Project No. 202319010719).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.