INTRODUCTION
Intervertebral disc degeneration (IDD) is the main cause of lower back pain in adults, causing disability and a significant socioeconomic burden globally [1]. The intervertebral disc is an avascular organ with a gel-like nucleus pulposus (NP) at its core, surrounded by the annulus fibrosus and supported by cartilage endplates. Native NP cells maintain extracellular matrix homeostasis, support biomechanical function, and preserve the gelatinous nature of NP tissue [2]. In IDD, the key pathological changes are a decrease in the NP cell population and a reduction in the proliferative capacity of the remaining cells, leading to biomechanical alterations, discogenic pain, and spinal dysfunction [3,4]. Current treatments focus on pain relief due to the lack of effective pharmacological options. Understanding the mechanisms and molecular characteristics of IDD progression is crucial for identifying therapeutic targets [5].
Ferroptosis is a form of programmed cell death that is characterized by the accumulation of lipid peroxides and iron-dependent reactive oxygen species (ROS) within cells [6,7]. This process has been implicated in various pathological conditions, including cancer, neurodegeneration, and ischemia-reperfusion injury [8]. Ferroptosis is a critical pathological hallmark of IDD and is characterized primarily by the excessive accumulation of Fe2+ and lipid ROS [9,10]. Understanding the correlation between IDD and ferroptosis may reveal possibilities for potential therapeutic interventions. Recent studies have shown that lysosomal permeability can be a key regulator of ferroptosis [11,12]. Lysosomes contain a significant amount of redox-active iron as a result of the degradation of iron-containing materials [13]. When lysosomal membranes become more permeable due to stress or damage, they can release ROS and Fe2+ into the cytoplasm, which can then contribute to lipid peroxidation and further production of ROS [12]. Additionally, lysosomes contain enzymes such as cathepsins, which can directly promote ferroptotic cell death [12,14]. However, the mechanisms by which lysosomal membrane permeability (LMP) is altered under pathological conditions remain poorly understood.
In our recent study, we conducted a comprehensive investigation using transcriptomic, proteomic, metabolomic and lipidomic approaches to identify the molecular features and perturbed pathways associated with IDD [5,15]. The integration of multiomics data has revealed a crucial pathway regulated by the lysophosphatidylcholine acyltransferase 1-phosphatidylcholine (LPCAT1-PC) axis in the maintenance of the biomembrane system. The maintenance of lysosomal membrane integrity relies on PC as an essential lipid component, with LPCAT1 serving as the key enzyme in the regulation of PC synthesis [16,17]. These findings prompted us to further investigate the potential role of the LPCAT1-PC axis in the regulation of lysosomal biogenesis, LMP and lysosomal-dependent ferroptosis. Specifically, we sought to determine (1) whether LMP is an important factor involved in NP cell ferroptosis; (2) whether LMP is caused by downregulation of the LPCAT1-PC axis; and (3) whether the LPCAT1-PC axis has the potential to protect human NP cells from ferroptosis by facilitating lysosomal repair.
MATERIALS AND METHODS
1. Patient Samples
The present study was approved by the Ethics Committee of the First Affiliated Hospital of USTC (approval No. 2020-N(H)-049), and written informed consent was obtained from all participants. NP samples were collected from patients undergoing discectomy procedures. Prior to surgery, all patients underwent standardized magnetic resonance (MR) imaging of the lumbar spine. Disc degeneration was assessed using MR T2-weighted images based on the Pfirrmann classification system. Grade I–II disc degeneration was categorized as early-stage disc degeneration, whereas grade III–V disc degeneration was classified as advanced-stage disc degeneration [15]. NP tissues were obtained from patients who underwent surgical procedures at our institution. Afterward, the tissues were dissected into small fragments and subsequently subjected to enzymatic digestion with 0.2% type II collagenase (Gibco, USA) for 4 hours at 37°C. Following digestion, the resulting cell suspension was filtered through a sterile mesh and washed twice with phosphate-buffered saline (PBS). After centrifugation, the isolated cells were resuspended and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 100-U/mL penicillin, and 100-mg/mL streptomycin in a humidified incubator containing 5% CO2 at 37°C. The culture medium was changed every 3 days [5,15,18]. Subsequent experiments were conducted using second-passage NP cells.
2. Transcriptomic, Proteomic, Metabolomic, and Lipidomic Sequencing
All sequencing was conducted by Ouyi Biotech (Shanghai, China). Detailed methodologies can be found in our previously published work [5,15,18,19]. For RNA-seq, genes were considered significantly differentially expressed if they reached a difference threshold with a p-value <0.05 and a fold change >2 or <0.5. For proteomic sequencing, proteins were identified as upregulated and downregulated if they had a p-value <0.05 and a fold change >1.2 or <0.83, respectively. The criteria for identifying significantly differentially abundant metabolites were a variable importance projection (VIP) value >1 and a p-value <0.05.
Raw RNA-Seq data are available in the National Center for Biotechnology Information Sequence Read Archive under accession number PRJNA1100191. Proteomics data have been deposited to the ProteomeXchange Consortium via the iProX partner repository with dataset identifier PXD051351. In terms of multiomics analysis, the correlation between protein expression levels and their corresponding transcript levels was assessed, and the relationships among differentially expressed genes and proteins were visualized using a quadrant plot. Functional enrichment analyses, including Gene Ontology (GO) annotation, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and gene set enrichment analysis (GSEA), were carried out for both genes and proteins. Furthermore, the correlations among genes, proteins, and metabolites were investigated by mapping differentially expressed genes, differentially expressed proteins, and differentially abundant metabolites to the KEGG database [5].
3. Western Blotting
The proteins were extracted using RIPA buffer supplemented with the protease inhibitor PMSF (phenylmethylsulfonyl fluoride) (Beyotime, China, Cat No. ST507). The proteins were subsequently separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, after which they were transferred onto a polyvinylidene difluoride membrane. To minimize nonspecific binding, the membranes were blocked with 5% skim milk and then incubated overnight with a primary antibody. The primary antibodies used were as follows: anti-CTSD (Proteintech, China, Cat No. 66534-1-Ig), anti-COX2 (Proteintech, China, Cat No. 27308-1-AP), anti-TF (Proteintech, China, Cat No. 17435-1-AP), anti-LPCAT1 (Proteintech, China, Cat No. 16112-1-AP), anti-CHOP (Proteintech, China, Cat No. 66741-1-Ig), anti-GRP78 (Zenbio, China, Cat No. 340494), anti-LAMP1 (Zenbio, China, Cat No. 516894), anti-FTH1 (Huabio, China, Cat No. ET1610-78), and anti-GPX4 (Huabio, China, Cat No. ET1706-45). After being washed, the membranes were incubated with a horseradish peroxidase (HRP)-conjugated secondary antibody for 1 hour.
4. Measurement of Lipid ROS Levels
Lipid ROS levels were quantified using the lipid peroxidation sensor C11-BODIPY 581/591 (Thermo Fisher Scientific, USA, Cat No. D3861) following the manufacturer’s protocol. Briefly, approximately 2×104 NP cells were cultured in 15-mm coronal dishes until they reached 70% to 80% confluency and then incubated with 10 μM C11-BODIPY 581/591 at 37°C for 30 minutes [20]. The cells were subsequently gently washed with PBS. The fluorescence was observed using a confocal microscope. The mean fluorescence intensity of the images was analyzed using ImageJ software.
5. Subcellular Fractionation
A total of 1×106 NP cells were collected for subcellular fractionation and the preparation of cytosolic and lysosomal-enriched fractions utilizing a Lysosome Enrichment Kit (Solarbio, China, Cat No. EX2670). Briefly, the cell pellet was resuspended in PBS and subjected to homogenization using a dounce homogenizer on ice with approximately 20 strokes. Afterward, the homogenate was subjected to differential centrifugation following the manufacturer’s protocol to isolate the crude lysosomal fractions, while the resulting supernatant fractions were collected as cytosolic fractions [21,22]. The isolated subcellular fractions were stored at -80°C until further analysis.
6. Lysosomal Enzyme Assay
The activities of cathepsin D (CTSD; Proteintech, China, Cat No. KE00421) and alpha-N-acetylglucosaminidase (NAGLU; Sigma-Aldrich, USA, Cat No. CS0780) were determined following the protocols provided by the manufacturers. The enzyme activities in the lysosomal and cytosolic fractions were estimated on the basis of the relative fold change in optical density [21].
7. LysoTracker Staining
Lysosomes were evaluated through LysoTracker Red staining (Beyotime, China, Cat No. C1046), in accordance with the manufacturer’s guidelines. The cells were incubated with LysoTracker Red at a concentration of 50 nM at 37°C for 30 minutes [23]. Afterward, the cells were imaged by confocal microscopy.
8. FerroOrange Staining
The intracellular Fe2+ iron levels were assessed using FerroOrange (CST, USA, Cat No. 36104). NP cells were seeded in a coronal dish and subjected to the indicated treatment. Serum-free medium containing 1 mM FerroOrange was added to the NP cells, which were then incubated at 37°C for 30 minutes [20].
9. Transmission Electron Microscopy
Human NP cells were fixed overnight in a solution of 2.5% glutaraldehyde and then immersed in 1% osmium tetroxide for 1 hour at 4°C. Afterward, the cells underwent a stepwise dehydration process using increasing concentrations of ethanol starting at 30%, followed by 50%, 70%, and finally 90%, with each step lasting 30 minutes. Afterward, the cells were immersed 3 times in pure ethanol for 30 minutes each. The dehydrated samples were then infiltrated with an embedding medium mixed with propylene oxide. Ultrathin sections with a thickness of approximately 70 nm were prepared using an LKB-V ultramicrotome (LKB, Sweden) and mounted on copper grids. These sections were then stained with lead citrate for 10 minutes at room temperature and rinsed 3 times with deionized distilled water. Further staining was carried out using uranyl acetate for 30 minutes under room temperature conditions, followed by another 3 washes with deionized distilled water [18]. Finally, the samples were examined using a JEM-1400 transmission electron microscope (JEOL, Japan) for imaging.
10. PC Content Analysis
The PC content was quantified using a colorimetric/fluorometric assay kit (BioVision, USA, Cat No. K576-100) according to the manufacturer’s instructions. Samples, reagents, and buffer were added to 96-well plates and incubated in the dark for 3 hours, after which the absorbance at 570 nm was measured with a microplate reader [5]. The results are expressed as ratios relative to those of control or untreated cells.
11. Cell Transfection
To overexpress LPCAT1, full-length human LPCAT1 cDNA was cloned and inserted into the pCDH-CMV-MCSEF1-puro vector. In addition, the target sequence of the LPCAT1 shRNA-1 plasmid was 5´-tacccggatcagacacatttc-3´, that of the LPCAT1 shRNA-2 plasmid was 5´-acggaaagtggccacagataa-3´, and that of the LPCAT1 shRNA-3 plasmid was 5´-agataggtattgcggagtttg-3´. Briefly, NP cells were seeded in 35 mm dishes overnight, after which the plasmids were transfected using Lipofectamine 2000 (Invitrogen, USA, Cat No. 12566014) and Opti-MEM (Invitrogen, USA, Cat No. 31985062) according to the manufacturer’s instructions. The transfection efficacy was measured by Western blotting (WB).
12. Histology and Immunostaining Assays
Human NP tissues were fixed in 4% paraformaldehyde for 24 hours, embedded in paraffin, sectioned into 5-μm slices, and stained with HE, Masson and Alcian blue. Rat intervertebral discs were fixed, decalcified over 4 weeks, dehydrated, paraffin-embedded, sectioned into 5 μm slices, and stained with hematoxylin-eosin (HE), Safranin‑O (S-O), and Alcian blue. Histological scoring was conducted using an established system for rat intervertebral discs and was based on the following 5 criteria: (1) cellularity of the annulus fibrosus; (2) morphology of the annulus fibrosus; (3) boundary between the annulus fibrosus and NP; (4) cellularity of the NP; and (5) morphology of the NP. In this scoring system, the grading scale ranged from 5 to 15, with a normal disc assigned 1 point per category, while a disc exhibiting severe degeneration received 3 points per category based on the evaluation of HE, Masson and Alcian blue staining [24,25].
For immunohistochemistry (IHC) analysis, tissue sections were treated with 3% hydrogen peroxide to inhibit endogenous peroxidase activity, followed by antigen retrieval in 0.125% trypsin at 37°C. The sections were then washed with PBS, blocked with 10% goat serum, and incubated with primary antibodies overnight at 4°C. The sections were subsequently incubated overnight at 4°C with primary antibodies targeting LPCAT1 (Proteintech, China, Cat No. 16112-1-AP), COX2 (Proteintech, China, Cat No. 27308-1-AP), TF (Proteintech, China, Cat No. 17435-1-AP), FTH1 (Huabio, China, Cat No. ET1610-78), GPX4 (Huabio, China, Cat No. ET1706-45), 4-HNE (Abcam, China, Cat No. ab48506), anti-CHOP (Proteintech, China, Cat No. 66741-1-Ig), and anti-GRP78 (Zenbio, China, Cat No. 340494), followed by incubation with HRP-linked secondary antibodies for 1 hour at room temperature. Hematoxylin was used as a counterstain for visualization.
For immunofluorescence (IF) staining, cells were plated at an appropriate density into 15-mm glass-bottom culture dishes. Once the desired confluency was achieved, the cells were fixed with 10% buffered formalin and permeabilized. Primary antibodies anti-CHOP (Proteintech, China, Cat No. 66741-1-Ig), anti-GRP78 (Zenbio, China, Cat No. 340494), LAMP1 (Zenbio, China, Cat No. 516894), CTSD (Proteintech, China, Cat No. 66534-1-Ig), and anti-GFP (Proteintech, China, Cat No. pabg1) were applied, followed by overnight incubation at 4°C. Detection was performed using secondary antibodies conjugated with Alexa Fluor 488 or Alexa Fluor 594.
13. Animal Experiments
The animal experiments were reviewed and approved by the Animal Care and Use Committee of the First Affiliated Hospital of USTC (approval No. 2022-N(A)-116). Male Sprague-Dawley rats, aged 3 months and weighing between 250 and 300 g, were utilized to create the IDD model [5,18,26]. The rats were housed under standard conditions with a 12/12-hour light/dark cycle and provided ad libitum access to food and water. After a 1-week acclimatization period, the animals were randomly assigned to 7 groups using a computer-generated random number sequence: a normal control group (NC, n=6), a puncture-induced IDD group (IDD, n=6), an N-dodecylimidazole (NDI)-induced IDD group (NDI, n=6), an IDD group treated with LPCAT1 adeno-associated virus (AAV) (IDD+LPCAT1, n=6), an NDI group treated with LPCAT1 AAV (NDI+LPCAT1, n=6), an IDD group treated with PC (IDD+PC, n=6), and an NDI group treated with PC (NDI+PC, n=6).
The rats were anesthetized via inhalation of isoflurane (induction at 4% and maintenance at 2%–2.5% in oxygen). The tail was disinfected with povidone-iodine. Under fluoroscopic guidance, the Co8/9 disc level was identified. For puncture-induced IDD, a 20-gauge needle was inserted percutaneously into the center of the disc, advanced to a depth of 5 mm to penetrate the NP, rotated 360°, and maintained in position for 30 seconds to ensure adequate injury to the annulus fibrosus and disruption of the NP tissue. The needle was then carefully withdrawn, and the puncture site was redisinfected. For the NDI-induced IDD model, a 33-gauge Hamilton needle (Hamilton, Switzerland) was inserted into the disc center under fluoroscopic guidance, and 3 μL of 1.0% (w/v) NDI solution was slowly injected. The needle remained in place for 30 seconds after injection to minimize leakage and promote distribution within the disc space.
Postoperatively, the rats received a single dose of analgesic immediately after surgery and again 12 hours later. One week after models construction, LPCAT1-overexpressing AAV (constructed by Shanghai Genomeditech Company, China) or PC (Solarbio, China, Cat#: L8260) was injected intradiscally. A volume of 3 μL containing 1.2×1010 vector genomes of LPCAT1 AAV or PC (2 mg/mL) was slowly administered using a 33-gauge needle. The needle remained in place for 30 seconds after injection to minimize leakage and promote distribution within the disc space. This intradiscal injection was repeated weekly for 3 consecutive weeks. The following animals were excluded from the final analysis: (1) death due to anesthetic complications or (2) the development of surgical site infection.
14. Radiological Evolution
Five weeks after the final transfection, all the rats underwent MR examination. Scanning was performed on a GE Discovery MR 750W 3.0T scanner. A dedicated rodent coil (MS160-3T; Medcoil, China) was used, with the rat positioned prone to center the tail within the isocenter. High-resolution T2-weighted fast spin-echo sequences were acquired in the sagittal plane with the following typical parameters: TR (time to repetition)/TE (time to echo)=2,500/160 msec, a slice thickness of 0.8 mm, and an interslice spacing of 0.5 mm, with a field of view measuring 320×320 mm.
The extent of IDD on the MR images was assessed according to the Pfirrmann scale, in which grade I demonstrated a homogeneous, hyperintense signal with clear structure and normal height. Grade II is inhomogeneous yet hyperintense with clear structure and normal height. Grade III shows an intermediate signal intensity, becoming inhomogeneous with unclear structure and potentially slightly decreased height. Grade IV is hypointense and inhomogeneous, with the loss of structural distinction and moderately decreased height. Grade V is hypointense with a collapsed disc space and no structural distinction [19].
15. Statistical Analysis
The data are presented as the mean±standard deviation from at least 3 independent experiments. Statistical differences between 2 groups were determined using Student’s t-test or the Mann-Whitney U-test for normally or nonnormally distributed data, respectively. For multigroup comparisons, one-way analysis of variance followed by Tukey post hoc test for multiple comparisons or the Kruskal-Wallis test followed by Dunn post hoc test with Bonferroni adjustment was applied, depending on the data distribution. The relationships between 2 quantitative variables were assessed using Spearman rank correlation test. Statistical significance was set at p<0.05.
RESULTS
1. Ferroptosis Was Detected During IDD
The degree of IDD was assessed based on T2-weighted MR images, and the degree of histological degeneration in the NP tissues was evaluated (Fig. 1A and B). During the course of IDD, the expression of markers of ferroptosis, namely lipid ROS and FerroOrange, significantly increased (Fig. 1C–F). We subsequently quantified the protein levels of COX2, TF, GPX4, and FTH1. The results demonstrated that COX2 and TF levels increased in the advanced stage, whereas GPX4 and FTH1 levels decreased (Fig. 1G and H). Furthermore, IHC staining for COX2, TF, and GPX4 corroborated the heightened ferroptosis activity in the advanced stage (Fig. 1I and J). These findings suggest that ferroptosis is involved in IDD.
2. Lysosomal Dysfunction Is the Key Molecular Feature of IDD
In our previous study, we conducted integrative transcriptomic and proteomic sequencing to identify the molecular features of IDD, and the study workflow is illustrated in Fig. 2A (created by Figdraw). GSEA of both the transcriptomic and proteomic data revealed a significant association between lysosomal dysfunction and IDD (Fig. 2B and C). Transmission electron microscopy (TEM) revealed impaired structural integrity of the lysosomes in the degenerated NP cells (Fig. 2D). Compared with early-stage IDD, advanced-stage IDD resulted in higher concentrations of lysosomal enzymes in the cytosolic fractions, accompanied by a reduction in the enzyme content within the lysosomal fractions (Fig. 2E–J). IF staining of CTSD and LAMP1 revealed that CTSD exhibited a more diffuse pattern and resulted in fewer puncta in the advanced stage, indicating impaired lysosomal biogenesis and LMP in IDD (Fig. 2K).
3. LMP Induced Ferroptosis in NP Cells In Vivo and In Vitro
To further investigate the correlation between lysosomal dysfunction and ferroptosis, we used NDI to induce disruption of the lysosomal membrane [22]. NDI induced the leakage of lysosomal contents from the lysosome into the cytosol (Fig. 3A–G) in NP cells. Additionally, ferroptotic levels increased after NDI treatment (Fig. 3H–M). Furthermore, in vivo experiments revealed that NDI resulted in IDD and ferroptosis in a mouse model (Fig. 3N–S). These findings collectively indicate that elevated LMP promotes ferroptosis in IDD.
4. Altered Lysosomal Membrane Lipid Composition Was Detected in IDD
To elucidate the mechanism underlying LMP, we utilized a metabolomic approach to investigate alterations in the metabolic profiles of entire NP cells. The analysis revealed a total of 175 metabolites with a VIP value >1 and p-value <0.05 between the 2 groups (Fig. 4A–C). The KEGG pathway mapper revealed remarkable alterations in the “glycerophospholipid metabolism (Has00564)” pathway, which is involved in lysosomal membrane biogenesis (Fig. 4D). We then prepared lysosome-enriched fractions from early- and advanced-stage cells to analyze the lipid composition of lysosomal membranes. As shown in Fig. 4E–G, our metabolic profiling results revealed a significant difference between the early and advanced stages, and the analysis revealed a total of 65 differential metabolites between the 2 groups. KEGG pathway mapper was used to identify the most strongly impacted pathway (Fig. 4H). An analysis integrating the data from the sequencing of whole NP cells and lysosome-enriched fractions revealed the “glycerophospholipid metabolism (Has00564)” pathway and downregulated PC abundance in whole NP cells and lysosomes (Fig. 4I–K).
Through the integration of transcriptomics, proteomics, and metabolomics for whole NP cells and lipidomics for lysosomes, a crucial pathway regulated by the LPCAT1-PC axis was revealed. This pathway is classified under “glycerophospholipid metabolism (Has00564)” (Fig. 4L). In this pathway, LPCAT1 levels significantly decreased in the advanced stage (Fig. 4M and N). In the validation cohort, we collected 29 NP tissue samples from the early-stage and 36 from those in the advanced-stage (Supplementary Table 1). IHC staining revealed that the proportion of LPCAT1-positive cells was significantly lower in the advanced stage than in the early stage (Fig. 4O and P). In addition, the expression of LPCAT1 was negatively correlated with disc degeneration grade (r=-0.897, p<0.001) (Fig. 4Q). Overall, these findings suggested that decreased LPCAT1 expression may be involved in lysosomal membrane composition changes, mediated LMP and exacerbated ferroptosis.
5. Downregulation of the LPCAT1-PC Axis Leads to LMP and Ferroptosis in NP Cells
To further elucidate the functional mechanisms of LPCAT1 in lysosomal dysfunction, LMP and ferroptosis, NP cells were transfected with the short hairpin-LPCAT1 (SH-LPCAT1) plasmid, in which SH-2 attained high transfection efficiency and was chosen for subsequent experiments (Fig. 5A and B). CTSD activity and NGAL activity increased in the cytosolic fraction but decreased in the lysosomal fraction after transfection (Fig. 5C and D). The content of lysosomal enzymes in the cytosolic fractions increased after SH-LPCAT1 transfection, which was accompanied by a decrease in the enzyme content in the lysosomal fractions (Fig. 5E–H). IF staining of CTSD and LAMP1 revealed that CTSD exhibited a more diffuse pattern and resulted in fewer puncta upon LPCAT1 inhibition (Fig. 5I).
Next, we explored whether LPCAT1 inhibition leads to NP cell ferroptosis. Markers of ferroptosis, namely, lipid ROS and FerroOrange, were significantly elevated after transfection with the SH-LPCAT1 plasmid (Fig. 5J–M). WB analysis revealed that LPCAT1 inhibition resulted in the upregulation of COX 2 and TF expression but the downregulation of GPX4 and FTH1 expression (Fig. 5N and O). TEM revealed severe disruption of the mitochondrial morphology following sh-LPCAT transfection, characterized by disintegration of the cristae structure and vacuolization, indicating ferroptosis (Fig. 5P). In contrast, the transfection of LPCAT1 in advanced-stage cells significantly alleviated the leakage of lysosomal enzymes from lysosomes into the cytosol (Supplementary Fig. 1A–D), accompanied by a reduction in ferroptosis (Supplementary Fig. 1E–J). Taken together, these data suggest that the downregulation of the LPCAT1-PC axis may contribute to LMP and the induction of ferroptosis in NP cells.
6. The LPCAT1–PC Axis Normalizes the Lysosomal Membrane Lipid Composition and Attenuates LMP
To determine whether the activation of the LPCAT1–PC axis can mitigate lysosomal membrane damage and LMP, we treated NDI-induced NP cells with either an LPCAT1 plasmid or PC. NDI induces the leakage of lysosomal enzymes into the cytoplasm, whereas LPCAT1 or PC prevents NDI-induced damage to lysosomes (Fig. 6A–F). IF costaining of CTSD and LAMP1 revealed that CTSD exhibited a less diffuse pattern and had more puncta within lysosomes cocultured with NDI and LPCAT1 or PC (Fig. 6G).
We subsequently prepared lysosome-enriched fractions to analyze the lipid composition of lysosomal membranes. Importantly, significant differences were detected between the lysosomal lipid profiles of the NDI- and LPCAT1-treated NDI groups (Fig. 6H and I). As shown in Fig. 6J, our analysis revealed significant differences in the abundance of phospholipids between NDI and NDI+LPCAT1. Enrichment analysis also revealed distinct components involved in the “glycerophospholipid metabolism” pathway (Has00564) (Fig. 6K). Furthermore, several subset classes of PCs significantly increased after LPCAT1 treatment (Fig. 6L). Lipidomic analysis revealed a similar trend for the NDI group versus the PC+NDI group (Fig. 6M–Q). These findings collectively indicate that the LPCAT1-PC axis mitigates lysosomal damage and LMP while promoting the formation of lysosomal membranes.
7. LPCAT1-PC Axis Facilitates the Interaction Between the ER and Lysosomes
We conducted a comprehensive analysis to further investigate the underlying mechanism of action associated with the LPCAT1-PC axis in facilitating lysosomal repair. PCA and volcano heat mat analysis revealed greater variations in protein expression after LPCAT1 treatment (Fig. 7A and B). Next, we investigated the potential biological functions of these proteins. KEGG and GO analyses revealed distinct components involved in the cellular processes “lysosome,” “lysosome lumen,” and “endoplasmic reticulum (ER) lumen” (Fig. 7C and D). GSEA revealed significant items, including “lysosome,” “lysosomal lumen,” and “lysosomal membrane” (Fig. 7E). The proteomic analysis for PC treatment yielded similar results (Fig. 7F–J). These findings indicate that the LPCAT1-PC axis enhances the interaction between the ER and lysosomes.
Lysosomal repair involves the formation of membrane contact sites between damaged lysosomes and the ER to enable efficient, rapid lipid transport for repair (Fig. 7K). As expected, treatment with NDI resulted in impaired ER-lysosome contact, whereas the LPCAT1-PC axis provided protection against this effect and increased lysosomal PC abundance (Fig. 7L–N). Additionally, the LPCAT1-PC axis promotes the contact between lysosomes and the ER in advanced-stage IDD (Fig. 7O). Collectively, these findings suggest that the LPCAT1-PC axis enhances the interaction between the ER and lysosomes.
8. The LPCAT1-PC Axis Ameliorates Ferroptosis Through the ER Pathway
We subsequently investigated whether the LPCAT1-PC axis can inhibit LMP and ferroptosis through the ER pathway. After being transfected with LPCAT1 or PC, the NP cells were treated with tunicamycin (TM), which disrupts protein glycosylation to induce ER stress. The level of ER stress significantly increased after TM disturbance (Fig. 8A–D). The interaction between the ER and lysosomes was impaired by TM (Fig. 8E). LPCAT1-PC significantly attenuated LMP, whereas this effect was blocked by treatment with TM (Fig. 8F–J). In addition, the LPCAT1-PC axis resulted in a reduction in ferroptosis levels. However, the presence of TM led to an increase in the expression of these markers associated with ferroptosis to some extent (Fig. 8K–P). Taken together, these results indicate that the LPCAT1-PC axis protects human NP cells from ferroptosis by facilitating lysosomal repair through interactions with the ER.
9. Therapeutic Efficacy of the LPCAT1-PC Axis in an IDD Model
To examine the therapeutic effects of the LPCAT1-PC axis in vivo, we used a rat model of IDD in our study, and a schematic illustration is shown in Fig. 9A. GFP (green fluorescent protein) signals were clearly visible in the NP tissue 5 weeks after the final transfection, indicating successful transfection of LPCAT1 AAV (Supplementary Fig. 2A–C) and further promotion of LPCAT1 protein expression (Supplementary Fig. 2D and E). MR images revealed that treatment with LPCAT1 or PC significantly improved the Pfirrmann scores in the IDD models (Fig. 9B and C). Additionally, LPCAT1 or PC mitigated IDD progression, as determined by histological evaluation using HE, S-O and Alcian blue staining (Fig. 9D and E). Moreover, treatment with LPCAT1 or PC decreased the expression of GRP78 and CHOP in the IDD group (Supplementary Fig. 2F and G) and further preserved lysosomal formation (Fig. 9F and G). Ferroptosis was evaluated using Bodipy-lipids, FerrOrange, and IHC staining for COX2, TF, GPX4, and FTH1, which revealed that the IDD models presented a high degree of ferroptosis, with these results partially reversed following treatment with LPCAT1 or PC (Fig. 9H–M).
Taken together, these results indicate that the LPCAT1-PC axis protects human NP cells from ferroptosis by facilitating lysosomal repair through interactions with the ER (Fig. 10, created by Figdraw).
DISCUSSION
By employing an integrated transcriptomic and proteomic analysis strategy to elucidate the molecular features of IDD, we investigated lysosomal dysfunction in NP cells. Specifically, in advanced-stage IDD, the normal structure of the lysosomes in NP cells was disrupted with increased LMP, and the contents of the lysosomes leaked into the cytoplasm. Lysosomes play a vital role in maintaining cellular homeostasis through their involvement in diverse functions, including the recycling of damaged organelles, degradation of macromolecules, and participation in intracellular signaling and plasma membrane repair [27]. The leakage of lysosomes, as well as LMP, not only leads to the loss of lysosomal functionality but also results in the release of Fe2+, ROS and cathepsins that directly damage intracellular components [12]. Previous studies have revealed that LMP is an important pathological mechanism in many diseases, including traumatic brain injury, skin flap ischemia and traumatic spinal cord injury [21,22]. In the current study, a higher degree of LMP was noted in advanced-stage IDD. Moreover, the pharmacological interference of lysosomes resulted in increased leakage of lysosomal contents and cathepsins into the cytoplasm, both in vivo and in vitro, indicating the induction of a ferroptotic phenotype. Here, we hypothesized that LMP, which is involved in the pathological mechanisms of IDD, might be a promising therapeutic target for treatment.
The alteration of the lipid composition in the lysosomal membrane plays a crucial role in determining the permeability of lysosomes. The primary constituents of lysosomal membrane lipids include PC, cholesterol, phosphatidylethanolamine, and sphingomyelin [28]. The unique lipid composition of the lysosomal membrane is essential for maintaining its structural integrity and regulating the entry and exit of various molecules. Changes in lipid composition can significantly impact lysosomal function by affecting membrane fluidity and stability, particularly alterations in specific lipids such as cholesterol or PC, which may compromise lysosomal permeability [22,29]. This disruption can lead to impaired degradation processes within cells, the accumulation of undigested material, and the leakage of lysosomal contents into the cytosol, ultimately contributing to lysosome-dependent cell death. Sarkar et al. [21] demonstrated that PLA2G4A hydrolyzes fatty acyl linkages of glycerophospholipids, leading to an increase in lysophospholipid membrane abundance and a decrease in PC abundance. Similarly, Lou et al. [22] demonstrated that LMP was the primary cause of lysosomal dysfunction and necroptosis in ischemic flaps, with changes in the lipid composition of the lysosomal membrane mediated by PLA2G4E (phospholipase A2 group IVE) being responsible for LMP occurrence. Therefore, understanding how alterations in lipid composition affect lysosomal function is not only important for elucidating basic biological processes but also has significant implications for therapeutic interventions.
Accordingly, we isolated lysosome-enriched fractions from both early- and advanced-stage IDD samples to investigate alterations in the lipid composition of the lysosomal membrane. Lipidomic analysis revealed a substantial reduction in total lysosomal PC content in advanced-stage samples, with the involvement of the glycerophospholipid metabolism pathway. PC plays a crucial role as one of the major components of lysosomal membranes, accounting for approximately 30% of their total membrane lipid composition [28]. The synthesis of PC is essential for maintaining and forming these membranes during lysosomal biogenesis, which is an important cellular process involved in recycling damaged or unnecessary cellular components [17]. The observed decrease in PC levels within lysosomal membranes raises concerns about potential detrimental effects on their structural integrity and function. A compromised lipid composition could lead to disruptions such as LMP and leakage of lysosomal contents [21]. Notably, various mechanisms regulate the phospholipid composition within lysosomal membranes. One such mechanism involves ER-lysosome lipid transport, in which lipids are transferred between the ER and lysosomes to maintain proper membrane composition [30,31]. Additionally, pathological degradation by phospholipases can also impact these components. Investigating the interactions and influences of diverse molecules on lysosomal phospholipid compositions may provide insights into potential therapeutic targets.
In our previous study, we reported dysfunctions in various organelles, including mitochondria, the ER, and lysosomes, in NP cells with advanced-stage IDD. Through the integration of multiomics analyses and functional validation, we identified a pivotal pathway regulated by the LPCAT1-PC axis. LPCAT1 is a vital enzyme that plays a crucial role in the lipid synthesis pathway known as the Lands cycle [32]. Specifically, LPCAT1 catalyzes the acylation reaction between lysophosphatidylcholine and fatty acyl-CoA to produce PC [5]. By actively participating in this enzymatic process, LPCAT1 ensures an adequate supply of PC to meet cellular physiological demands. Additionally, LPCAT1 contributes to the repair of peroxidation chain damage in cell biomembranes through the replacement of oxidized acyl chains to maintain membrane integrity [16]. In a study conducted by Bi et al. [33], LPCAT1 modulated plasma membrane composition by increasing the level of saturated PC, which is essential for the efficient transduction of oncogenic signals. Upregulation of LPCAT1 leads to an increased proportion of saturated fatty acids incorporated into cell membranes; this results in increased levels of phospholipid saturation and decreased levels of polyunsaturated fatty acids. Consequently, this mechanism protects cells from membrane damage caused by phospholipid peroxidation and ultimately inhibits ferroptosis in various tumor cells [16]. The findings of our study suggest that upregulation of the LPCAT1-PC axis can effectively alleviate LMP and ferroptosis in NP cells. However, the detailed mechanisms by which the LPCAT1-PC axis regulates lysosomal repair require further elucidation.
After the proteomic data following LPCAT1 or PC treatment were compared, a series of proteins associated with various biological functions were identified. KEGG enrichment analysis revealed significant associations with “endoplasmic reticulum,” “lysosomal lumen,” and “lysosomes.” Additionally, different GO terms were related to “lysosomal lumen (GO:0043202),” “lysosome (GO:0005764),” and “lysosomal membrane (GO:0005765).” Furthermore, fluorescence tracker staining revealed increased interactions between the ER and lysosomes in advanced-stage NP cells treated with LPCAT1 or PC, whereby the ER envelops the lysosomes. These results suggest that the LPCAT1-PC axis facilitates the interaction between the ER and lysosomes. LPCAT1 is an ER-localized protein that promotes the synthesis of PC. The interaction between the ER and lysosomes serves as the structural basis for lipid transport, enabling the transfer of lipids from the ER to lysosomes [31]. This process not only ensures a sufficient supply of phospholipids for lysosomal membranes but also facilitates lipid remodeling within these compartments. Consequently, dynamic changes in lipid composition can impact various aspects of lysosomal function, such as membrane fluidity and permeability. These findings suggest that the LPCAT1–PC axis may enhance the interaction between the ER and lysosomes, thereby promoting the repair of lysosomal membranes and alleviating LMP and ferroptosis.
When lysosomal membranes are compromised, cells initiate a complex response to restore their integrity. Damaged lysosomes can be repaired through ESCRT (endosomal sorting complex required for transport) complexes or degraded via autophagy; however, prompt lysosomal repair can still occur in cells lacking crucial ESCRT or autophagy genes, suggesting the existence of alternative pathways for lysosomal repair and quality control within the cellular system [34-36]. Recent studies have identified a PITT (phosphoinositide-initiated membrane tethering and lipid transport) pathway that is activated upon LMP [31]. This pathway involves the accumulation of PI4K2A (phosphatidylinositol 4 kinase type 2α) on damaged lysosomes, facilitating direct lipid transfer from the ER to lysosomes and supporting membrane repair. Lysosomal repair by the LPCAT1-PC axis was blocked during ER stress, highlighting the importance of contact between the ER and lysosomes in lysosomal repair. These contacts, often referred to as membrane contact sites, allow for the direct transfer of lipids and other molecules between the 2 organelles, thereby enhancing cellular lipid homeostasis and signaling pathways [37]. Moreover, these contacts are dynamic structures that can respond to cellular needs [38]. Here, our research offers new perspectives on the mechanistic function of the LPCAT1-PC axis in lysosomal repair processes in NP cells and elucidates the therapeutic potential of the LPCAT1-PC axis in mitigating ferroptosis.
This study has several limitations. First, human IDD is a chronic condition that progresses gradually over time, whereas the IDD model represents a traumatic, acute model that does not fully mimic the human IDD process. IDD is a multifactor and complex process, and generating an in vitro IDD model induced by a single factor that accurately encompasses the etiology and pathogenesis of IDD is challenging. Although the in vitro model cannot fully simulate the etiology and pathological mechanism of IDD, we can choose a model reasonably according to the phenotype or pathogenic pathway [39,40]. In this study, we observed significant upregulation of ferroptosis and lysosomal damage markers in a puncture-induced IDD model after 8 weeks. This method may provide a potentially viable model for investigating the efficacy of potential therapeutic targets. Second, in the investigation of ferroptosis, we focused on the function of lysosomes and did not consider alternative signaling pathways. These avenues of research will be explored in subsequent studies. Third, since IDD was considered the origin of pain, we did not validate the IDD models in combination with assessments of thermal hyperalgesia and mechanical allodynia. Further research is needed to address this limitation.
CONCLUSION
Our findings indicate that LPCAT1 and its downstream metabolite, PC, function to enhance the interactions between the ER and lysosomes. This strengthened interaction facilitates the remodeling and restoration of phospholipids within the lysosomal membrane, thereby alleviating LMP and ultimately promoting the evasion of ferroptosis. Pharmacological modulation of the LPCAT1-PC axis, both in vivo and in vitro, has shown considerable potential as a therapeutic strategy for the treatment of IDD.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Supplement
Print
