DDRGK1 preserves intervertebral disc development through ufmylation Authors: Mingkuan Lu, Tangjun Zhou, Xiao Yang, Kexin Liu, Kewei Rong, Peixiang Ma, An Qin, Jie Zhao Published: 2025 ABSTRACT UFMylation, similar to ubiquitination, is a unique post-translational modification which is indispensable in hematopoiesis, neurogenesis and chondrogenesis. However, its role in intervertebral disc development remains unclear. In this study, we focused on DDRGK domain containing protein 1 (DDRGK1), a pivotal component involved in UFMylation, and generated Ddrgk1fl/fl; Acan-CreERT2 (Ddrgk1cKO) mice to explore DDRGK1’s regulatory function in the nucleus pulposus and cartilage endplate. We found that Ddrgk1 conditional knockout led to severe retardation of spinal growth, disruption of disc cellularity and initiation of disc degeneration during early postnatal phase. Furthermore, Ddrgk1 conditional knockout in late postnatal phase resulted in profound degeneration of mouse discs, mainly characterized by substantially reduced thickness of cartilage endplate. In addition, Ddrgk1cKO mice exhibited exacerbated disc degeneration compared to the WT mice after the lumbar spine instability surgery. RNA sequencing of disc cells from Ddrgk1cKO mice showed upregulation of genes related to apoptosis, matrix metalloproteinase activation, extracellular matrix (ECM) degradation and endoplasmic reticulum (ER) unfolded protein response after Ddrgk1 conditional knockout. Immunohistochemical analysis further verified increased apoptosis, ECM disruption and ER stress in both nucleus pulposus and cartilage endplate after Ddrgk1 conditional knockout. In summary, this study demonstrated that DDRGK1 preserves the normal cellularity and structure of intervertebral discs by regulating the cell fate of nucleus pulposus and cartilage endplate cells, maintaining the ER homeostasis and regulating the metabolic balance of ECM.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00018-025-06034-8. Introduction The intervertebral disc is a fibrocartilaginous tissue located between adjacent vertebrae [1]. It permits limited movement and distributes compressive loads across adjacent vertebrae [2]. The intervertebral disc comprises three key components: nucleus pulposus, annulus fibrosus and cartilage endplate [3, 4]. The nucleus pulposus is a highly hydrated and gelatinous tissue composed of nucleus pulposus cells and abundant ECM [5]. The ECM secreted by nucleus pulposus cells is crucial for withstanding compressive loads and maintaining a stable cellular microenvironment [6]. The annulus fibrosus surrounding the nucleus pulposus constrains the position and shape of nucleus pulposus during axial compression and distributes stress uniformly across the vertebral body [7]. The cartilage endplate, a hyaline cartilage layer approximately 0.6–1 mm thick, covers the superior and inferior surfaces of the nucleus pulposus [8, 9]. Chondrocytes are the only cell type in the cartilage endplate which secrete ECM, mainly composed of collagen Ⅱ and proteoglycans, to maintain the structure and function of the cartilage endplate [10–12]. The cartilage endplate connects the vertebral body and nucleus pulposus, serving as a mechanical barrier and molecular channel from nearby blood vessels to the avascular intervertebral disc [13–15]. Intervertebral disc degeneration is a common degenerative disease of the musculoskeletal system [7, 16]. As a multifactorial disease, the pathogenesis of intervertebral disc degeneration remains incompletely understood. Notably, balanced interactions between the ECM and disc cells play a critical role in the homeostasis of intervertebral disc [17, 18]. Thus ECM degradation has been considered as an initiating and pivotal event in the intervertebral disc degeneration [19, 20]. In addition, the apoptosis of nucleus pulposus and cartilage endplate cells is closely associated with the progression of intervertebral disc degeneration [14, 21–24]. Cartilage endplate calcification has also been highlighted as a key contributor which alters mechanical properties and impairs nutrient supply, ultimately exacerbating intervertebral disc degeneration [25, 26]. Furthermore, ER stress, a condition occurs when a large number of misfolded proteins accumulate in the ER, could also disrupt the metabolic balance of the ECM, trigger the apoptosis of the nucleus pulposus cells and ultimately lead to the aggravation of intervertebral disc degeneration [27–29]. Ubiquitin folding modifier 1 (UFM1) conjugation system is a recently characterized ubiquitin-like modification system. This modification, known as UFMylation, is catalyzed by the sequential action of the ubiquitin activating enzyme 5 (UBA5), the UFM1 conjugase 1 (UFC1) and the UFM1 ligase 1 (UFL1), and plays a critical role in maintaining ER homeostasis and embryonic development [30–32]. DDRGK1 is an essential component of UFMylation which forms a stable complex with UFL1 and contributes to the covalent attachment of UFM1 to substrate proteins [30, 33]. Embryonic knockout of Ddrgk1 has been confirmed to cause hematopoietic defects, leading to embryonic death by E12.5 due to anemia [34, 35]. Furthermore, DDRGK1 depletion reduces p53 stability, prevents p53 accumulation under DNA damage conditions and promotes tumor growth, underscoring the critical role of DDRGK1 in maintaining genomic integrity and determining cell fate [36]. Notably, DDRGK1 has been shown to be indispensable in cartilage development. Loss-of-function mutations in DDRGK1 have been proven to be linked to severe chondrodysplasia through its regulation of SOX9, a key transcription factor in chondrogenesis [34, 37]. Conditional knockout of Ddrgk1 in mouse mesenchymal cells led to severe limb shortening, joint abnormalities and growth plate development disorders characterized by shortened proliferative region and enlarged hypertrophic region [37]. In addition, DDRGK1 may modulate ER homeostasis and chondrogenesis by regulating the stability of inositol requiring enzyme 1α (IRE1α) via UFMylation [38, 39]. However, no study has verified whether DDRGK1 regulates the survival and development of nucleus pulposus and cartilage endplate cells through UFMylation. To address this, we employed Ddrgk1 conditional knockout mice to investigate the underlying molecular mechanisms. Materials and methods Materials All the key reagents and instrumentation involved in this study were listed in Table 1 with their source, region and identifier.Table 1Key reagents or resources involved in this studyREAGENT or RESOURCESSOURCECOUNTRY/REGIONIDENTIFIERChemicals2 × Phanta UniFi Master MixVazymeNanjing, Jiang Su, ChinaCat# P5264-HydroxytamoxifenMedChemExpressShanghai, ChinaCat# HY-169505×SDS-PAGE Sample Loading BufferYeasenShanghai, ChinaCat# 20315ES2050× TAE BufferSangon BiotechShanghai, ChinaCat# B548101AgaroseBiosharpBeijing, ChinaCat# BS081Australia Origin Fetal Bovine SerumAvantorPennsylvania, USACat# 76294-180BCA Protein Assay KitThermo Fisher ScientificWilmington, USACat# A55864Collagenase IIAbsinShanghai, ChinaCat# abs47048001Corn oilSigma AldrichSaint Louis, MO, USACat# C8267CycloheximideSelleckShanghai, ChinaCat# S7418DAB solutionDakoHamburg, GermanyCat# K3468DAPISolarbioBeijing, ChinaCat# C0060DL2000 Plus DNA MarkerVazymeNanjing, Jiangsu, ChinaCat# MD101DMEM/F-12 (1:1) basic(1X)Thermo Fisher ScientificWilmington, USACat# 11330500EDTAServicebioWuhan, Hubei, ChinaCat# G1105H&E Staining KitServicebioWuhan, Hubei, ChinaCat# G1005Hieff® qPCR SYBR Green Master Mix (Low Rox Plus)YeasenShanghai, ChinaCat# 11202ESITS Cell Culture SupplementOriCellGuangzhou, Guangdong, ChinaCat# ITSS-10201kira6SelleckShanghai, ChinaCat# S8658MG132SelleckShanghai, ChinaCat# S2619Millex®-GPMerck MilliporeCork, IRLCat# SLGPR33RBParaformaldehydeServicebioWuhan, Hubei, ChinaCat# G1101PBSHyClonePA, USACat# SH30256.01Phosphatase Inhibitor Cocktail ⅠTargetMolShanghai, ChinaCat# C0002Phosphatase Inhibitor Cocktail ⅡTargetMolShanghai, ChinaCat# C0003PrimeScript™ RT Master MixTakaraBeijing, ChinaCat# RR036AProtease Inhibitor CocktailTargetMolShanghai, ChinaCat# C0001PVDF membraneBio-RadShanghai, ChinaCat# 1620264Quick Genotyping Assay Kit for Mouse TailBeyotimeShanghai, ChinaCat# D7283SRIPA Lysis BufferBeyotimeShanghai, ChinaCat# P0013CRNA Library Prep SetMGI TechShenzhen, Guangdong, ChinaCat# 1000006383Safranin O-Fast Green Staining Reagent For Bone TissueServicebioWuhan, Hubei, ChinaCat# G1053Skimmed milk powderServicebioWuhan, Hubei, ChinaCat# GC310001SuperRedBiosharpBeijing, ChinaCat# BS354BSurePAGE™GenScriptNanjing, Jiangsu, ChinaCat# M00657TamoxifenMedChemExpressShanghai, ChinaCat# HY-13757ATBS/Tween BufferEpizyme BiotechShanghai, ChinaCat# PS103TRIzolSigma AldrichSaint Louis, MO, USACat# 15596018AntibodiesAggrecan Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 13880-1-APAnti-ATF-4 antibodyAbcamUKCat# ab31390Anti-Caspase-3 Rabbit pAbServicebioWuhan, Hubei, ChinaCat# GB11767Anti-Collagen I antibodyAbcamUKCat# ab316222Anti-Collagen II antibodyAbcamUKCat# ab34712Anti-IRE1 antibodyAbcamUKCat# ab37073Anti-Ki67 antibodyAbcamUKCat# ab16667Anti-rabbit IgG (H+L) (DyLight 800 4X PEG Conjugate)Cell Signaling TechnologyDenver, MA, USACat# 5151Anti-SOX9 antibodyAbcamUKCat# ab185230BAK Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 29552-1-APBAX Recombinant Rabbit mAbStarterHangzhou, Zhejiang, ChinaCat# S0B0230Beta Actin Monoclonal antibodyProteintechWuhan, Hubei, ChinaCat# 66009-1-IgCaspase-7 Recombinant Rabbit mAbStarterHangzhou, Zhejiang, ChinaCat# S0B1309Collagen Type X Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 26984-1-APDDRGK1 Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 21445-1-APGoat Anti-Mouse IgG H&L (Alexa Fluor® 647)AbcamUKCat# ab150115GRP78/BIP Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 11587-1-APMMP-13 (E3Q4Y) Rabbit mAbCell Signaling TechnologyShanghai, ChinaCat# 67329MMP3 Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 17873-1-APMMP-9 (N-terminal) Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 10375-2-APExperimental models Ddrgk1fl/fl miceGempharmatechJiangsu, ChinaNO. T009014 Acan−CreERT2 miceGempharmatechJiangsu, ChinaNO. T055140Key instrumentationDigital Pathology Slide ScannerKFBIOZhejiang, ChinaKF-PRO-020Digital Radiography SystemFaxitron BiopticsUSAMultiFocusInverted phase contrast microscopeOLYMPUSTokyo, JapanCKX41LightCycler®480Roche DiagnosticsShanghai, ChinaN/AMicrotomeLeica BiosystemsHeidelberg, GermanyRM2016NanoDrop SpectrophotometerThermo Fisher ScientificWilmington, USAND-1000OdesseyLI-COR BiosciencesLincoln, NE, USAN/AProFlex™ PCR SystemThermo Fisher ScientificWilmington, USACat# 4484073SequencerMGI TechShenzhen, Guangdong, ChinaDNBSEQSoftwareBrandWebsiteVersionAdobe Illustratorhttps://www.adobe.com/cn/25.0Dr.Tomhttps://biosys.bgi.com/2.0GOhttp://www.geneontology.orgN/AGraphPad Prismhttps://www.graphpad.com/10.1.2GSEAhttps://www.gsea-msigdb.org/gsea/index.jsp4.3.3ImageJhttps://imagej.org1.54fKEGGhttps://www.kegg.jpN/AK-Viewerhttps://www.kfbio.cn/k-viewer1.7.1.1SPSShttps://www.ibm.com/29.0.0TBtoolshttps://github.com/CJ-Chen/TBtools-II/releases2.154 Table 1Key reagents or resources involved in this studyREAGENT or RESOURCESSOURCECOUNTRY/REGIONIDENTIFIERChemicals2 × Phanta UniFi Master MixVazymeNanjing, Jiang Su, ChinaCat# P5264-HydroxytamoxifenMedChemExpressShanghai, ChinaCat# HY-169505×SDS-PAGE Sample Loading BufferYeasenShanghai, ChinaCat# 20315ES2050× TAE BufferSangon BiotechShanghai, ChinaCat# B548101AgaroseBiosharpBeijing, ChinaCat# BS081Australia Origin Fetal Bovine SerumAvantorPennsylvania, USACat# 76294-180BCA Protein Assay KitThermo Fisher ScientificWilmington, USACat# A55864Collagenase IIAbsinShanghai, ChinaCat# abs47048001Corn oilSigma AldrichSaint Louis, MO, USACat# C8267CycloheximideSelleckShanghai, ChinaCat# S7418DAB solutionDakoHamburg, GermanyCat# K3468DAPISolarbioBeijing, ChinaCat# C0060DL2000 Plus DNA MarkerVazymeNanjing, Jiangsu, ChinaCat# MD101DMEM/F-12 (1:1) basic(1X)Thermo Fisher ScientificWilmington, USACat# 11330500EDTAServicebioWuhan, Hubei, ChinaCat# G1105H&E Staining KitServicebioWuhan, Hubei, ChinaCat# G1005Hieff® qPCR SYBR Green Master Mix (Low Rox Plus)YeasenShanghai, ChinaCat# 11202ESITS Cell Culture SupplementOriCellGuangzhou, Guangdong, ChinaCat# ITSS-10201kira6SelleckShanghai, ChinaCat# S8658MG132SelleckShanghai, ChinaCat# S2619Millex®-GPMerck MilliporeCork, IRLCat# SLGPR33RBParaformaldehydeServicebioWuhan, Hubei, ChinaCat# G1101PBSHyClonePA, USACat# SH30256.01Phosphatase Inhibitor Cocktail ⅠTargetMolShanghai, ChinaCat# C0002Phosphatase Inhibitor Cocktail ⅡTargetMolShanghai, ChinaCat# C0003PrimeScript™ RT Master MixTakaraBeijing, ChinaCat# RR036AProtease Inhibitor CocktailTargetMolShanghai, ChinaCat# C0001PVDF membraneBio-RadShanghai, ChinaCat# 1620264Quick Genotyping Assay Kit for Mouse TailBeyotimeShanghai, ChinaCat# D7283SRIPA Lysis BufferBeyotimeShanghai, ChinaCat# P0013CRNA Library Prep SetMGI TechShenzhen, Guangdong, ChinaCat# 1000006383Safranin O-Fast Green Staining Reagent For Bone TissueServicebioWuhan, Hubei, ChinaCat# G1053Skimmed milk powderServicebioWuhan, Hubei, ChinaCat# GC310001SuperRedBiosharpBeijing, ChinaCat# BS354BSurePAGE™GenScriptNanjing, Jiangsu, ChinaCat# M00657TamoxifenMedChemExpressShanghai, ChinaCat# HY-13757ATBS/Tween BufferEpizyme BiotechShanghai, ChinaCat# PS103TRIzolSigma AldrichSaint Louis, MO, USACat# 15596018AntibodiesAggrecan Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 13880-1-APAnti-ATF-4 antibodyAbcamUKCat# ab31390Anti-Caspase-3 Rabbit pAbServicebioWuhan, Hubei, ChinaCat# GB11767Anti-Collagen I antibodyAbcamUKCat# ab316222Anti-Collagen II antibodyAbcamUKCat# ab34712Anti-IRE1 antibodyAbcamUKCat# ab37073Anti-Ki67 antibodyAbcamUKCat# ab16667Anti-rabbit IgG (H+L) (DyLight 800 4X PEG Conjugate)Cell Signaling TechnologyDenver, MA, USACat# 5151Anti-SOX9 antibodyAbcamUKCat# ab185230BAK Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 29552-1-APBAX Recombinant Rabbit mAbStarterHangzhou, Zhejiang, ChinaCat# S0B0230Beta Actin Monoclonal antibodyProteintechWuhan, Hubei, ChinaCat# 66009-1-IgCaspase-7 Recombinant Rabbit mAbStarterHangzhou, Zhejiang, ChinaCat# S0B1309Collagen Type X Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 26984-1-APDDRGK1 Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 21445-1-APGoat Anti-Mouse IgG H&L (Alexa Fluor® 647)AbcamUKCat# ab150115GRP78/BIP Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 11587-1-APMMP-13 (E3Q4Y) Rabbit mAbCell Signaling TechnologyShanghai, ChinaCat# 67329MMP3 Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 17873-1-APMMP-9 (N-terminal) Polyclonal antibodyProteintechWuhan, Hubei, ChinaCat# 10375-2-APExperimental models Ddrgk1fl/fl miceGempharmatechJiangsu, ChinaNO. T009014 Acan−CreERT2 miceGempharmatechJiangsu, ChinaNO. T055140Key instrumentationDigital Pathology Slide ScannerKFBIOZhejiang, ChinaKF-PRO-020Digital Radiography SystemFaxitron BiopticsUSAMultiFocusInverted phase contrast microscopeOLYMPUSTokyo, JapanCKX41LightCycler®480Roche DiagnosticsShanghai, ChinaN/AMicrotomeLeica BiosystemsHeidelberg, GermanyRM2016NanoDrop SpectrophotometerThermo Fisher ScientificWilmington, USAND-1000OdesseyLI-COR BiosciencesLincoln, NE, USAN/AProFlex™ PCR SystemThermo Fisher ScientificWilmington, USACat# 4484073SequencerMGI TechShenzhen, Guangdong, ChinaDNBSEQSoftwareBrandWebsiteVersionAdobe Illustratorhttps://www.adobe.com/cn/25.0Dr.Tomhttps://biosys.bgi.com/2.0GOhttp://www.geneontology.orgN/AGraphPad Prismhttps://www.graphpad.com/10.1.2GSEAhttps://www.gsea-msigdb.org/gsea/index.jsp4.3.3ImageJhttps://imagej.org1.54fKEGGhttps://www.kegg.jpN/AK-Viewerhttps://www.kfbio.cn/k-viewer1.7.1.1SPSShttps://www.ibm.com/29.0.0TBtoolshttps://github.com/CJ-Chen/TBtools-II/releases2.154 Key reagents or resources involved in this study Animals Ddrgk1fl/fl mice and Acan-CreERT2 mice were purchased from GemPharmatech Co., Ltd., Jiangsu, China. Upon arrival, the mice were housed in an SPF-grade laboratory animal facility, maintained at a temperature of 26–28℃ and a humidity of 50%−65%. They were fed a standard diet and had free access to fresh water. Apart from necessary experimental procedures, no additional disturbances were made to their environment. Experimental manipulations were conducted at least one week after their arrival to allow the mice to adapt to the new living conditions. In this study, Ddrgk1fl/fl mice were first crossed with Acan-CreERT2 mice to generate Ddrgk1fl/+; Acan-CreERT2 heterozygous mice. These heterozygotes were then crossed to produce Ddrgk1cKO mice. The mice were fasted overnight prior to the experimental procedures, with free access to water. All experimental procedures involving mice in this study strictly adhered to the Regulations on the Administration of Laboratory Animals issued by State Scientific and Technological Commission of China, and were approved by the Laboratory Animal Committee of Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine. All mice were on a C57BL/6J background. Tamoxifen was dissolved in corn oil at 12.5 mg/ml and administered at 75 mg/kg i.p. to induce Acan-CreERT2 expression. For the early-postnatal group, administration occurred on postnatal days 1 to 3, and for the late-postnatal group, it was given for 5 consecutive days at one month of age of mice. Upon reaching the specified time point, the mice were euthanized. Approximately 0.5 cm of the mouse tail tissue was collected for genotype identification. The dorsal skin was incised with tissue scissors to isolate the spine. Then the spine was either immersed in a 4% paraformaldehyde (PFA) solution or the intervertebral discs were isolated for RNA extraction and sequencing analysis. Cell isolation and culture To isolate primary mouse chondrocytes, neonatal Ddrgk1fl/fl and Ddrgk1cKO mice (3–5 each, not injected with tamoxifen) aged 3 days were euthanized and immersed in 75% ethanol for 15 min. Subsequently, the knee joint was isolated from both hind limbs of the mice after the skin was removed. The synovium and muscle tissue were stripped carefully. The cartilage samples were then minced into pieces (0.5–1 mm in diameter) and digested in DMEM/F12 with 1% collagenase II at 37℃ for 4 h. After centrifugation (300 × g for 5 min), the cells were resuspended in complete medium (DMEM/F12 with 5% FBS and 1% penicillin-streptoycin). The primary mouse chondrocytes were cultured in complete medium with 1% insulin-transferrin-selenium (ITS) cell culture supplement at 37℃ and 5% CO2. The medium was refreshed every 3 days. 4-hydroxytamoxifen was added to the medium at 4µM after the first passage for 3 days to induce the conditional knockout of Ddrgk1 in the cells. High density culture To evaluate chondrocyte differentiation, 1.5 × 105 primary chondrocytes were resuspended in 10 µl medium and then seeded in 24-well plates. After 1 h of adhesion at 37℃, 500 µl of DMEM/F12 medium with 1% ITS cell culture supplement, 5% FBS and 1% penicillin-streptoycin was added into each well. After a 24-hour incubation 37℃ and 5% CO2, cells were treated with or without 6.25nM thapsigargin (Tg) and 0.6µM kira6. The medium was refreshed every 2 days. Following 7 days of incubation, the cell masses were fixed and staining with Alcian blue for 24 h at room temperature. Western blotting Primary mouse chondrocytes were cultured for 7 days, and total proteins were extracted using RIPA lysis buffer containing phosphatase and protease inhibitors. The extracted protein was quantified using BCA method and incubated with loading buffer at 95℃ for 10 min. Equal protein amounts were loaded onto a 4–20% SurePAGE™ gel for electrophoresis and then transferred to a 0.22 μm PVDF membrane. The membrane was blocked with 5% bovine serum albumin in TBS-Tween buffer (TBST) for 1 h at room temperature. Following TBST washes, it was then incubated with the primary antibodies (Table 1) at 4℃ overnight. The next day, membranes were washed with TBST for 3 times and incubated with anti-rabbit IgG (H + L) secondary antibody for 1 h at room temperature in the dark. Protein bands were then detected using a LI-COR Odyssey fluorescence imaging system. Lumbar spine instability (LSI) model construction The construction of the LSI surgery model closely replicates the process of human intervertebral disc degeneration, including degenerative manifestations such as reduced disc height, decreased vacuoles in the nucleus pulposus, ectopic bone formation in the cartilage endplate, and rupture of the annulus fibrosus [40]. The model was constructed as follows: 3-month-old WT and Ddrgk1cKO mice were anesthetized with isoflurane and placed in a prone position on the operating table. Then, lower back skin of mouse was prepared and sterilized. After the L5 vertebra was located, a longitudinal incision was made to expose the lumbar vertebrae. The posterior paraspinal muscles were dissected to reveal the L3-L5 spinous processes, which were resected along with the supraspinous and interspinous ligaments. Following suturing and sterilizing the skin, the mice were returned to their original environment and observed on alternative days. At the conclusion of the experiment, spinal tissue was obtained using the aforementioned method for further experimentation. H&E and safranin o-fast green staining At the time of euthanasia, the spines in mice were isolated and fixed in 4% PFA solution for 1 day. Spine samples were then decalcificated in 10% ethylene diamine tetraacetic acid solution for 2–3 weeks, dehydrated by graded ethanol and embedded in paraffin. Tissue sections were then stained by H&E or Safranin O-Fast Green staining solution according to the manufacturer’s instructions. After the sections were mounted, images were captured using the slide scanner for further analysis. Histological analyses Cartilage endplate thickness was calculated as the average thickness of five equal parts of endplate along the sagittal plane of spine. The cartilage proportion in endplate was determined using Image J software in Safranin O-Fast Green stained sections. The mean value from obtained from 3 to 5 intervertebral disc segments per mouse sample served as a data point. Histological scoring of intervertebral disc degeneration evaluates the cell density and morphological features of the annulus fibrosus and nucleus pulposus, as well as the boundary between them [41–43]. This scoring system includes five degeneration categories, each of which scores from 0 (indicating normal disc condition) to 3 (indicating severe degeneration). Detailed scoring information is presented in Table 2.Table 2Scoring system for degenerative changes in intervertebral discScoring items and scoresScoring criteriaNP morphology0Round shape and the NP constitutes >75% of the disc area1Round shape and the NP constitutes 50-75% of the disc area2Discontinuous round shape and the NP constitutes 25-50% of the disc area3Irregular shape and the NP constitutes <25% of the disc areaNP cellularity0Mostly star-shaped cells with proteoglycan matrix located peripherally and distributed evenly1Partially star-shaped and partially round cells, more star-shaped than round cells2More round cells than star-shaped cells, separated by a dense proteoglycan matrix3Mostly round cells separated by a dense proteoglycan matrixAF morphology0Well-arranged collagen lamellae without rupture1Inward bulging, ruptured, or serpentine fibers constitute <25% of the AF2Inward bulging, ruptured, or serpentine fibers constitute 25-50% of the AF3Inward bulging, ruptured, or serpentine fibers constitute >50% of the AFAF cellularity0Fibroblasts comprise >90% of the cells1Fibroblasts comprise 75-90% of the cells2Fibroblasts comprise 25-75% of the cells3Chondrocytes comprise >75% of the cellsBoundary of NP and AF0Normal without any interruption1Minimal interruption2Moderate interruption3Severe interruptionTotal:15 Table 2Scoring system for degenerative changes in intervertebral discScoring items and scoresScoring criteriaNP morphology0Round shape and the NP constitutes >75% of the disc area1Round shape and the NP constitutes 50-75% of the disc area2Discontinuous round shape and the NP constitutes 25-50% of the disc area3Irregular shape and the NP constitutes <25% of the disc areaNP cellularity0Mostly star-shaped cells with proteoglycan matrix located peripherally and distributed evenly1Partially star-shaped and partially round cells, more star-shaped than round cells2More round cells than star-shaped cells, separated by a dense proteoglycan matrix3Mostly round cells separated by a dense proteoglycan matrixAF morphology0Well-arranged collagen lamellae without rupture1Inward bulging, ruptured, or serpentine fibers constitute <25% of the AF2Inward bulging, ruptured, or serpentine fibers constitute 25-50% of the AF3Inward bulging, ruptured, or serpentine fibers constitute >50% of the AFAF cellularity0Fibroblasts comprise >90% of the cells1Fibroblasts comprise 75-90% of the cells2Fibroblasts comprise 25-75% of the cells3Chondrocytes comprise >75% of the cellsBoundary of NP and AF0Normal without any interruption1Minimal interruption2Moderate interruption3Severe interruptionTotal:15 Scoring system for degenerative changes in intervertebral disc Cartilage endplate degeneration can be assessed by a scoring system based on established literature [44, 45]. Key criteria include chondrocyte proliferation, cartilage disorganization, cartilage cracks, microfracture, new bone formation and bony sclerosis. The specific scoring details are presented in Table 3.Table 3Scoring system for degenerative changes in cartilage endplateScoring items and scoresScoring criteriaCell proliferation0Normal cellularity1Localized cell proliferation2Diffuse cell proliferation3Extensive cell proliferation4Scar/tissue defectsCartilage organization0Well-structured hyaline cartilage1Cartilage irregularities (or accompanied by obliterated vessels)2Disorganized matrix with thinning3Complete cartilage disorganization with defects4Scar/tissue defectsCartilage cracks0Absent1Rarely present2Present in intermediate amounts3Abundantly present4Scar/tissue defectsMicrofracture0Absent1Present2Scar/tissue defectsNew bone formation0Absent1Present2Scar/tissue defectsBony sclerosis0Absent1Present2Scar/tissue defectsTotal:18 Table 3Scoring system for degenerative changes in cartilage endplateScoring items and scoresScoring criteriaCell proliferation0Normal cellularity1Localized cell proliferation2Diffuse cell proliferation3Extensive cell proliferation4Scar/tissue defectsCartilage organization0Well-structured hyaline cartilage1Cartilage irregularities (or accompanied by obliterated vessels)2Disorganized matrix with thinning3Complete cartilage disorganization with defects4Scar/tissue defectsCartilage cracks0Absent1Rarely present2Present in intermediate amounts3Abundantly present4Scar/tissue defectsMicrofracture0Absent1Present2Scar/tissue defectsNew bone formation0Absent1Present2Scar/tissue defectsBony sclerosis0Absent1Present2Scar/tissue defectsTotal:18 Scoring system for degenerative changes in cartilage endplate Intervertebral disc height change was assessed by the disc height index. Specifically, the height of intervertebral disc and adjacent vertebral bodies were measured at the midline and 25% of the disc width bilaterally. The disc height index was then determined as the average of three disc height measurements divided by the average height of the adjacent vertebral bodies. Tissue RNA extraction and reverse transcription-quantitative PCR (RT-qPCR) After euthanasia, the mouse’s dorsal skin was incised to expose the spine, which was promptly isolated and submerged in pre-chilled phosphate buffer saline (PBS). Subsequently, microsurgical tools were employed to excise the muscle surrounding the spine and to collected the intervertebral discs into a centrifuge tube under a microscope. RNA was extracted from intervertebral discs using TRIzol, isopropanol and ethanol in sequence. RNA concentration was measured with a NanoDrop. The extracted RNA was subjected to reverse transcription using RT Master Mix to obtain the cDNA template. After the cDNA was diluted, a qPCR reaction system was prepared with SYBR Green Master Mix and specific primers. In the LightCycler 480, qPCR reaction conditions were as follows: Pre-denaturation (95℃ for 5 min), 40 cycles of denaturation (95℃ for 10 s), annealing (60℃ for 20 s) and extension (72℃ for 30 s), concluding with a melting curve stage. The obtained Ct values were then used to calculate the relative expression levels of target genes using the 2−ΔΔCt method [46]. The primers were designed using NCBI blast and the Actb gene served as an internal control (Table 4).Table 4Primers used for RT-qPCRGene NameOrganismsAccession NumberForward Sequence (5'→3')Reverse Sequence (5'→3')ActbMus musculusNM_007393.5TGT CCA CCT TCC AGC AGCGC AGC TCA GTA ACA GTC CAggrecanMus musculusNM_007424.3CGC TAC CCC ATC TCC AAGCCC ACC CCG AAG AAG TTTAtf4Mus musculusNM_009716.3CCT CAG ACA GTG AAC CCA AGGA GTG GAA GAC AGA ACC CBak1Mus musculusNM_007523.3CTC CCC GGA ACC TAT GAT TTCC CCG ATA CAT TTG GCTBaxMus musculusNM_007527.4TGC GTC CAC CAA GAA GCCCA CCC GGA AGA AGA CCBiPMus musculusNM_022310.3AGG AGG ACA AGA AGG AGG AGAG TGA AGG CCA CAT ACG ACaspase-3Mus musculusNM_009810.3TGT GCT AGA AAC GAA AGG GCCA TAA GGA GGC CAG GACaspase-7Mus musculusNM_007611.3TTC GGT TCC CTT GTC ATCTGG CTT GGT TTC CAT CTCCol10a1Mus musculusNM_009925.4GCC CTC AAG GTC CCA TAGCC GTT TTC ACC TCT TCT TCol1a1Mus musculusNM_007742.4CGA TGG ATT CCC GTT CGA GTGAG GCC TCG GTG GAC ATT AGCol2a1Mus musculusNM_031163.4GGC CAG GAT GCC CGA AAA TACG ATC ACC TCT GGG TCC TTDdrgk1Mus musculusNM_029832.3CTG TTA TTC CAG CCC AGG AGGCTC ACG TTG GGA CTC TAG GCIre1αMus musculusNM_023913.2CGA GCC ATG AGA AAC AAG AGGA AGC GGG AAG TGA AGMmp13Mus musculusNM_008607.2CTT TTC ACT GCG AGC GTTAAT GTA AGG CCA CCT CCA CMmp3Mus musculusNM_010809.3CCC CAC TCA CTT TCT CCAACT GTT CCA GGC CCA TCSox9Mus musculusNM_011448.4TGA AGA TGA CCG ACG AGC AGCGGA TGC ACA CGG GGA ACT TA Table 4Primers used for RT-qPCRGene NameOrganismsAccession NumberForward Sequence (5'→3')Reverse Sequence (5'→3')ActbMus musculusNM_007393.5TGT CCA CCT TCC AGC AGCGC AGC TCA GTA ACA GTC CAggrecanMus musculusNM_007424.3CGC TAC CCC ATC TCC AAGCCC ACC CCG AAG AAG TTTAtf4Mus musculusNM_009716.3CCT CAG ACA GTG AAC CCA AGGA GTG GAA GAC AGA ACC CBak1Mus musculusNM_007523.3CTC CCC GGA ACC TAT GAT TTCC CCG ATA CAT TTG GCTBaxMus musculusNM_007527.4TGC GTC CAC CAA GAA GCCCA CCC GGA AGA AGA CCBiPMus musculusNM_022310.3AGG AGG ACA AGA AGG AGG AGAG TGA AGG CCA CAT ACG ACaspase-3Mus musculusNM_009810.3TGT GCT AGA AAC GAA AGG GCCA TAA GGA GGC CAG GACaspase-7Mus musculusNM_007611.3TTC GGT TCC CTT GTC ATCTGG CTT GGT TTC CAT CTCCol10a1Mus musculusNM_009925.4GCC CTC AAG GTC CCA TAGCC GTT TTC ACC TCT TCT TCol1a1Mus musculusNM_007742.4CGA TGG ATT CCC GTT CGA GTGAG GCC TCG GTG GAC ATT AGCol2a1Mus musculusNM_031163.4GGC CAG GAT GCC CGA AAA TACG ATC ACC TCT GGG TCC TTDdrgk1Mus musculusNM_029832.3CTG TTA TTC CAG CCC AGG AGGCTC ACG TTG GGA CTC TAG GCIre1αMus musculusNM_023913.2CGA GCC ATG AGA AAC AAG AGGA AGC GGG AAG TGA AGMmp13Mus musculusNM_008607.2CTT TTC ACT GCG AGC GTTAAT GTA AGG CCA CCT CCA CMmp3Mus musculusNM_010809.3CCC CAC TCA CTT TCT CCAACT GTT CCA GGC CCA TCSox9Mus musculusNM_011448.4TGA AGA TGA CCG ACG AGC AGCGGA TGC ACA CGG GGA ACT TA Primers used for RT-qPCR RNA-seq Once RNA quality was confirmed using an Agilent 2100 bioanalyzer, sequencing was conducted on the extracted RNA. Following BGI’s protocol, key sequencing steps included transcriptome library construction, sequencing via the DNBSEQ platform, quality control of raw reads, data filtering, alignment with the reference genome. Downstream bioinformatic analyses included KEGG pathway enrichment, GO term enrichment and gene set enrichment analysis (GSEA) to elucidate the functional implications of the transcriptomic profiles. Immunohistochemical staining Suitable tissue slices were selected to dewax and rehydrate using xylene and graded ethanol. The tissue antigen was retrieved using a microwave. After blocked by 3% bull serum albumin at room temperature for 30 min and washed with PBS, the tissue then incubated with appropriate primary antibody at 4℃ overnight. The next day, the secondary antibody was added on the tissue to incubate for 1 h. Then the slices were stained with DAB solution and Harris hematoxylin in sequence monitored under microscope. Finally, the tissue slices were dehydrated by graded ethanol and xylene and mounted with neutral balsam for image capture and further analysis. The positive cell rate of Ki67, TUNEL, SOX9 and IRE1α was defined as the ratio of total positive cells to total cells in the cartilage endplate or nucleus pulposus from 3 to 5 lumbar disc segments of each mouse. The other indicators were measured as the proportion of the positive area by Image J. n represented the biological replicates of mice. Counting was performed blinded to genotype or modeling using identical exposure and threshold settings. TdT mediated dUTP nick end labeling (TUNEL) staining TUNEL staining was employed to identify apoptotic signals in intervertebral discs. Briefly, paraffin sections of spinal tissue were treated with xylene and graded ethanol to dewax and rehydrate. Then the sections underwent proteinase K treatment for 30 min at 37℃, followed by three PBS washes. After permeabilized by Triton X-100 for 20 min, the sections were incubated with a TdT-dUTP reaction mixture (2:29 ratio) for 2 h at 37℃. To block endogenous peroxidase activity, the sections were immersed into a 3% hydrogen peroxide solution for 15 min. Then converter-POD solution was added for 30 min incubation at 37℃. After treated with DAB working solution, cells exhibiting TUNEL-positive nuclear staining were identified as undergoing apoptosis. The apoptosis rate at specific region was determined by dividing the number of apoptotic cells by the total cell count. Statistical analysis Statistical analysis of the experimental data in this study was performed using IBM® SPSS® Statistics V29 software and GraphPad Prism 10 software. Data are presented as mean ± standard deviation (mean ± SD). For mouse data that followed a normal distribution, an unpaired two-tailed Student’s t-test was used to compare the differences between two groups. For mouse data that did not follow a normal distribution, the Mann-Whitney test was employed to assess statistical significance. In the volcano plot, genes were determined as differentially expressed based on screening criteria of |log2(cKO/WT)| > 1 and p < 0.05. In the GSEA analysis, a p-value < 0.05 and a false discovery rate (FDR) q-value < 0.25 were considered significant. For all other tests, a p-value < 0.05 was deemed significant. In the bar charts, * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001, **** represents p < 0.0001, and “ns” indicates no significant difference between the data. Results Ddrgk1 conditional knockout impeded spinal development in early postnatal phase To evaluate the importance of DDRGK1 in intervertebral disc development early postnatally, we generated Ddrgk1cKO mice to induce the knockout of Ddrgk1 in nucleus pulposus and cartilage endplate cells after birth (Fig. 1A, Figure S1). We administered Ddrgk1cKO and wild-type (WT) mice with tamoxifen (75 mg/kg, i.p., P1-P3) and examined the structural changes from 5 d to 38 d (Fig. 1B). Safranin O-Fast Green staining images showed significant abnormalities in the spinal structure of Ddrgk1cKO mice (Fig. 1C). First, the WT mice exhibited a continuous increase in vertebral height over time. Initially, there was no notable difference between Ddrgk1cKO and WT mice at 5 d and 8d. However, the vertebral growth in Ddrgk1cKO mice was greatly decelerated from 12 d and almost stagnated at 18 d and 22d. Significant reductions were all detected in vertebral height of Ddrgk1cKO mice at 12 d, 22 d and 30 d compared to the WT group (Fig. 1, D, G and J). Second, Ddrgk1cKO mice exhibited structural similarities with WT mice at 5 d and 8 d, with no notable macroscopic alterations in the nucleus pulposus, cartilage endplate and annulus fibrosus. However, the nucleus pulposus in Ddrgk1cKO mice began to deform at 12 d, transitioning from an oval to a rectangular at 22 d and eventually to an irregular shape at 30d. Additionally, the cartilage endplate displayed structural distortion and disordered cartilage layers in Ddrgk1cKO mice from 12 d (Fig. 1C). Nevertheless, no significant differences in disc height were detected at 12 d, 22 d and 30 d between Ddrgk1cKO and WT mice (Fig. 1, E, H and K). Moreover, the disc height index of Ddrgk1cKO mice significantly increased that of WT mice at the same time points considering the changes in vertebral height (Fig. 1, F, I and L). These results indicated pronounced structural abnormalities in both vertebrae and discs after Ddrgk1 conditional knockout in the early postnatal phase of development.Fig. 1Ddrgk1cKO mice showed abnormal spinal structure in early postnatal life. (A) Schematic design of Ddrgk1 conditional knockout in mice. (B) Timeline of tamoxifen injection and sample harvest of Ddrgk1cKO and WT mice within 38 days postnatally. (C) Safranin O-Fast Green staining of partial lumbar segment of the WT and Ddrgk1cKO mice from 5 to 38 days of age. Black arrows in each image are used to indicate the approximate height of vertebrae. Scale bar: 400 μm. (D-F) Statistical analysis of (D) vertebral height, (E) disc height and (F) disc height index of WT and Ddrgk1cKO mice at 12d. (G-I) Statistical analysis of (G) vertebral height, (H) disc height and (I) disc height index of WT and Ddrgk1cKO mice at 22d. (J-L) Statistical analysis of (J) vertebral height, (K) disc height and (L) disc height index of WT and Ddrgk1cKO mice at 30d. Data are represented as mean ± SD. n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1cKO mice showed abnormal spinal structure in early postnatal life. (A) Schematic design of Ddrgk1 conditional knockout in mice. (B) Timeline of tamoxifen injection and sample harvest of Ddrgk1cKO and WT mice within 38 days postnatally. (C) Safranin O-Fast Green staining of partial lumbar segment of the WT and Ddrgk1cKO mice from 5 to 38 days of age. Black arrows in each image are used to indicate the approximate height of vertebrae. Scale bar: 400 μm. (D-F) Statistical analysis of (D) vertebral height, (E) disc height and (F) disc height index of WT and Ddrgk1cKO mice at 12d. (G-I) Statistical analysis of (G) vertebral height, (H) disc height and (I) disc height index of WT and Ddrgk1cKO mice at 22d. (J-L) Statistical analysis of (J) vertebral height, (K) disc height and (L) disc height index of WT and Ddrgk1cKO mice at 30d. Data are represented as mean ± SD. n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1 conditional knockout disturbed cellular structure and induced degeneration of intervertebral discs in early postnatal phase To further investigate the abnormal phenotype in the disc of Ddrgk1cKO mice in early postnatal phase, we selected 12 d, 22 d, and 30 d of age as 3 time points after tamoxifen administration (75 mg/kg, i.p., P1-P3) for detailed scoring and statistical analysis of nucleus pulposus and cartilage endplate regions (Fig. 2, A, E and I, Figure S2). Quantitative analysis showed that cartilage endplate thickness in Ddrgk1cKO mice was significantly reduced by approximately 5.70%, 30.48%, and 43.26% at 12 d, 22 d, and 30 d compared to WT mice (Fig. 2, B, F and J). Furthermore, the cartilage endplate thickness in Ddrgk1cKO mice progressively decreased with age, suggesting developmental defects in the cartilage endplate cells. In addition, Ddrgk1cKO mice exhibited significant degenerative changes in cartilage endplate, manifested as abnormal chondrocyte proliferation in the central and peripheral region and disorganized cartilage structure in the central region at all three time points (Fig. 2, A and C, E and G, I and K). We also found significant degenerative changes in the nucleus pulposus of Ddrgk1cKO mice (Fig. 2, A and D, E and H, I and L). First, the nucleus pulposus of Ddrgk1cKO mice maintained an oval shape at 12 d, but by 22 d and 30 d, its morphology became increasingly irregular. Second, the nucleus pulposus cells of Ddrgk1cKO mice became larger and rounder with significant alterations in the ECM compared to the WT mice at 12d. These structural changes in the nucleus pulposus cells and ECM progressed further by 22 d and 30 d in the Ddrgk1cKO mice.Fig. 2Ddrgk1cKO mice exhibited significant disc degeneration within 30 days. (A), (E) and (I) Representative Safranin O-Fast Green staining images of lumbar disc from WT or Ddrgk1cKO mice at (A) 12 d, (E) 22 d, and (I) 30d. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate (EP), peripheral region of superior cartilage endplate and central nucleus pulposus (NP), where the scale bar represents 50 μm. (B-D) Statistical analysis of (B) cartilage endplate thickness, (C) cartilage endplate degeneration score, (D) histological score of disc degeneration of WT and Ddrgk1cKO mice at 12d. (F-H) Statistical analysis of (F) cartilage endplate thickness (G) cartilage endplate degeneration score, (H) histological score of disc degeneration of WT and Ddrgk1cKO mice at 22d. (J-L) Statistical analysis of (J) cartilage endplate thickness (K) cartilage endplate degeneration score, (L) histological score of disc degeneration of WT and Ddrgk1cKO mice at 30d. Data are represented as mean ± SD. n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 Ddrgk1cKO mice exhibited significant disc degeneration within 30 days. (A), (E) and (I) Representative Safranin O-Fast Green staining images of lumbar disc from WT or Ddrgk1cKO mice at (A) 12 d, (E) 22 d, and (I) 30d. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate (EP), peripheral region of superior cartilage endplate and central nucleus pulposus (NP), where the scale bar represents 50 μm. (B-D) Statistical analysis of (B) cartilage endplate thickness, (C) cartilage endplate degeneration score, (D) histological score of disc degeneration of WT and Ddrgk1cKO mice at 12d. (F-H) Statistical analysis of (F) cartilage endplate thickness (G) cartilage endplate degeneration score, (H) histological score of disc degeneration of WT and Ddrgk1cKO mice at 22d. (J-L) Statistical analysis of (J) cartilage endplate thickness (K) cartilage endplate degeneration score, (L) histological score of disc degeneration of WT and Ddrgk1cKO mice at 30d. Data are represented as mean ± SD. n = 6. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 Taken together, these findings indicated that Ddrgk1cKO mice exhibited significant disc development abnormalities, demonstrated the critical role of Ddrgk1 gene in the early postnatal development and morphological maintenance of intervertebral discs in mice. Ddrgk1 conditional knockout accelerated intervertebral disc degeneration in late postnatal phase To examine the impact of Ddrgk1 conditional knockout on the intervertebral discs in the late postnatal phase, we induced Ddrgk1 conditional knockout in mice by tamoxifen (75 mg/kg, i.p., 5 consecutive days) at 1 month old and harvested specimens for analysis at 6 months of age (Fig. 3A).Fig. 3Ddrgk1cKO mice exhibited severe degenerative features in intervertebral discs in later developmental phase. (A) Schematic diagram of establishment for the late stage of intervertebral disc development in WT and Ddrgk1cKO mice. Tamoxifen was administered intraperitoneally at 1 month of age, at 75 mg/kg daily for 5 consecutive days. Mice were harvested at 6 months old. (B-D) (B) Gross observation images, (C) dorsoventral X-ray images and (D) statistical comparison of body length of 6-month-old WT and Ddrgk1cKO mice. n = 8. (E) Focused X-ray images of the lumbar spine region. T12 represents the 12th thoracic vertebra. L1-L6 represent the 1 st to 6th lumbar vertebrae. (F) Statistical comparison of the length from L1 to L6. n = 8. (G) Safranin O-Fast Green staining of partial lumbar segment of the WT and Ddrgk1cKO mice at 6 months of age. Black arrows in each image are used to indicate the approximate height of vertebrae. Scale bar: 600 μm. (H-J) Statistical analysis of (H) disc height, (I) vertebral height and (J) disc height index of WT and Ddrgk1cKO mice at 6 months. n = 6. (K) Representative Safranin O-Fast Green staining images of lumbar disc from 6-month-old WT or Ddrgk1cKO mice. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate, peripheral region of superior cartilage endplate and central nucleus pulposus, where the scale bar represents 50 μm. (L-O) Statistical analysis of (L) cartilage endplate thickness, (M) proportion of cartilage tissue in cartilage endplate, (N) cartilage endplate degeneration score and (O) intervertebral disc degeneration score of WT and Ddrgk1cKO mice at 6 months old. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data. First, 6-month-old Ddrgk1cKO mice were significantly shorter in body length compared to wild-type mice (Fig. 3, B-D). Furthermore, the length from L1 to L6 of Ddrgk1cKO mice was also significantly shorter than that of WT mice (Fig. 3, E and F). Safranin O-Fast Green staining images of lumbar segments in 6-month-old Ddrgk1cKO and WT mice revealed a significant reduction in both disc and vertebral body heights, yet with no significant difference in the comparison of disc height index (Fig. 3, G-J). Then, more detailed analysis was conducted to examine the degenerative changes in the cells and matrices of different disc components of the Ddrgk1cKO mice (Fig. 3K, Figure S3). Notably, 6-month-old Ddrgk1cKO mice exhibited pronounced degenerative alterations across the sampled lumbar levels (L1-L6) consistently. First, the cartilage endplate of Ddrgk1cKO mice was characterized by substantially reduced thickness in the central region and distinct ossification in the peripheral region with few remaining cartilaginous cells (Fig. 3K). Statistical analysis revealed that both the thickness and the cartilage tissue proportion of endplate in the Ddrgk1cKO mice were significantly reduced compared to WT mice (18.12 ± 7.13 μm vs. 68.96 ± 4.28 μm, 8.21 ± 2.30% vs. 68.14 ± 11.15%) (Fig. 3, L and M). Additionally, featured mainly by distinct cartilage tissue defect in endplate, the cartilage endplate degeneration score of Ddrgk1cKO mice was significantly higher than WT mice (Fig. 3N). Notable degenerative changes were also observed in nucleus pulposus of 6-month-old Ddrgk1cKO mice (Fig. 3, G and K). The nucleus pulposus of 6-month-old Ddrgk1cKO mice showed severe structural deformation with large and round cells separated by dense areas of proteoglycan matrix. In brief, the disc degeneration score significantly increased considering these changes in discs of 6-month-old Ddrgk1cKO mice (Fig. 3O). Ddrgk1cKO mice exhibited severe degenerative features in intervertebral discs in later developmental phase. (A) Schematic diagram of establishment for the late stage of intervertebral disc development in WT and Ddrgk1cKO mice. Tamoxifen was administered intraperitoneally at 1 month of age, at 75 mg/kg daily for 5 consecutive days. Mice were harvested at 6 months old. (B-D) (B) Gross observation images, (C) dorsoventral X-ray images and (D) statistical comparison of body length of 6-month-old WT and Ddrgk1cKO mice. n = 8. (E) Focused X-ray images of the lumbar spine region. T12 represents the 12th thoracic vertebra. L1-L6 represent the 1 st to 6th lumbar vertebrae. (F) Statistical comparison of the length from L1 to L6. n = 8. (G) Safranin O-Fast Green staining of partial lumbar segment of the WT and Ddrgk1cKO mice at 6 months of age. Black arrows in each image are used to indicate the approximate height of vertebrae. Scale bar: 600 μm. (H-J) Statistical analysis of (H) disc height, (I) vertebral height and (J) disc height index of WT and Ddrgk1cKO mice at 6 months. n = 6. (K) Representative Safranin O-Fast Green staining images of lumbar disc from 6-month-old WT or Ddrgk1cKO mice. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate, peripheral region of superior cartilage endplate and central nucleus pulposus, where the scale bar represents 50 μm. (L-O) Statistical analysis of (L) cartilage endplate thickness, (M) proportion of cartilage tissue in cartilage endplate, (N) cartilage endplate degeneration score and (O) intervertebral disc degeneration score of WT and Ddrgk1cKO mice at 6 months old. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data. Therefore, Ddrgk1 conditional knockout could lead to severe accelerated disc degeneration, further indicating the indispensable role of DDRGK1 in maintaining disc homeostasis and delaying the onset of disc degeneration. Ddrgk1 conditional knockout aggravates disc degeneration under the condition of lumbar spine instability In order to investigate the impact of DDRGK1 on the progression of lumbar disc degeneration, we administered tamoxifen (75 mg/kg, i.p., 5 consecutive days) at 1 month old, conducted LSI surgery at 3 months old and harvested specimens for analysis at 6 months old in both WT and Ddrgk1cKO mice (Fig. 4A). In the Safranin O-Fast Green stained sections, we observed that WT mice displayed notable degenerative changes after surgery, including local ossification of the cartilage endplate and minor vacuolar changes in nucleus pulposus cells (Fig. 4B). In contrast, Ddrgk1cKO mice exhibited more pronounced degeneration. Specifically, the cartilage endplate of Ddrgk1cKO mice showed a significant reduction in thickness and a significant loss of cartilage components after LSI surgery (Fig. 4, B-E), whereas the WT mice, despite partial ossification in endplate, were able to maintain a certain thickness and structural integrity of cartilage endplate. Moreover, Ddrgk1cKO mice showed extensive vacuolization of nucleus pulposus cells, indicating widespread changes in the ECM properties (Fig. 4, B and F). Therefore, we believed that under the pathological condition of lumbar instability, the intervertebral disc degeneration of Ddrgk1cKO mice was far more severe and extensive than the WT mice.Fig. 4Ddrgk1cKO mice exhibited severe disc degeneration after lumbar spine instability surgery. (A) Schematic diagram of establishment of LSI surgery in WT and Ddrgk1cKO mice. Tamoxifen was administered intraperitoneally at 1 month of age, at 75 mg/kg daily for 5 consecutive days. The surgery was conducted at 3 months of age and mice were harvested at 6 months of age. (B) Representative Safranin O-Fast Green staining images of lumbar discs from WT or Ddrgk1cKO mice after LSI surgery. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate, peripheral region of superior cartilage endplate and central region of nucleus pulposus, where the scale bar represents 50 μm. (C-F) Statistical analysis of (C) cartilage endplate thickness, (D) proportion of cartilage tissue in cartilage endplate, (E) cartilage endplate degeneration score, (F) intervertebral disc degeneration score of WT and Ddrgk1cKO mice after LSI surgery. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data. Ddrgk1cKO mice exhibited severe disc degeneration after lumbar spine instability surgery. (A) Schematic diagram of establishment of LSI surgery in WT and Ddrgk1cKO mice. Tamoxifen was administered intraperitoneally at 1 month of age, at 75 mg/kg daily for 5 consecutive days. The surgery was conducted at 3 months of age and mice were harvested at 6 months of age. (B) Representative Safranin O-Fast Green staining images of lumbar discs from WT or Ddrgk1cKO mice after LSI surgery. The leftmost image shows the general view of the intervertebral disc region, where the scale bar represents 300 μm. Yellow labels 1–3 correspond to the magnified images of the central region of superior cartilage endplate, peripheral region of superior cartilage endplate and central region of nucleus pulposus, where the scale bar represents 50 μm. (C-F) Statistical analysis of (C) cartilage endplate thickness, (D) proportion of cartilage tissue in cartilage endplate, (E) cartilage endplate degeneration score, (F) intervertebral disc degeneration score of WT and Ddrgk1cKO mice after LSI surgery. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data. RNA-seq revealed transcriptional reprogramming in intervertebral discs after Ddrgk1 conditional knockout Next, to detect the transcriptomic alterations after Ddrgk1 conditional knockout, we performed RNA sequencing analysis on intervertebral disc cells from WT and Ddrgk1cKO mice at 2 months after tamoxifen injection (75 mg/kg, i.p., 5 consecutive days) at 1 month old (Fig. 5A). Principle component analysis (PCA), volcano plot, and cluster analysis revealed distinct gene expression profiles in Ddrgk1cKO mice compared to WT, identifying 1,385 differentially expressed genes between the two groups (Fig. 5, B-D). GSEA revealed Ddrgk1 deletion significantly upregulated the genes associated with apoptosis (e.g., Bak1, Casp3, Casp7 and Casp8) activation of matrix metalloproteinases (e.g., Mmp3 and Mmp9), ECM degradation (e.g., Col1a1 and Fn1), and ER unfolded protein response (e.g., Ern1(Ire1), Eif2ak3, Xbp1 and Atf6b) in intervertebral disc cells (Fig. 5, E-H; Figure S4). The KEGG enrichment analysis revealed that several signal transduction pathways, including MAPK, TNF, NF-kappa B, and Ras, were notably enriched, suggesting activation of inflammation-related pathways after Ddrgk1 knockout (Fig. 5I). Furthermore, significant enrichment of SUMOylation-related pathways highlighted a potential impact of Ddrgk1 knockout on ubiquitin-like modification pathways in disc cells (Fig. 5J).Fig. 5Ddrgk1cKO mice showed transcriptional reprogramming in intervertebral disc tissues. (A) Schematic illustration of tamoxifen administration (75 mg/kg, i.p., 5 consecutive days) at 1 month old and sampling for RNA-seq at 2 months old of both WT and Ddrgk1cKO mice. n = 3. (B) PCA based on the differentially expressed genes (DEGs) showing clustering of transcriptomic profiles of intervertebral disc cells. (C) Volcano plot of DEGs for Ddrgk1cKO vs. WT. 408 genes were up-regulated and 977 gene were down-regulated based on the screening criteria of |log2(cKO/WT)| > 1 and p < 0.05. (D) Heat map of hierarchical cluster based on the DEGs in panel (C). The normalization method is z-score by row. Some of the DEGs of interest were labeled on the right of the panel. (E-H) GSEA plots of (E) apoptosis, (F) activation of matrix metalloproteinases, (G) degradation of the ECM and (H) ER unfolded protein response for Ddrgk1cKO vs. WT. A p-value < 0.05 and an FDR q-value < 0.25 indicate significant differences. (I and J) Bubble plots of DEGs enriched in (I) KEGG term “Signal transduction” and (J) Reactome term “SUMOylation” for Ddrgk1cKO vs. WT Ddrgk1cKO mice showed transcriptional reprogramming in intervertebral disc tissues. (A) Schematic illustration of tamoxifen administration (75 mg/kg, i.p., 5 consecutive days) at 1 month old and sampling for RNA-seq at 2 months old of both WT and Ddrgk1cKO mice. n = 3. (B) PCA based on the differentially expressed genes (DEGs) showing clustering of transcriptomic profiles of intervertebral disc cells. (C) Volcano plot of DEGs for Ddrgk1cKO vs. WT. 408 genes were up-regulated and 977 gene were down-regulated based on the screening criteria of |log2(cKO/WT)| > 1 and p < 0.05. (D) Heat map of hierarchical cluster based on the DEGs in panel (C). The normalization method is z-score by row. Some of the DEGs of interest were labeled on the right of the panel. (E-H) GSEA plots of (E) apoptosis, (F) activation of matrix metalloproteinases, (G) degradation of the ECM and (H) ER unfolded protein response for Ddrgk1cKO vs. WT. A p-value < 0.05 and an FDR q-value < 0.25 indicate significant differences. (I and J) Bubble plots of DEGs enriched in (I) KEGG term “Signal transduction” and (J) Reactome term “SUMOylation” for Ddrgk1cKO vs. WT To validate the RNA-seq findings, we isolated primary chondrocytes from WT and Ddrgk1cKO mice to verify changes in gene transcription and protein expression via RT-qPCR and WT assays (Figure S5 and Figure S6). First of all, effective in vitro knockout of Ddrgk1 was confirmed at both transcriptional and protein levels (Figure S5A and S6A). In Ddrgk1cKO chondrocytes, transcriptional levels of Ire1, Grp78 and Atf4 were elevated, aligning with the RNA-seq findings (Figure S4C; Figure S5, B-D). Western blot analysis revealed an increase in BiP levels and a decrease in IRE1α, indicating the activation of ER stress and possible reduction of stability of IRE1α after Ddrgk1 knockout (Figure S6A). Secondly, the pro-apoptotic factors, such as BAK1, BAX, Caspase-3 and Caspase-7, were upregulated (Figure S5, E-H; Figure S6B). Furthermore, increased expression of MMP3, MMP9, and MMP13, along with decreased SOX9 expression and Acan transcription, collectively indicated the remodeling of the cartilage ECM (Figure S5, I-L; Figure S6C). These results suggested implied that DDRGK1 plays a crucial role in preserving the intervertebral disc structure by regulating nucleus pulposus and cartilage endplate cell survival and apoptosis, ECM metabolic balance and ER homeostasis. Furthermore, the conditional knockout of Ddrgk1 appeared to induce alterations in multiple inflammatory signaling pathways and post-translational modification processes. Ddrgk1 conditional knockout induced cell apoptosis and suppressed proliferation Combining the degeneration of disc structure observed in histology and the upregulation of apoptosis-related genes by transcriptomics in Ddrgk1cKO mice, we then conducted Ki67 and TUNEL immunohistochemical staining on disc tissues from Ddrgk1cKO and WT mice at 12 d, 22 d, 30 d, 6 m and after LSI modeling to further validate these findings (Fig. 6, A and B). Statistical analysis demonstrated significant reductions of Ki67 positive rates in the nucleus pulposus and cartilage endplate cells of Ddrgk1cKO mice at all time points (Fig. 6C), while TUNEL positive rates were significantly elevated (Fig. 6D). Specifically, the Ki67 positive rates were 6.73 ± 1.49% and 0.28 ± 0.31% in nucleus pulposus cells and 7.13 ± 0.72% and 0.68 ± 0.53% in cartilage endplate cells of WT and Ddrgk1cKO mice at 12d. In contrast, the TUNEL positive rates were 4.64 ± 2.10% and 38.52 ± 9.86% in nucleus pulposus cells and 6.19 ± 3.19% and 37.54 ± 7.86% in cartilage endplate cells of WT and Ddrgk1cKO mice at 12d. Besides, cell density of nucleus pulposus decreased at 22 d, 30 d, 6 m; while cell density of cartilage endplate remained unchanged in early phase but decreased at 6 m (Figure S7). Taken together, these results suggested that Ddrgk1 conditional knockout significantly suppressed cell proliferation and enhanced apoptosis in the nucleus pulposus and cartilage endplate during early and late development phases and degeneration condition.Fig. 6Ddrgk1cKO mice exhibited increased apoptosis and decreased proliferation in intervertebral discs. (A and C) (A) Ki67 immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (C) Statistical analyses of Ki67+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (B and D) (B) TUNEL immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of TUNEL+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. Scale bar: 50 μm. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1cKO mice exhibited increased apoptosis and decreased proliferation in intervertebral discs. (A and C) (A) Ki67 immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (C) Statistical analyses of Ki67+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (B and D) (B) TUNEL immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of TUNEL+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. Scale bar: 50 μm. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1 conditional knockout inhibited cartilage development and disrupted ECM turnover in intervertebral discs To investigate the effects of Ddrgk1 conditional knockout on cartilage tissue development and ECM turnover in mouse intervertebral discs, we performed immunohistochemical staining for SOX9, MMP3, and Aggrecan on discs from WT and Ddrgk1cKO mice at the time points mentioned previously (Fig. 7). The staining results indicated a significant decrease of SOX9 expression (Fig. 7, A and D) and Aggrecan expression (Fig. 7, C and F) and a significant increase of MMP3 expression (Fig. 7, B and E) in Ddrgk1cKO mice compared to WT mice. Specifically, the SOX9 positive rates were 68.24 ± 14.32% and 16.14 ± 13.16% in nucleus pulposus cells and 53.19 ± 8.07% and 8.01 ± 5.57% in cartilage endplate cells of WT and Ddrgk1cKO mice at 6 months of age. In contrast, the relative MMP3 expression rates were 3.46 ± 0.53% and 13.68 ± 2.61% in nucleus pulposus and 1.43 ± 0.81% and 7.55 ± 1.11% in cartilage endplate of WT and Ddrgk1cKO mice at 6 months of age. Besides, to examine expression patterns of collagen in nucleus pulposus and cartilage endplate after Ddrgk1 conditional knockout, immunohistochemical staining for COL1A1, COL2A1, and COL10A1 was then conducted on discs from WT and Ddrgk1cKO mice at the time points mentioned previously (Figure S8). The results showed that, compared with WT mice, Ddrgk1cKO mice exhibited significantly elevated expression of COL1A1 (Figure S8, A and D) and COL10A1 (Figure S8, C and F) but reduced expression of COL2A1 (Figure S8, B and E) in the nucleus pulposus and cartilage endplate at the detected time points.Fig. 7Ddrgk1cKO mice showed diminished SOX9 and Aggrecan expression and increased MMP3 expression in intervertebral discs. (A and D) (A) SOX9 IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of SOX9+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (B and E) (B) MMP3 IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (E) Quantitative analyses of MMP3 expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (C and F) (C) Aggrecan IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (F) Quantitative analyses of Aggrecan expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. Scale bar: 50 μm. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1cKO mice showed diminished SOX9 and Aggrecan expression and increased MMP3 expression in intervertebral discs. (A and D) (A) SOX9 IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of SOX9+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (B and E) (B) MMP3 IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (E) Quantitative analyses of MMP3 expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. (C and F) (C) Aggrecan IHC staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (F) Quantitative analyses of Aggrecan expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. n = 6. Scale bar: 50 μm. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data In brief, these data implied that Ddrgk1 conditional knockout inhibited cartilage differentiation, disrupted the ECM turnover and disc homeostasis during disc development and degeneration. Ddrgk1 conditional knockout triggered ER stress in intervertebral discs To further explore the influence on ER stress after Ddrgk1 conditional knockout in intervertebral discs, we performed immunohistochemical staining for IRE1α, ATF4, and BiP at aforementioned time points(Fig. 8, A-C).Fig. 8Ddrgk1cKO mice exhibited elevated ER-stress levels in intervertebral discs. (A and D) (A) IRE1α immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of IRE1α+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. (B and E) (B) ATF4 immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (E) Quantitative analyses of ATF4 expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. (C and F) (C) BiP immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (F) Quantitative analyses of BiP expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above.Scale bar: 50 μm. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Ddrgk1cKO mice exhibited elevated ER-stress levels in intervertebral discs. (A and D) (A) IRE1α immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (D) Statistical analyses of IRE1α+ cells rate in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. (B and E) (B) ATF4 immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (E) Quantitative analyses of ATF4 expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above. (C and F) (C) BiP immunohistochemical staining images in the central region of nucleus pulposus and superior cartilage endplate of WT and Ddrgk1cKO mice at 12 d, 22 d, 30 d, 6 m, and after LSI surgery. (F) Quantitative analyses of BiP expression in the nucleus pulposus and cartilage endplate regions between WT and Ddrgk1cKO mice at the corresponding time points above.Scale bar: 50 μm. n = 6. Data are represented as mean ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns represents no statistical difference between data Statistical analysis showed that Ddrgk1cKO mice exhibited significantly lower positive rates of IRE1α (Fig. 8D). For instance, the IRE1α positive rates were 51.07 ± 5.51% and 22.58 ± 5.84% in nucleus pulposus cells as well as 46.40 ± 8.79% and 18.68 ± 3.33% in cartilage endplate cells of WT and Ddrgk1cKO mice at 30d. The expression of ATF4 was significantly increased in the nucleus pulposus and cartilage endplate cells at all time points after Ddrgk1 conditional knockout (e.g., 7.39 ± 3.38% and 18.07 ± 5.36% in nucleus pulposus, 6.36 ± 1.65% and 15.74 ± 2.60% in cartilage endplate of WT and Ddrgk1cKO mice at 22 d) (Fig. 8E). Except no significance of BiP expression was found in the nucleus pulposus after LSI modeling, significant increases of BiP expression were detected in other time points of nucleus pulposus and cartilage endplate in Ddrgk1cKO mice (e.g., 6.89 ± 2.92% and 21.03 ± 2.35% in nucleus pulposus, 2.28 ± 0.56% and 9.06 ± 3.08% in cartilage endplate of WT and Ddrgk1cKO mice) (Fig. 8F). These findings indicated that conditional knockout of Ddrgk1 could reduce the stability of IRE1α and induce ER stress during intervertebral disc development and degeneration. Discussion Our study demonstrated that DDRGK1 as a critical component of UFMylation is essential for preserving the tissue structure and cellular phenotype of the cartilage endplate and nucleus pulposus as well as supporting vertebral growth during the developing phase. We found that Ddrgk1 conditional knockout in both early or late postnatal phase could lead to severe structural abnormalities in the cartilage endplate and nucleus pulposus and significant disc degeneration in adulthood (Figure 1-3). Notably, significant structural abnormalities were observed in both nucleus pulposus and cartilage endplate inDdrgk1cKO mice in early postnatal phase, but no significant reduction was detected in disc height in Ddrgk1cKO mice (Figure 1 and 2). However, distinct cartilage endplate defect emerged as a key contributor to the significant decrease in disc height of Ddrgk1cKO mice in the late postnatal phase along with pronounced degenerative phenotype (Figure 3). RNA sequencing indicated the upregulation of apoptosis-related genes after Ddrgk1 conditional knockout (Figure 5E). Immunohistochemical staining confirmed that Ddrgk1 conditional knockout reduced cell proliferation (Ki67) and increased apoptosis (TUNEL) in nucleus pulposus and cartilage endplate cells (Figure 6). Interestingly,Ddrgk1 conditional knockout did not reduce cell density in nucleus pulposus at 12d and in cartilage endplate during early postnatal phase, but did so significantly in later phase (Figure S7); meanwhile, significant changes in ECM indicators were observed in both phases (Figure 7 and Figure S8), which implied that DDRGK1 may regulate cell fate and matrix turnover via distinct pathways. The deficiency of the components in UFM1 system has been shown to impede cartilage development. Homozygous loss-of-function mutations inDDRGK1 or heterozygous mutations in UFM1-specific protease 2 gene (UFSP2)could lead to spondylo-epi-metaphyseal dysplasias (SEMD), a chondrodysplasia characterized by abnormal development of spine, epiphysis and metaphysis [34, 47, 48]. Patients with SEMD may present with scoliosis, spinal compression fractures, flattened vertebral bodies, and insufficient mineralization of long bones [49]. Furthermore, a loss-of-function mutation in UFSP2 has also been reported to be associated with Beukes hip dysplasia, characterized by proximal femoral deformities and severe osteoarthropathy [50]. Further researches revealed that conditional knockout of Ddrgk1 in mesenchymal cells leads to progressively limb shortening and joint abnormalities, while mice with Ddrgk1 conditional knockout in osteoblasts did not show apparent phenotype of limbs [34, 37]. Collectively, these studies highlight the critical role of the UFM1 system in regulating cartilage development. ER stress and the activation of the unfolded protein response (UPR) play crucial roles in the abnormal development and accelerated degeneration of intervertebral discs caused by Ddrgk1 knockout. Our results showed significant enrichment of gene sets related to the ER unfolded protein response in intervertebral disc cells after Ddrgk1 knockout. Immunohistochemical staining revealed significantly increased levels of ATF4 and BiP, indicating ER stress induction and UPR pathway activation following Ddrgk1 knockout. Interestingly, we found that while the transcription level of Ire1 was upregulated afterDdrgk1 knockout (Figure S4C), its tissue protein level was significantly decreased compared to WT mice (Figure 8D), which was further validated by RT-qPCR and Western blot assays in primary chondrocytes in vitro (Figure S5B and Figure S6A). Besides, CHX assay indicated that proteasome-dependent pathway could be the principal pathway for IRE1α degradation (Figure S9). We also observed that modulation of ER stress through IRE1α pathway could partially alleviate the degenerative phenotype afterDdrgk1 conditional knockout in vitro (Figure S10). These findings suggested reduced IRE1α protein stability triggered ER stress following Ddrgk1conditional knockout. Current studies believe that the transient and moderate activation of IRE1α/XBP1s can reduce ER stress and mitigate cell damage, playing a positive regulatory role in cell survival [51]. By contrast, persistent ER stress and excessive activation of IRE1α can trigger pathways such as RIDD or JNK, inducing cell apoptosis [52]. Liu et al. found that DDRGK1 can regulate IRE1α stability by interacting with its kinase domain. They also found that loss of DDRGK1 in cancer cells and hematopoietic stem cells can accelerate IRE1α degradation and PERK activation, leading to apoptosis [39]. Additionally, in our previous studies, we found that DDRGK1 deficiency in chondrocytes has been shown to reduce IRE1α UFMylation, increase ubiquitin-mediated IRE1α degradation, decrease XBP1s expression, and activate the PERK/ATF4/CHOP-induced apoptotic pathway [38]. This study further investigated the alteration at both transcriptional and protein levels of IRE1α in discs after Ddrgk1 conditional knockout, while the XBP1 activity in disc cells was inferred by the activation of IRE1α/UPR signaling pathway and our prior work without direct quantification in this study. However, the mechanism of interaction between DDRGK1 and IRE1α remains controversial. Cai et al. reported upregulation of IRE1α and significant activation of the XBP1s pathway following Ddrgk1 knockout in intestinal cells [53]. Furthermore, in B cells, Ddrgk1 knockout has been found to significantly increase the expression and activation levels of IRE1α/XBP1 [54]. SOX9, a member of the SRY-related high-mobility group (HMG) box transcription factor family, is a unique master transcriptional factor expressed in the process of embryonic mesenchymal condensation for cartilage development and bone formation [55, 56]. SOX9 binds to the HMG sites in DNA to activate target genes such as Col2a1 and Acan, which encode important components of the cartilage ECM [57, 58]. Notably, SOX9 is subjected to degradation by the ubiquitin-proteasome system [59, 60]. Egunsola et al. reported UFMylation mediated by DDRGK1 inhibited the ubiquitination and proteasomal degradation of SOX9, thereby enhancing its stability [34]. Our immunohistochemical staining showed a significant decrease of SOX9 expression following Ddrgk1 conditional knockout, which was validated by RT-qPCR and Western blot assays in primary chondrocytes from Ddrgk1cKOmice. Besides, up-regulation of MMP3, MMP9, MMP13, COL1A1 and COL10A1, coupled with down-regulation of Aggrecan and COL2A1, consistently indicated abnormal cartilage catabolism and matrix phenotype disruption after Ddrgk1 conditional knockout (Figure 7, S5, S6 and S8). These findings indicated that the absence of DDRGK1 disrupted the UFMylation of SOX9, thereby diminishing its stability and impairing chondrogenic activity and cell maintenance in vivo. Admittedly, this research had several limitations which warrant further investigation. First, the present work focused mainly on the role of DDRGK1 in nucleus pulposus and cartilage endplate. However, we also observed some changes in the annulus fibrosus after Ddrgk1 conditional knockout that cannot be fully explained based on current findings. Thus, we planned to conduct further investigations into the specific role of DDRGK1 in the annulus fibrosus in future studies. Second, while RNA sequencing revealed transcriptional changes in disc cells, it lacked the spatial resolution to distinguish among the nucleus pulposus, cartilage endplate, and annulus fibrosus. Therefore, we planned to employ spatial transcriptomics to further analyze the intervertebral disc region of Ddrgk1cKO mice. In summary, our work provides detailed insights into intervertebral disc development and degeneration inDdrgk1cKOmice at the histological, transcriptional and protein levels. Conditional knockout ofDdrgk1 induces ER stress, accelerates ECM degradation and triggers apoptosis in nucleus pulposus and cartilage endplate cells, thereby profoundly exacerbating disc degeneration in both early and late postnatal phase. Collectively, this study underscores the pivotal role of DDRGK1 in preserving the normal cellularity and structure of intervertebral discs and offers a theoretical foundation for DDRGK1-targeting in treating degenerative disc diseases. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 (DOCX 3.58 MB) Supplementary Material 1 (DOCX 3.58 MB)