Effect Of Endoplasmic Reticulum Stress On Chondrogenesis Differentiation Of Mesenchymal Stem Cells By In Vitro Expansion

OBJECTIVE:Mesenchymal stem cells(MSCs)are promising candidates for cell therapeutic approaches in regenerative medicine because of their multipotentiality.MSCs can be harvested from many tissues,including bone marrow and adipose tissue.To obtain sufficient cells for transplantation in clinical treatment,in vitro expansion is always a necessary step for MSCs derived from bone marrow.However,long-or short-term in vitro cultures lead to replicative senescence and impaired multipotency of stem cells.Using a hydrogel system in our previous study,we found decay in telomerase activity and changes in chromosomal anomaly in bone marrow-derived MSCs(BMSCs)at passage 3(P3)compared to freshly isolated BMSCs(PO),thereby resulting in decreased chondrogenic potential of stem cells and a lower therapeutic effect in cartilage repair.However,the precise mechanism underlying the impaired pluripotency of stem cells by in vitro expansion remains unknown.Preliminary analysis based on the comparative profiling of P3 BMSCs vs.PO BMSCs in our previous study showed that endoplasmic reticulum(ER)stress and unfolded protein response(UPR)signaling were significantly different,demonstrating their association with in vitro culture.In this study,we investigated the effect of ER stress and changes in the UPR on the chondrogenic potential of BMSCs in vitro and BMSCs-based cartilage regeneration in vjvo using the ER stress inhibitor(4-phenylbutyrate(4-PBA))in PO BMSCs and the ER stress inducer(tunicamycin(TM))in P3 BMSCs.METHODS:Sixty male New Zealand white rabbits weighing 2.5-3 kg and aged 2 months were used.After the rabbits were anaesthetized with pentobarbital,a 16 G needle was used to puncture the bilateral femurs and tibias,and approximately 20 ml of bone marrow was aspirated.BMSCs were harvested using a bone marrow mononuclear cell isolation kit.BMSCs were separated into four groups as follows:(1)the PO group:fresh BMSCs were isolated from the bone marrow of untreated rabbits.(2)The PO+4-PBA group:fresh BMSCs were isolated from the bone marrow of rabbits which had been treated with 0.25 g/kg of 4-PBA by oral gavage once a day for 20 days.The dosage of 4-PBA was based on a published report.(3)The P3 group:PO BMSCs were cultured and expanded up to passage 3 in vitro.(4)P3+TM group:P3 BMSCs were cultured with 0.25 ?g/ml of TM for 48 h,as previously reported.Total RNA was extracted from the original PO and P3 samples using TRIzol reagent(Invitrogen).Microarray analysis was performed on an Agilent Array platform by Shanghai KangChen Biotech in three replicates.A P<0.05 and an absolute log base 2 fold changes greater than 1 were considered as the criteria for differentially expressed gene selection.KEGG pathway analysis was applied to determine key signaling pathways and relationships between these differentially expressed genes.The resulting data were log base 2 transformed and subjected to further analysis by hierarchical clustering with average linkage.Then,BMSCs were seeded in a collagen solution at a density of 1 ×107 cells/ml.The mixture of BMSCs and collagen solution was gelated and cultured in chondrogenic medium for 7,14 or 21 days.After 7,14,21 days,the samples in all groups were taken out for analysis.Quantitative real-time PCR were used to examine the gene expression of cartilage specific gene(ACAN,SOX9,COL2A1,COL1A1)and UPR-associated markers(ATF4,ATF6,XBP1).Western blot and Immunofluorescence were applied to detect the protein levels of UPR-associated markers and type ?collagen.DNA content and MTT assay were used to evaluate the proliferation of BMSCs.Safranin O staining and DMMB assay were used to detect the glycosaminoglycan(GAG)contents.We also investigated the effect of ER stress and changes in the UPR on the BMSCs-based cartilage regeneration in vivo.Both knees of the rabbits were subjected to the operation.After general anesthesia,a lateral parapatellar approach was used to expose the knee and a chondral-only defect of 4.0 mm in diameter was created in the medial area of each patellar groove.Then,collagen hydrogel loaded with BMSCs was injected into the defective area.The defects were filled as follows:(1)collagen hydrogel with PO BMSCs(PO group,n=30 knees);(2)collagen hydrogel with P0 BMSCs+4-PBA,which from rabbits were fed after 20 days of administration with 4-PBA(P0+4-PBA group,n=30 knees);(3)collagen hydrogel with P3 BMSCs(P3 group,n=30 knees);and(4)collagen hydrogel with P3 BMSCs+TM,which were treated with TM for 48 h(P3+TM group,n=30 knees).Rabbits were euthanized at 4,12,and 26 weeks after surgery and the repaired cartilage was used for further analysis.The gross morphological grading,mechanical evaluation,Histological and immunohistochemical examination were performed to evaluate cartilage regeneration of repaired tissues.RESULTS:The microarray analysis showed there were 1143 upregulated and 3181 downregulated transcripts in the P3 BMSCs compared with the PO BMSCs.By pathway enrichment analyses,we found differentially expressed genes(DEGs)involved in several pathways,including focal adhesion,PI3K-Akt,cell cycle,regulation of actin cytoskeleton,MAPK,UPR,toll-like receptor,and HIF-1.Based on hierarchical clustering analysis,we found that most genes involved in UPR were downregulated in P3 BMSCs.To validate gene expression levels in microarray analysis,several established UPR-associated markers were assessed by quantitative RT-PCR analysis.Three UPR sensors-ATF4,XBP1,and ATF6-were significantly decreased in P3 BMSCs compared to PO BMSCs(P<0.05),which verified the microarray analysis.The mRNA expression of ATF4,ATF6 and XBP1 in the UPR pathway was particularly high at the initial time and almost gradually decreased with time in all groups.The gene expression of ATF4,ATF6 and XBP1 was markedly lower in P3 BMSCs than in PO BMSCs at each time point(P<0.05).After treatment with 4-PBA in PO BMSCs,the expression of the three genes was significantly downregulated(P<0.05).In contrast,the expression of ATF4,ATF6 and XBP1 was significantly increased in P3 BMSCs after treatment with TM(P<0.05).Western blotting analysis confirmed that protein expression of ATF6,ATF4 and XBP1 was correspondingly reduced in P0+4-PBA compared with P0 and increased in P3+TM compared with P3.Immunofluorescence staining showed that there were more positive stain areas for ATF4,ATF6,and XBP1 in the P0 and P3+TM groups than the P0+4-PBA and P3 groups,which is in accordance with the results of quantitative RT-PCR and Western blotting.These findings suggested that the chemical chaperone ER stress inhibitor 4-PBA successfully suppressed UPR in PO BMSCs,and the ER stress inducer TM effectively induced ER stress and UPR in P3 BMSCs.To further uncover the effect of ER stress and UPR variation on chondrogenic differentiation in BMSCs,we assessed the mRNA expression level of cartilage-specific markers,including ACAN,SOX9 and COL2A1,by quantitative RT-PCR analysis.We found that the expression of cartilage-specific genes was increased in a time-dependent manner in all the groups.In comparison,expression levels of the cartilage specific markers were significantly lower in P3 BMSCs than in PO BMSCs(P<0.05).Moreover,4-PBA administration significantly downregulated the expression of the cartilage markers in PO BMSCs(P<0.05),while TM treatment promoted the gene expression(P<0.05).As shown in the Western blotting analysis,protein expression of COL2A1,which is the hyaline cartilage marker,was decreased significantly in the PO+4-PBA group compared with the PO group at different time points(P<0.05).Meanwhile,its expression was significantly elevated in the P3+TM group compared with the P3 group(P<0.05),in agreement with the quantitative RT-PCR findings.This was confirmed by inmunofluorescence staining of COL2A1 and semi-quantitative analysis measured after 21 days.P0 BMSCs showed intense positive staining but P3 BMSCs exhibited almost negative staining.4-PBA administration resulted in slightly positive staining,which was much weaker than P0 BMSCs.TM led to enhanced positive staining in P3 BMSCs.Taken together,these data suggested that ER stress and UPR suppression impaired the chondrogenic differentiation ability of P0 BMSCs,while ER stress and UPR activation promoted chondrogenic differentiation of P3 BMSCs.GAG content was lower in P0+4-PBA compared with PO(P<0.05)but was higher in P3+TM vs.P3 BMSCs(P<0.05).This was also confirmed by safranin-0 staining,with more positive staining in PO BMSCs than P0+4-PBA and in P3+TM than P3,indicating that 4-PBA was associated with a reduction in UPR,inhibiting the synthesis of chondrogenic glycosaminoglycan in PO BMSCs,and TM-mediated UPR induction accelerated the chondrocytic matrix in P3 BMSCs.Taken together,these results suggested that ER stress and changes in the UPR influenced chondrogenic differentiation in BMSCs cultured in vitro.To further examine the effect of ER stress and UPR change on BMSCs-based therapy for cartilage defects,we created a cartilage defect model.We examined the repair efficacy of the cartilage defect by macroscopic assessment,biomechanical evaluation,histological and type ? collagen immunohistochemical staining.At each time point,the PO BMSCs therapy group showed a better therapeutic effect than the others.In contrast,the P3 groups had lower scores than the other groups.4-PBA delays cartilage restoration while the ER stress enhancer TM accelerates healing.CONCLUSION:In conclusion,we showed that the decline in chondrogenic potential of stem cells after in vitro culture and expansion may be mediated by ER stress and the UPR pathway.Activation of ER stress and UPR promoted chondrogenesis in expanded BMSCs,favorable for therapy of cartilage defects.In addition,their inhibition weakened the chondrogenic differentiation of freshly isolated BMSCs and was unfavorable to cartilage regeneration.