https://orcid.org/0000-0001-6604-9972
Abstract
Mandibuloacral dysplasia type A (MADA) is a rare genetic progeroid syndrome associated with lamin A/C (LMNA) mutations. Pathogenic mutations of LMNA result in nuclear structural abnormalities, mesenchymal tissue damage and progeria phenotypes. However, it remains elusive how LMNA mutations cause mesenchymal-derived cell senescence and disease development. Here, we established an in vitro senescence model using induced pluripotent stem cell-derived mesenchymal stem cells (iMSCs) from MADA patients with homozygous LMNA p.R527C mutation. When expanded to passage 13 in vitro, R527C iMSCs exhibited marked senescence and attenuation of stemness potential, accompanied by immunophenotypic changes. Transcriptome and proteome analysis revealed that cell cycle, DNA replication, cell adhesion and inflammation might play important roles in senescence. In-depth evaluation of changes in extracellular vesicle (EV) derived iMSCs during senescence revealed that R527C iMSC-EVs could promote surrounding cell senescence by carrying pro-senescence microRNAs (miRNAs), including a novel miRNA called miR-311, which can serve as a new indicator for detecting chronic and acute mesenchymal stem cell (MSC) senescence and play a role in promoting senescence. Overall, this study advanced our understanding of the impact of LMNA mutations on MSC senescence and provided novel insights into MADA therapy as well as the link between chronic inflammation and aging development.
Introduction
Lamin A/C (LMNA) is a widely expressed protein component of the nuclear inner membrane that interacts with chromatin to regulate gene expression (1,2). Mutations in the LMNA gene cause laminopathies, resulting in tissue and organ damage of various interstitial origins (3,4). How LMNA regulates cell senescence and gene expression remains unclear. Mandibuloacral dysplasia type A (MADA) is a rare genetic progeroid syndrome associated with missense mutations in LMNA (R527C, R527H, and A529V), characterized by type A lipodystrophy, skeletal and craniofacial dysmorphic, metabolic abnormalities and varying degrees of accelerated aging (5–8). Because of the limited number of patient samples, current research remains insufficient for understanding the specific mechanism of the disease and providing a cure. The mechanism underlying the accelerated cellular aging and disease development resulting from LMNA gene mutation requires further exploration and elucidation.
Mesenchymal stem cells (MSCs) are a key therapeutic tool in regenerative medicine and tissue engineering (9,10). However, during in vivo or in vitro culture, the functional degeneration of MSCs occurs with donor age and disease status, causing a gradual loss of their stemness and limiting their application. MSCs undergo replicative senescence during long-term culture in vitro, which is characterized by a change in cell morphology from spindle-like to flattened, decreased proliferation ability, increased mitochondrial and lysosome density and altered mitochondrial membrane potential (11–13). Induced pluripotent stem cells (iPSCs) restore terminally differentiated somatic cells to a totipotent state by introducing specific transcription factors. Because iPSCs have the exact genetic background of the donor and possess pluripotency, using iPSC/iPSC-derived stem cells to establish disease models will help further study the mechanism of disease occurrence and development, providing an excellent platform for screening effective therapeutic methods (14,15). However, effectively delaying or reversing the senescence of iPSC-derived MSCs (iMSCs) in vitro remains the key for advance their applications.
MicroRNAs (miRNAs), a special class of RNA consisting of about 22 nucleotides, are regarded as important contributors to gene regulation, primarily modulating post-transcriptional processes by affecting the translation and stability of targeted mRNAs (16,17). Extracellular vesicles (EVs) are heterogeneous vesicles secreted by cells under stimuli such as differentiation and senescence, with diameters ranging from 30 nm to 5 μm. EVs are considered important carriers of cell-to-cell communication by transporting special proteins, nucleic acids and lipids (18,19). MSCs can interfere with the senescence process of neighboring cells by secreting and releasing EVs into the extracellular environment, and their secretory properties change with senescence. Using stem cell-derived EVs and interfering with substances carried by them to modify their effects has become a new approach for diseases.
As the affected tissues in laminopathies are mainly derived from mesenchyme, we have chosen MSCs as the relevant cell model to investigate the impact of LMNA mutations. This study aims to investigate the impact of LMNA R527C mutation on MSC senescence and further explore the influence of MSC-derived EVs on their senescence. To achieve this goal, we focused on changes in cellular mRNA, cellular protein and EV small RNA during the replicative senescence of LMNA R527C iMSCs to explore the molecular mechanism of replicative senescence and the role of EVs in cell senescence. The reasons for the functional changes in LMNA R527C MSCs have been discovered for the first time, and a novel miRNA (called miR-311) may be used as a new target to detect and regulate MSC senescence.
Results
Cell phenotype of LMNA R527C iMSCs during in vitro replicative senescence
R527C iMSCs underwent replicative senescence after prolonged in vitro culture from passage 6 (P6) to P13 and exhibited changes in morphology as well as specific functional properties. After prolonged culture, the cell area increased, morphology flattened and the cells became longer. In addition, cell proliferation decreased (Fig. 1A and B). Increased expression of p16INK4A and p21 is widely considered a marker of cell senescence. Compared with R527C iMSCs at the early senescence (P6), the expression of p21 and p16INK4A was upregulated significantly at the late senescence (P13) (Fig. 1C and D). The mRNA expression of stemness genes (Nanog and SOX2) was significantly downregulated with prolonged culture (Fig. 1E). In the examination of differentiation potential, we found that the osteocyte and chondrocyte differentiation potentials of P13 cells were weaker than those of the P6 group (Fig. 1F). We also detected the density of intracellular lysosomes and mitochondria, and found that both were significantly increased in the P13 group (Fig. 1G and H). In the detection of mitochondrial membrane potential, we observed that the P13 groups displayed a higher proportion of JC-1 monomers with green fluorescence and a lower proportion of JC-1 aggregates with red fluorescence, indicating that replicative senescence had reduced mitochondrial membrane potential (Fig. 1I). Together, R527C iMSCs exhibited obvious hallmarks of cell senescence and a decreased potential for stemness during in vitro culture.
Changes in senescent markers and stemness properties of R527C iMSCs during serial passages. (A) Representative images of R527C iMSC morphology at P6 and P13. Scale bar: 500 μm. (B) Proliferation of R527C iMSCs at P6 and P13. (C) Protein expression levels of cell cycle markers of R527C iMSCs at P6 and P13. GAPDH serves as an internal control. The mRNA levels of cell cycle genes (D) and stemness genes (E) of R527C iMSCs at different passages. (F) Osteogenic, adipogenic and chondrogenic differentiation of R527C iMSCs at P6 and P13. (G) The lysosomal density of R527C iMSCs at P6 and P13. Scale bar: 200 μm. (H) The mitochondrial density of iMSC at P6 and P13. Scale bar: 200 μm. (I) The mitochondrial membrane potential of iMSC at P6 and P13. Scale bar: 200 μm. (J) Staining activity analysis of osteogenic and adipogenic differentiation. (K) Fluorescence density analysis of lysosomal staining, mitochondrial staining and JC-1 staining. n = 3, **P < 0.01; ***P < 0.001 by Student’s t-test.
Transcriptome of LMNA R527C iMSCs during in vitro replicative senescence
To understand the mechanism behind the replicative senescence of R527C iMSCs, we performed transcriptomic analysis on P6 and P13 cells. A total of 2894 differentially expressed genes (DEGs) were identified using screening criteria of P-value < 0.05 and |log2 FC| > 1 (fold change), with 1813 genes upregulated and 1081 downregulated (Fig. 2A). Interleukins were predominantly upregulated in high passage iMSCs, and the top three candidates (interleukin 11 (IL11), IL1RAPL1 and IL4I1) were associated with a pro-inflammatory response. Similarly, platelet-derived growth factors (PDGFs), transforming growth factors (TGFs), vascular endothelial growth factors (VEGFs), matrix metallopeptidases (MMPs) and collagen-related genes were mainly upregulated in in vitro culture. The genes for chondroitin sulfate proteoglycans (CSPGs), laminins and fibronectin were upregulated, whereas integrins showed to be differentially expressed in in vitro culture. Together, these changes in gene expression might reflect the more pro-inflammatory phenotype of late senescent R527C iMSCs. The cyclin-dependent kinase (CDK)-related genes were predominantly downregulated in P13 iMSCs. CDKN2A (encoding p16INK4A, a senescence biomarker) was upregulated in P13 iMSCs, whereas CDKN2C (encoding p18) was downregulated. Tumor necrosis factor (TNF)- and bone morphogenetic protein (BMP)-related genes were mainly upregulated in the P13 groups, as well as the insulin-like growth factors (IGFs) and human leukocyte antigens (HLAs) family. The sirtuin- and homeobox B (HOXB)-related genes were differentially expressed, and differentially expressed CD markers were also identified as potential surface features of the cells (Fig. 2B and C). Collectively, the data suggested that the immunomodulatory properties of R527C iMSCs shifted toward a pro-inflammatory phenotype during in vitro culture.
Transcriptome of R527C iMSCs during in vitro culture. (A) Volcano plot exhibited DEGs in P6 and P13 iMSCs from the mRNA-sequencing analysis. (B) Genes related to interleukins, growth factors, integrin and matrix genes. (C) Genes related to CDKs, TNFs, BMPs, IGFs, HLAs, Sirts, HOXBs and CD markers. The values were the log2 ratios of P13 cells to P6 cells, with a value of 1 indicating a 2-fold increase. Negative values indicate higher gene expression in the P6 group, whereas positive values indicate higher gene expression in the P13 group. n = 3.
Gene ontology (GO) enrichment analysis provided information on biological processes (BP), cellular components (CC) and molecular functions (MF) of the obtained DEGs in this study. The dominant categories for BP were nuclear division, organelle fission and mitotic nuclear division. For CC, extracellular matrix, chromosome centromeric region and condensed chromosome centromeric region were the most abundant entries. For MF, extracellular matrix structural constituent conferring tensile strength, glycosaminoglycan binding and receptor regulator activity were the most abundant catalogs (Fig. 3A). In addition, we observed a significant decrease in the representation of nuclear, chromosome, microtubule and DNA replication gene sets among P13 cells based on the top six downregulated GO terms (Fig. 3B). Next, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis provided information on regulated biological pathways. The top five enriched pathways were DNA replication, cell cycle, protein digestion and absorption, axon guidance and extracellular matrix (ECM)–receptor interaction (Fig. 3C). Functional analysis revealed that DEGs were associated with key functions, including nuclear division, chromosome segregation, DNA replication and cell adhesion.
Functional classification of GO enrichment and KEGG pathway clusters in the R527C iMSC transcriptome. (A) GO analysis for DEGs between P6 and P13 iMSCs for biological processes, cellular components and molecular functions. (B) The top six downregulated GO terms. (C) KEGG analysis for DEGs between P6 and P13 iMSCs. n = 3.
Proteome of LMNA R527C iMSCs during in vitro replicative senescence
To complement the transcriptional analyses, we also examined the protein-level response to prolong culture. Figure 4A showed a total of 363 differentially expressed proteins (DEPs, 112 upregulated and 251 downregulated) identified with screening criteria of P-value < 0.05 and |log2FC| > 1. GO and KEGG pathway enrichment analyses were performed to understand the biological functions of R527C iMSC DEPs during replicative senescence. The BP of GO enrichment included DNA replication and nucleic acid metabolic process, the CC mainly involved the nucleus and extracellular region, and the MF mainly involved DNA binding, endopeptidase inhibitor activity and ion binding (Fig. 4B). In addition, we also noted that the top three downregulated GO terms indicated strongly lower representation of DNA replication and metabolic, nuclear, extracellular region and endopeptidase inhibitor activity gene sets in P13 cells (Fig. 4C). KEGG pathway enrichment analysis indicated that the enriched pathways involved in downregulated DEPs included phosphatidylinositol 3 kinases (PI3K)—protein kinase B (Akt) signaling pathway and nuclear factor κB (NF-κB) signaling pathway (Fig. 4D). Functional analysis revealed that DEPs were associated with key functions such as DNA replication, nucleic acid metabolism, immunity and inflammation.
Proteomic analysis of R527C iMSCs during in vitro culture. (A) Volcano plot exhibited DEPs in P6 and P13 iMSCs from proteome analysis. (B) GO analysis for DEGs between P6 and P13 iMSCs for biological processes, cellular components and molecular functions. (C) The top three downregulated GO terms. (D) KEGG analysis for DEPs between P6 and P13 iMSCs. (E) Venn diagram showed overlap and variation of DEGs and DEPs in P6 and P13 iMSCs. (F) The correlation between DEGs and DEPs in P6 and P13 iMSCs. (G) Correlation functional analysis for DEGs and DEPs in P6 and P13 iMSCs. n = 3.
A total of 27 723 genes and 3907 proteins were identified in transcriptomic and proteomic analyses. A Venn diagram of DEGs and DEPs in R527C iMSCs from prolonged culture revealed a total of 61 DEGs (Fig. 4E). Correlation analysis of transcriptome and proteome expression levels in R527C iMSCs during prolonged culture revealed a weak linear relationship between transcriptome and proteome expression, as indicated by the Pearson’s correlation coefficient of 0.16 (Fig. 4F). Subsequently, GO and KEGG pathway enrichment analyses were performed on the correlated DEGs and DEPs. The BP of GO enrichment included chromatin organization, macromolecule metabolic process, ossification, regulation of the apoptotic process and small molecule biosynthetic process. The CC mainly involved the extracellular region, intermediate filament and nucleus. The MF mainly involved DNA binding, endopeptidase inhibitor activity and ion binding. KEGG pathway enrichment indicated that the significant pathways included cell cycle, DNA replication, PI3K-Akt signaling pathway, nucleotide oligomerization domain (NOD)-like receptor signaling pathway, NF-κB signaling pathway and focal adhesion (Fig. 4G). Combined results indicated that processes such as cell cycle, DNA replication, inflammation and cell adhesion might play important roles in the replicative senescence of R527C iMSCs.
Characteristics of LMNA R527C iMSC-EVs
EVs play a crucial role in cell-to-cell communication and precisely regulate the developmental processes of receptor cell senescence and inflammation. To further investigate the impact of LMNA R527C mutation on MSC senescence, we compared the characteristics, functions and carried miRNAs between R527C iMSC-EVs at early and late senescence stages. Transmission electron microscopy (TEM) and nanoparticle tracking analysis (NTA) revealed that both R527C iMSC-P6 and -P13 cells produced a heterogeneous population of EVs with diameters ranging from ~50 to 200 nm, exhibiting typical morphological features (Fig. 5A and B). Purified P6-EV and P13-EV had comparable particle counts (3.73 ± 0.2 × 108 and 6.05 ± 0.3 × 108, respectively). Western blots showed that both P6-EV and P13-EV tested positive for established exosome markers (Alix, Tsg101, CD81 and CD63) and lacked the cellular marker (glyceraldehyde-3-phosphate dehydrogenase, GAPDH) (Fig. 5C).
Characterization of iMSC-derived EVs. (A) TEM images of the different-sized EVs. Scale bar: 200 nm. (B) NTA analysis of the size and concentration of EVs. (C) Western blot analysis of the surface marker of EVs. n = 3.
Effect of EVs on senescence characteristics of LMNA R527C iMSCs
To explore the functional differences between EVs derived from P6 and P13 cells, two types of EV treatments were tested for their effects on R527C iMSC senescence, including cell proliferation (Fig. 6A and B), senescence-associated-β-galactosidase (SA-β-gal) staining (Fig. 6C), mitochondrial density (Fig. 6D) and mitochondrial membrane potential changes (Fig. 6E). Here, we considered P6 as the early stage of senescence and P12 as the late stage. At the early stage of R527C iMSCs, P6-EV had little effect on cell proliferation compared with the control group, whereas P13-EV significantly attenuated the cell proliferation (Fig. 6A), and P13-EV significantly increased the proportion of SA-β-gal staining positive cells (Fig. 6C). At the late stage of R527C iMSCs, P6-EV significantly promoted the proliferation of R527C iMSCs compared with the control (Fig. 6B). Then, we analyzed the effect of two EV treatments on mitochondrial density and found that P6-EV incubation significantly reduced the mitochondrial fluorescence intensity of R527C iMSCs in both early and late stage, whereas P13-EV incubation significantly enhanced the mitochondrial fluorescence intensity during early senescence (Fig. 6D). Decreased mitochondrial membrane potential is a hallmark of cellular oxidative stress damage. In the early stage, P6-EV treatment slightly increased red fluorescence, whereas P13-EV treatment decreased red fluorescence and increased green fluorescence, indicating that P13-EV incubation reduced the mitochondrial membrane potential in early senescence iMSCs. In the late stage, P6-EV incubation significantly improved R527C iMSC mitochondrial membrane potential (Fig. 6E). Together, the effects of P6-EV and P13-EV on R527C iMSC senescence were inconsistent.
Alterations of senescent characteristics in R527C iMSCs during EV incubation. (A) Cell counts of iMSCs in early senescence after EV incubation. (B) Cell counts of iMSCs in late senescence after EV incubation. *P < 0.05; **P < 0.01; ***P < 0.001 by Student’s t-test. (C) SA-β-gal activity of early senescent iMSCs after 2 days of EV incubation. Scale bar: 1000 μm. (D) Mitochondrial density of early and late senescent iMSCs after 2 days of EV incubation. Scale bar: 200 μm. (E) Mitochondrial membrane potential of early and late senescent iMSCs after 2 days of EV incubation. Scale bar: 200 μm. (F) Fluorescence density analysis of mitochondrial staining. (G) Fluorescence density analysis of JC-1 staining. The concentration used for EV incubation was 2 × 107 particles/mL. P < 0.05 by one-way ANOVA with Tukey’s post hoc test. n = 3.
Small RNA sequencing of LMNA R527C iMSC-EVs
To further analyze the functional differences between P6- and P13-EVs, we conducted a small RNA sequencing analysis on EVs. A total of 935 miRNAs (925 known miRNAs and 10 novel miRNAs) were identified in EVs from the P6 and P13 cells. The main content was composed of known miRNAs whereas the novel ones were present at a relatively lower level. The differentially expressed miRNAs (DEMs) were screened out from the P6- and P13-EVs using a P-value < 0.05 and |log2FC| > 1 as screening criteria (Fig. 7A). A total of 15 DEMs (six upregulated and nine downregulated) were identified, as shown in the miRNA expression pattern cluster analysis heatmap (Fig. 7B), which may reflect functional differences between P6- and P13-EVs.
Identification of differentially expressed miRNAs between P6 and P13 iMSC-derived EVs. (A) Volcano plot showed DEMs in P6 and P13 iMSC-derived EVs based on small RNA sequencing. (B) Hierarchical cluster analysis of these DEMs. (C) Secondary structure of miR-311. (D) Expression of miR-311 in P6 and P13 R527C iMSC-EVs. (E) Expression of miR-311 in R527C iMSCs at different passages. P < 0.05 by one-way ANOVA with Tukey’s post hoc test. (F) Expression of miR-311 in WT-iMSCs and primary hMSCs at different passages. (G) Expression of miR-311 in doxorubicin-induced WT-iMSC senescence. n = 3, *P < 0.05. ***P < 0.001 by Student’s t-test.
Using miREvo and mirdeep2 software to predict new miRNA and analyze the expression differences, we found that the secondary structure of miR-311 precursor is shown in Fig. 7C (with a mature sequence of GUGGGGUGUUAGGUCAUUC), and the expression of miR-311 was differentially in P6- and P13-EVs. The quantitative real-time PCR (qPCR) results revealed a significant upregulation of miR-311 in P13-EV compared with P6-EV (Fig. 7D), and the expression of miR-311 in cells increased with passages (Fig. 7E), which was observed in both wild type (WT)-iMSCs and primary hMSCs (Fig. 7F). WT-iMSCs (passage 6) were exposed to 100 nm doxorubicin in culture medium for a duration of 2 days, resulting in acute senescence. Then, the cells were collected for the detection of miR-311 expression. The miR-311 expression was significantly increased in the doxorubicin-induced acute senescence treatment group (Fig. 7G). Together, miR-311 could be regarded as a senescence-associated molecular marker of MSCs, reflecting both chronic and acute degrees of cell senescence.
Effect of miR-311 on senescence characteristics of LMNA R527C iMSCs
The upregulated miRNAs (miR-7704, miR-133b, miR-106a, miR-183-5p, miR-363-3p) in P13-EVs have been shown to exhibit expression changes during cell senescence and participate in the senescence process. However, the role of novel miRNA (miR-311) is unknown. Next, we analyzed the function of miR-311 in R527C iMSC senescence by transiently transfecting miR-311 mimic and inhibitor into R527C iMSCs to achieve overexpression or silencing (Fig. 8A). Silencing miR-311 promoted cell proliferation in late-stage iMSCs (Fig. 8B). Detection of senescence-related proteins in late-stage iMSCs also revealed that overexpression of miR-311 increased the protein levels of p16INK4A and p21, whereas silencing reduced their levels (Fig. 8C). In iMSCs, overexpression of miR-311 increased mitochondrial fluorescence intensity whereas silencing decreased mitochondrial intensity (Fig. 8D). In early senescent iMSCs, overexpression of miR-311 decreased mitochondrial membrane potential. Silencing miR-311 enhanced red fluorescence and reduced green fluorescence, indicating that silencing miR-311 could improve the mitochondrial membrane potential (Fig. 8E). The data suggested that miR-311 might be involved in promoting R527C iMSC senescence.
Effect of miR-311 on R527C iMSCs. (A) Expression of miR-311 in R527C iMSCs after miR-311 mimic and inhibitor transfection. (B) Cell counts of iMSCs in early senescence after miR-311 transfection. ***P < 0.001 by Student’s t-test. (C) Protein levels of cell cycle genes in late senescent iMSCs after transfection with miR-311 mimic and inhibitor. β-Actin serves as an internal control. (D) Mitochondrial density in early and late senescent iMSCs after transfection with miR-311 mimic and inhibitor. Scale bar: 200 μm. (E) Mitochondrial membrane potential of iMSCs in early senescence after transfection with miR-311 mimic and inhibitor. Scale bar: 200 μm. (F) Fluorescence density analysis of mitochondrial staining. (G) Fluorescence density analysis of JC-1 staining. P < 0.05 by one-way ANOVA with Tukey’s post hoc test. n = 3.
Discussion
Laminopathies represent a group of genetic disorders with different clinical phenotypes caused by mutations in the LMNA gene, including adipose tissue disorders (mandibuloacral dysplasia) (6), cardiac muscle disorders (dilated cardiomyopathy) (20), neurological diseases (Charcot–Marie–Tooth disease) (21), skeletal muscle disorders (congenital muscular dystrophy) (22), skin diseases (restrictive dermopathy) (23) and progeria (24,25). LMNA is involved in cell proliferation and differentiation by regulating chromatin organization and transcription. LMNA mutations have specific effects on MSC differentiation and proliferation (26). The previous results showed that R527C iMSCs exhibited accelerated senescence, including changes in cell morphology, proliferation and differentiation potential, compared with normal iMSCs of the same passages in in vitro expansion (results not shown). After long-term culture in vitro, the heterogeneity of hMSCs increased, leading to cell senescence and a decline in primitive phenotype. During the senescent process, MSCs gradually lost their typical fibroblast-like spindle morphology and became irregular and abnormal, with decreased proliferation and impaired mitochondrial function (27). In this research, we found that replicative senescent R527C iMSCs exhibited senescence-specific features, including enlarged cell morphology, delayed cell proliferation, increased expression of cell cycle-related genes and density of lysosomes and mitochondria, decreased expression of stemness genes and reduced mitochondrial membrane potential. Under similar passages and conditions as human bone marrow (hBM)-MSCs, R527C iMSCs also displayed certain characteristics of cell senescence.
The differentiation potential of R527C MSCs was also reduced by replicative senescence induced through serial culture. Our focus is on the effect of LMNA mutation on MSC differentiation potential from its primary impact on mesenchymal tissue. We found that the adipogenic differentiation potential of R527C iMSCs was slightly affected by replicative senescence, and P6 cells exhibited higher osteogenic and chondrogenic differentiation potentials than P13 cells. The differentiation of MSCs into osteogenic and adipogenic lineages is mutually inhibitory, with inducers of adipocyte differentiation capable of suppressing the formation of osteoblasts. Both P4 and P8 hBM-MSCs were able to differentiate into adipocyte-like cells, but their osteogenic differentiation potential was significantly impaired, suggesting that senescent MSCs might preserve their potential for adipogenesis better than for osteogenesis (28). Reports on the effect of LMNA on adipogenic differentiation have shown controversial results, which may be attributed to differences in the cellular models studied. Mutated LMNA has the potential to alter the balance between osteogenic and adipogenic potential in a specific manner, which may currently only be speculative. Therefore, further studies are required to validate the phenomena observed in this study regarding the influence of senescence on the adipogenic differentiation potential of MSCs.
The stemness characteristics of R527C iMSCs undergo changes in their stemness genes and surface markers with serial passages, which increased the risk of heterogeneity. Replicative senescence may alter the stemness and function of iMSCs. However, the effects of surface-specific markers on the functional properties of MSC during long-term culture remain unclear. In this study, a replicative senescence model was developed by serial culturing MSCs, and the observed senescence characteristics of R527C iMSCs underscored the need to assess and identify genetic profiles.
Transcriptome analysis revealed that the immunophenotype of R527C iMSCs shifted toward a pro-inflammatory state after serial passages, and differential expression of growth factors, integrins and matrix proteins was also observed. Interleukins (IL1RAPL1, IL4I1 and IL11), which were highly expressed in P13 cells, are associated with pro-inflammatory responses. These changes in immunophenotype and immunomodulation suggested that the immunophenotype might play a crucial role in LMNA mutant MSC senescence. In cardiac laminopathies, the clinical manifestations associated with patients carrying pathogenic LMNA mutation were found to correlate with the degree of inflammation, as indicated by the upregulation of pro-inflammatory cytokines. IL-1RA may serve as a biomarker associated with conduction defects and arrhythmic manifestations in LMNA patients exhibiting a dominant cardiac phenotype (29). The expression of IL-1RA and TGF-β2 was also upregulated in patients with striated muscle laminopathies (30). This study also observed the high expression of collagen (COL4A2, COL6A3, COL15A1, etc.) and MMPs (MMP3, MMP1, MMP13, etc.) in R527C iMSC-P13. Alterations in matrix proteins (MMP9, COL1A1 and COL5A1) could potentially impair the osteogenic differentiation properties of MSCs (31). The integrin family can regulate intracellular signaling cascades and cell-to-cell adhesion, thereby promoting cell proliferation and differentiation (32–34). In R527C iMSC-P13, the most significant changes were observed in ITGA8 and ITGB4. Among these changes, ITGA8 was found to regulate multiple cellular activities, such as adhesion and proliferation (35). Considering that replicative senescence could exert greater pressure on cell proliferation, ITGA8 is speculated to be more susceptible to change under the stress of replicative senescence.
We conducted functional analyses of DEGs and DEPs to gain a deeper understanding of the mechanism underlying cell senescence in R527C iMSCs, intending to elucidate the significance of the observed changes. GO analysis of DEGs and DEPs indicated that nuclear- and chromosome-related gene sets were significantly altered in the cellular processes of serially passaged R527C iMSCs. Nuclear changes are a vital feature of cell senescence, involving DNA damage, nuclear structural changes and chromatin rearrangements that affect gene transcription, and the secretion of senescence-associated secretory phenotype (SASP). We observed that the gene sets related to nuclear, chromosome, microtubule and DNA replication were significantly downregulated in P13 cells compared with P6 cells in transcriptomics. Similarly, the gene sets associated with nuclear, DNA replication, nucleic acid metabolism and serine-type endopeptidase inhibitor activity was strongly downregulated in P13 cells compared with P6 cells in proteomics. We speculated that nuclear- and chromosomal-related changes were closely associated with the replicative senescence of R527C iMSCs. It has been reported that aging could lead to increased serine endopeptidase activity in mouse skin (36). In this study, the gene set for serine-type endopeptidase inhibitor activity was found to be downregulated in P13 cells, suggesting that serine endopeptidase may contribute to R527C iMSC senescence. However, further investigation is necessary to draw definitive conclusions.
KEGG analysis showed that the biological functions enriched by DEGs and DEPs were related to cell cycle, inflammation, cell adhesion molecules and the signaling pathways (PI3K-Akt, NF-κB and NOD-like receptor signaling pathways). Cell senescence is a permanent cell cycle arrest in response to mitogenic stimuli, triggering changes in secretory phenotype and resistance to cell death (37). The PI3K-Akt pathway has been found to be involved in cell senescence, and the acceleration of MSC senescence in systemic lupus erythematosus patients was observed with enhanced PI3K-Akt signaling (38). NF-κB activity has been reported to increase in aging and aging-related chronic diseases. In progeria mice, NF-κB was randomly activated in multiple cell types during aging, and inhibition of NF-κB reduced DNA oxidative damage and delayed cell senescence (39). Meanwhile, PI3K-Akt, NF-κB and NOD-like receptor signaling pathways have also been reported to be involved in inflammation (40–42). The cell-derived omics data in this study supported the immunophenotypic transfer of R527C iMSCs during in vitro expansion and demonstrated a unique expression change pattern caused by LMNA mutation, which could be used to reveal the senescence mechanism of R527C iMSCs.
We analyzed the changes in characteristics, functions and carried miRNAs of R527C iMSCs-derived EVs to gain insights into how LMNA mutations affect senescence by altering the secretory phenotype. In response to aging and senescence, cells increased EV production. NTA results indicated that the average particle size and concentration of P13-EV were larger than those of P6-EV, consistent with observations made in BM-MSCs derived from late passage cultures or aged donors. These findings indicated that cell senescence also affected EV production in LMNA-mutated iMSCs. Senescent cells are capable of direct intercellular communication through the secretion of EVs. The functional analysis results showed that P13-EV aggravated the senescence characteristics of early senescence cells, whereas P6-EV alleviated the senescence characteristics of late passage cells, which was similar to the effect of BM-MSC-derived EVs on senescence. Senescent cells can transmit senescence signals by secreting EV transfer regulators, such as miRNAs and proteins, which trigger neighboring cells to respond efficiently and rapidly. Next, we analyzed the changes in small RNA expression profiles between P6- and P13-EVs and identified 15 specifically dysregulated miRNAs, including a predicted novel miRNA (miR-311). MiR-311 was not only highly expressed in P13-EV but also gradually upregulated in cells with the serial passage, indicating its potential as a senescence-related biomarker of MSCs. The functional results indicated that miR-311 promoted the senescence of LMNA-mutated iMSCs. In addition, other upregulated miRNAs (miR-7704, miR-133b, miR-106a, miR-183-5p and miR-363-3p) have been shown to exhibit changes in expression during cell senescence. Specifically, the expression of miR-7704 was found to be upregulated in exosomes derived from human vascular smooth muscle cells undergoing replicative senescence (43). The downregulation of miR-133 in MSC-EVs and serum significantly attenuated renal injury induced by a unilateral ureteral obstruction in elderly rats (44). In a mouse model of myocardial infarction, miR-183-5p in MSC-exosomes inhibited cardiomyocyte senescence by regulating the high mobility group box-1 (HMGB1)/extracellular regulated protein kinases (ERK) pathway (45). In rat BM-MSCs, miR-363-3p inhibited osteogenic differentiation and promoted adipogenic differentiation and cell senescence by targeting TNF receptor-associated factor 3 (TRAF3) (46). In this regard, we hypothesized that senescent R527C iMSCs could accelerate surrounding cell senescence by secreting EVs containing high levels of these pro-senescent miRNAs. Further investigations are required to establish conclusive evidence.
Materials and Methods
Cell culture and transfection
The LMNA R527C iMSC line and WT-iMSC line used in this study were derived from peripheral blood cells of patients with the homozygous LMNA p.R527C mutation (gifted by Professor Wei Shu, Guangxi Medical University) and healthy donors, respectively. The blood samples were collected from patients and healthy donors with written informed consent. For patients, the consent was obtained from their parents and signed by themselves, which included permission for the use of patient and healthy donor data for publication. The relevant experiment was approved by the medical ethics committee (Shenzhen Medical Ethics Committee, Shenzhen Luohu People’s Hospital Medical Ethics Committee, China). IPS cells were generated from peripheral blood cells using the Cytotune-IPS 2.0 Sendai reprogramming kit (Invitrogen, USA) and differentiated into MSCs according to our previous method (47). The iPSCs and iMSCs utilized have been well validated in our previous study (A. Padhiar, X. Yang, Z. Li, J. Liao, I. Ali, W. Shu, A. Chishti, L. He, G. Alam, A. Faqeer, Y. Zhou, S. Zhang, T. Wang, T. Liu, M. Zhou, G. Wang, X. Zou and G. Zhou, manuscript in preparation, preprint at https://doi.org/10.1101/2202.08.31.504639). The primary human MSC line was obtained from Nuwacell (Cat #RC02003, China). These MSC lines were maintained in MSC basal medium (Dakewe, China) containing a serum-free supplement (EliteGro-Advanced; Dakewe) at 37°C and 5% CO2 in an incubator.
Transient transfection of miRNA mimics or inhibitors (100 nm) was performed on LMNA R527C iMSCs using Lipofectamine 3000 reagent (Life Technologies, USA). The sequences of miRNA mimics and inhibitors are listed in Supplementary Material, Table S1. Cellular morphological changes, including alterations in overall morphology and cell size, were assessed under an inverted microscope.
Multipotential differentiation of LMNA R527C iMSCs
Osteogenic differentiation
The iMSCs were grown to 80–90% confluency before replacing the MSC basal medium with an osteogenic differentiation medium. The differentiation lasted for 21 days and required a medium change every 2 days. The differentiation was visualized by Alizarin Red S staining (MSC osteo-staining kit; Vivacell, China) under a microscope.
Adipogenic differentiation
The iMSCs were grown to 80–90% confluency before replacing the MSC basal medium with adipogenic differentiation medium. The differentiation lasted for 14 days and required medium change every 3 days. The differentiation was visualized by Oil Red O staining (MSC adipo-staining kit; Vivacell) under a microscope.
Chondrogenic differentiation
The iMSCs were grown to 80–90% confluency before replacing the MSC basal medium with a chondrogenic differentiation medium. The differentiation lasted for 21 days and required medium change every 3 days. The differentiation was visualized by Alcian blue 8GX staining (MSC chondro-staining kit; Vivacell) under a microscope.
Cell proliferation
R527C iMSCs were seeded onto a culture plate, and cell proliferation was detected using the WST-1 cell proliferation and cytotoxicity assay kit (Beyotime Biotechnology, China). The culture medium was incubated at 37°C for 2 h after adding 0.1 vol of water-soluble tetrazolium reagent, and the absorbance was measured at a wavelength of 450 nm using a microplate reader (Bio-Tek, USA).
Quantitative real-time PCR
For qPCR analysis of genes and miR-311 expression, sample RNA extraction was performed using a miRNA kit (for cell mRNA and miRNA; Omega Scientific, USA) and exoRNeasy serum/plasma starter kit (for EV-RNA; QIAGEN, Germany). The cDNA synthesis was carried out using the PrimeScript RT reagent kit with gDNA eraser (Takara, Japan) or Mir-X miRNA first-strand synthesis kit (Clontech, USA). These genes and miR-311 expression were quantified using the LightCycler 480 SYBR Green I Master kit on the LightCycler 480 Detection System (Roche, Switzerland). Each experiment was performed in triplicate. Data were calculated using the 2−△△Ct method, and GAPDH and U6 were detected as internal controls. The primer sequences used for qPCR are listed in Supplementary Material, Table S2.
Western blot
For western blotting, R527C iMSCs were lysed with radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime Biotechnology) containing 1% protease inhibitor (Sigma-Aldrich, USA). The protein samples were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene difluoride membranes (Merck Millipore, Germany), which were subsequently blocked with TBST buffer containing 5% bovine albumin. Blots were incubated with appropriate primary and corresponding secondary antibodies, followed by color development using a super-enhanced chemiluminescence detection reagent (Yeasen Biotechnology, China). Primary antibodies against the following proteins were used in this study: p16, Proteintech, 10883-1-AP; p21, Proteintech, 10355-1-AP; GAPDH, Proteintech, 60004-1-lg; β-actin, Proteintech, 81115-1-RR; Alix, Abcam, ab275377; Tsg101, Abcam, ab125011; CD63, Abcam, ab134045; CD81, Cell Signaling Technology, 56039. GAPDH and β-actin were detected as internal controls.
Immunofluorescence and cell count
The cells were stained with mito-tracker reagent (Beyotime Biotechnology) for mitochondria and lyso-tracker reagent (Beyotime Biotechnology) for lysosomes before fixation in 4% paraformaldehyde at 37°C for 15 min. The nucleus was stained with diamidino-phenyl-indole (DAPI; Life Technologies, USA). The samples were then visualized under a microscope (Bio-Tek), and the fluorescence intensity was measured by calculating the integrated density using ImageJ software.
After staining the cells with DAPI, the culture wells were scanned and photographed under a microscope. The DAPI was then segmented and counted using image analysis software.
Isolation and identification of EVs
The cultured medium was concentrated to a volume of 500 μL using an Amicon Ultra centrifugal filtration device (100 kDa; Merck Millipore). The excretion column (Izon Science, New Zealand) was used according to the protocol. The column was equilibrated at room temperature and rinsed with phosphate-buffered saline (PBS) before adding 500 μL of the sample. Immediately after adding the sample to the column, 500 μL of the fraction was collected. The morphology of EVs was detected by TEM, the size distribution of EVs was captured by NTA and EV markers were detected using western blot.
Mitochondrial membrane potential
The mitochondrial membrane potential was measured using the JC-1 mitochondrial membrane potential assay kit (Beyotime Biotechnology). The R527C iMSCs were treated, washed with PBS and then incubated with a medium containing JC-1 staining solution for 20 min at 37°C. Fluorescence microscopy analysis was performed to determine the fluorescence intensity, which was quantified using ImageJ software.
SA-β-gal assay
Senescent R527C iMSCs were detected using the SA-β-gal staining kit (Beyotime Biotechnology). The R527C iMSCs were fixed with 4% paraformaldehyde for 15 min at 37°C and then incubated overnight with SA-β-gal staining solution at 37°C. Subsequently, standard light microscopy was employed to analyze the results.
Transcriptomics of LMNA R527C iMSCs
Passages 6 and 13 (P6 and P13) of R527C iMSC groups were used for transcriptomic sequencing. Total RNA was isolated using TRIzol reagent, and its purity, integrity and concentration were examined. The total RNA was then used to synthesize cDNA and create a library, which was sequenced on a 150 paired-end Illumina platform (Illumina. San Diego, USA). The fragments per kilobase million (FPKM) of each gene were calculated based on the gene length to estimate the level of gene expression. DESeq2 software (1.20.0) was utilized for differential expression analysis between two comparison groups, and Benjamini and Hochberg’s false discovery rate was applied for test corrections. Genes with corrected P-value < 0.05 and |log2FC| > 1 were defined as DEGs. The software ClusterProfiler (3.4.4) was used to perform GO enrichment and KEGG pathway analysis on the DEGs, with three replicates for each passage.
Proteomics analysis of LMNA R527C iMSCs
Total protein was extracted from liquid nitrogen-cured samples, and its concentration was measured using the Bradford protein quantification kit. The extracted protein was subjected to trypsin digestion, and the resulting peptides were separated using a C18 Nano-Trap column and an EASY-nLC 1200 ultra-high performance liquid chromatography (UHPLC) system (Thermo Fisher Scientific, USA). The separated peptides were then analyzed using an Orbitrap Exploris 480 mass spectrometer equipped with FAIMS (Thermo Fisher Scientific) and an ion source of Nanospray Flex to generate raw data.
The results of protein quantification were analyzed statistically using a t-test. Proteins with significant quantitative differences between the P6 and P13 groups (P < 0.05, |log2FC| > 1) were considered DEPs. Protein functional annotation was performed using a public nonredundant database that included PANTHER, Pfam, PRINTS, ProDom, ProSiteProfiles and SMART. DEPs were used for volcanic map analysis, cluster heat map analysis and enrichment analysis of GO and KEGG. Significantly enriched GO terms and KEGG pathways were identified with a corrected P-value < 0.05. Each passage was replicated three times.
To investigate the correlation between transcriptome and proteomic data at different passages, we applied default screening criteria of P-value < 0.05 and |log2FC| > 1. The Pearson correlation coefficient was utilized to determine the relationship between mRNA and protein expression levels. Then, co-expressed DEGs and DEPs were subjected to enrichment analysis of GO terms and KEGG pathways.
LMNA R527C iMSC-EV small RNA sequence
The purification of total EV-RNA was carried out by the exoReasy kit (Qiagen, Germany). RNA purity was assessed using the NanoPhotometer spectrophotometer (IMPLEN, USA), concentration was measured with the Qubit RNA assay kit in a Qubit 2.0 fluorometer (Life, USA) and integrity was evaluated using the RNA Nano 6000 assay kit of Agilent Bioanalyzer 2100 system (Agilent Technologies, USA). Sequencing libraries were generated using NEBNext Multiplex small RNA library prep set for Illumina (NEB, USA). Library quality was assessed on the Agilent Bioanalyzer 2100 system using DNA high-sensitivity chips (USA). The library was sequenced using an Illumina Hiseq 2500/2000 platform, generating 50 bp single-end reads.
The miRNA expression levels were quantified as transcript per million. Differential expression analysis was performed using DESeq2 software (version 1.20.0), and miRNAs with a corrected P-value < 0.05 were defined as DEMs. Novel miRNAs were predicted based on the hairpin structure features of miRNA precursor using both miREvo and mirdeep2 software.
Statistical analysis
Representative experimental data were also reported, and all data were presented as means ± standard deviation. All experiments were performed three times using triplicate biological samples unless otherwise noted. The statistical significance of differences between the two experimental groups was analyzed using a t-test. P-value < 0.05 was considered statistically significant.
Conclusion
Our study showed that LMNA R527C iMSCs exhibited marked characteristics of cell senescence and attenuation of stemness potential, accompanied by immunophenotypic changes during in vitro expansion. The gene sets related to DNA replication, nuclear structure and chromatin showed the most significant changes during replicative senescence of LMNA R527C iMSCs, and the pathways such as cell cycle, DNA replication, inflammation and cell adhesion might play important roles in senescence. LMNA R527C iMSC-EVs might promote surrounding cell senescence by carrying high levels of pro-senescence miRNAs, including miR-311, which could serve as a novel biomarker for detecting chronic and acute MSC senescence and contribute to promoting senescence. Our findings have revealed the mechanism underlying LMNA mutation in MSC senescence, providing new insights into MADA therapy and the link between chronic inflammation and senescence development.
Conflict of interest statement. The authors declare no conflict of interest.
Acknowledgements
We thank Professor Wei Shu (Guangxi Medical University) for giving us the peripheral blood cells of patients with LMNA p.R527C homozygous mutation and healthy people for this study.
Funding
This work is supported, in part, by the National Natural Science Foundation of China (2072480, 32100603); Shenzhen Commission of Development Reform (Funding for Shenzhen Engineering Laboratory for Orthopedic Diseases and Regenerative Technologies).
Data availability
The transcriptome data have been deposited in the CNSA of CNGBdb under the accession number CNP0003607 (https://db.cngb.org). The proteome data have been deposited in the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD037639 (http://www.ebi.ac.uk/pride/archive). The small RNA sequencing data have been deposited in NCBI under the accession number PRJNA891637 (https://www.ncbi.nlm.nih.gov).
