Murine bone-derived mesenchymal stem cells undergo molecular changes after a single passage in culture

The rarity of the mesenchymal stem cell (MSC) population poses a significant challenge for MSC research. Therefore, these cells are often expanded in vitro, prior to use. However, long-term culture has been shown to alter primary MSC properties. Additionally, early passage primary MSCs in culture are often assumed to represent the primary MSC population in situ, however, little research has been done to support this. Here, we compared the transcriptomic profiles of murine MSCs freshly isolated from the bone marrow to those that had been expanded in culture for 10 days. We identified that a single passage in culture extensively altered MSC molecular signatures associated with cell cycling, differentiation and immune response. These findings indicate the critical importance of the MSC source, highlighting the need for optimization of culture conditions to minimize the impact on MSC biology and a transition towards in vivo methodologies for the study of MSC function.

cell cycle, DNA replication and mitosis, while downregulated genes were associated with an impairment to MSC differentiation and immunomodulatory capabilities.Overall, our results shed light on biological differences between primary MSCs and cultured MSCs which could have implications for their application.

A single passage of MSCs in vitro alters the expression of genes associated with essential biological processes
Fluorescence-activated cell sorting was used to isolate primary MSCs and cultured MSCs and RNA sequencing was performed to identify transcriptional profiles (Figs. 1, 2A).To assess variance within the dataset, we performed principal component analysis which identified distinct clustering between primary MSCs and cultured MSCs (Fig. 2B).Differential expression analysis identified a total of 2165 differentially expressed genes (DEGs) in cultured MSCs when compared to primary MSCs (|Log2FC|≥ 2, q-value ≤ 0.01).Of those genes, 952 were upregulated and 1213 were downregulated (Figs.2C, D).
To understand the potential biological ramifications of this altered gene expression we performed gene set enrichment analysis (GSEA).Genes sets associated with the cell cycle, DNA replication, P53 signalling pathways and protein synthesis pathways (i.e., ribosome, biosynthesis of amino acids, RNA transport) were significantly enriched in cultured MSCs (Fig. 3A).Conversely, downregulated genes were enriched for processes including extracellular matrix (ECM) receptor interactions, cytokine-cytokine receptor signalling, and the complement and coagulation cascades, which are the essential components of the innate immune response (Fig. 3B).Furthermore, the cAMP and cGMP signalling pathways, which are important for mediating a range of cellular responses to external stimuli, were also downregulated in cultured MSCs (Fig. 3B).Collectively, this indicates that the alterations to gene expression in cultured MSCs impacts on a number of essential biological processes.

Gene ontology enrichment analysis reveals that genes upregulated in cultured MSCs are associated with cell cycle and cell division
Further insight into the functional profile of cultured MSCs was gained by performing gene ontology (GO) enrichment analysis.Genes observed to be upregulated in cultured MSCs, such as cyclin A2 (Ccna2) (Log2FC = 3.69) and anillin (Anln) (Log2FC = 4.53), were enriched for biological processes including the 'cell cycle' and 'cell division' (Fig. 4A).Furthermore, processes such as 'chromosome segregation' , 'mitotic sister chromatid segregation' , 'mitotic cell cycle' and 'mitotic cytokinesis' , were also enriched in upregulated DEGs.In line with these findings, upregulated genes were associated with nuclear cellular components such as the 'chromosome' , 'centromeric region' , 'kinetochore' and 'spindle' (Fig. 4B).The upregulation of Ccna2 and Anln by MSCs following culture was further confirmed by quantitative reverse transcription polymerase chain reaction      www.nature.com/scientificreports/(RT-qPCR) (Fig. 4C).Taken together, we identified a clear transcriptomic enrichment for cell cycling and DNA replication-related processes.

Gene ontology enrichment analysis reveals that genes downregulated in cultured MSCs are associated with differentiation and immune responses
GO enrichment analysis of genes downregulated in cultured MSCs identified an enrichment for biological processes that are essential for physiological MSC function.For instance, genes associated with 'osteoblast differentiation', including myocilin (Myoc) (Log2FC = − 10.46), insulin-like growth factor 1 (Ig f1) (Log2FC = − 3.44) and transmembrane protein 119 (Tmem119) (Log2FC = − 3.44), were downregulated in cultured MSCs (Figs. 2C, 5A).Likewise, genes associated with 'ossification' and 'bone mineralization' were also downregulated in cultured MSCs, implying that cultured MSCs possess impaired osteogenic potential (Fig. 5A).Similarly, genes associated with promoting adipocyte differentiation ('positive regulation of fat cell differentiation') were also downregulated in cultured MSCs.In contrast, genes involved in the 'negative regulation of chondrocyte differentiation' were downregulated, demonstrating a potential bias towards the chondrogenic lineage in cultured MSCs (Fig. 5A).Overall, a clear perturbation to cultured MSC differentiation potential is evident at the molecular level.
Another essential function performed by MSCs is modulation of the immune response via interactions with B cells, T cells, natural killer cells and dendritic cells 1 .GO enrichment analysis identified that 'immune response' and 'complement activation' processes were downregulated in cultured MSCs.Specifically, cultured MSCs exhibited decreased expression of genes such as complement factor H (Cfh) (Log2FC = − 6.45) and ectonucleotide pyrophosphatase/phosphodiesterase 2 (Enpp2) (Log2FC = − 4.59) (Fig. 2C).These results suggest that the immunomodulatory capabilities of MSCs may also be impaired following a single passage in vitro.
Finally, GO enrichment analysis of cellular components identified that downregulated genes were associated with the 'extracellular region' , 'extracellular space' , 'cell surface' and 'plasma membrane' (Fig. 5B).In addition, GO enrichment analysis identified enrichment of biological processes including 'cell adhesion' and 'ECM organisation' , demonstrating that downregulated genes are important for intercellular interaction and regulating the ECM (Fig. 5A).Specifically, genes mediating cell-cell adhesion such as collagen type XIV alpha 1 (Col14a1) (Log2FC = − 8.11) and thrombospondin 3 (Thbs3) (Log2FC = − 7.24) were downregulated in cultured MSCs (Fig. 2C).Matrix metalloproteinase 2 (Mmp2) (Log2FC = − 3.51), which is responsible for ECM breakdown and remodelling was also downregulated in cultured MSCs (Fig. 2C).Downregulation of these genes in cultured MSCs compared to primary MSCs was confirmed by RT-qPCR (Fig. 5C).These findings suggest alterations to the mechanisms by which MSC's interact with the microenvironment following a single passage in vitro.

Discussion
Over the years, extensive research has established the role of MSCs in diseases such as cancer 20,21 .MSC-based regenerative therapies have been trialled and implemented for a range of injuries and illnesses 11,22 .However, the rarity of this stem cell population necessitates its expansion in culture prior to clinical use.The most commonly used passages for transplantation are from passage 3 to 7, however many studies have found that the passaging process inadvertently alters MSC biology and function [13][14][15][16][17][18] .To date, changes identified by culturing MSCs at increasing passages include alterations to DNA damage response, senescence, multipotent potential and immunomodulatory properties [13][14][15][16][17][18] .Others have shown that an increase in the passage number does not significantly impact the immunoprivilege of MSCs 23 .However, no study has directly compared changes between primary MSCs and MSCs after one passage.Here we provide unique insight into culture-induced molecular changes in MSCs following just a single passage.
Previous studies have demonstrated that MSCs have limited proliferative potential in vitro and undergo senescence with increasing cell passages 13,15 .However, it should be noted that these studies examined the biological aging of MSCs when the cells were already under maintenance in an in vitro setting.In contrast, our study explored the molecular impact of cell culture on MSCs using primary MSCs and cultured MSCs.In fact, we observed a clear distinction in the molecular signatures of cultured MSCs and primary MSCs, particularly in upregulation of genes involved in 'cell cycle' , 'cell division' and 'DNA replication' in cultured MSCs, suggesting that MSC turnover is increased following early passage in vitro.Our results identified that just 10 days in culture induced the upregulation of genes associated with cell cycling.Ccna2 and Anln are responsible for cellular proliferation and were the most significantly upregulated genes associated with 'cell cycle' and 'cell division' biological processes 24,25 .In support of this, we also found that the number of the cultured MSCs after 10 days in culture (average of 1,03,250 cells per mouse) was approximately 27-fold higher than the number of the primary MSCs (average of 3850 cells per mouse).Further studies should be conducted to further investigate the mechanistic role of these genes in this setting and exploit the strategy of genetically manipulating Ccna2 and Anln in vitro to overcome MSC senescence with increasing cell passages.In addition, western blot analysis could also be conducted to evaluate changes in the signaling molecules associated with cell cycle and cell division, such as cyclins, PCNA and Ki67.
Another important characteristic of MSCs is their trilineage differentiation potential.Whilst we previously confirmed that cultured MSCs can differentiate into osteoblasts, adipocytes and chondrocytes, our transcriptomic analyses suggest that differentiation potential may vary between primary MSCs and cultured MSCs 26 .For example, GO enrichment analysis identified that genes associated with 'osteoblast differentiation' , 'ossification' and 'bone mineralization' were downregulated in cultured MSCs.A reduction in MSC osteogenic potential has previously been noted following long-term culture 15,17 .However, our data provide striking evidence that the osteogenic potential of MSCs may be compromised as early as following a single passage in culture.Interestingly, one study has successfully demonstrated the feasibility of preconditioning MSCs with cytokines (e.g., tumor www.nature.com/scientificreports/necrosis factor-alpha) to promote osteogenesis 27 .These findings are clinically important as MSCs are being trialled for their potential to regenerate bone in human subjects 28 .GO enrichment analysis also identified changes to the adipogenic-and chondrogenic-lineage potentials of MSCs following culture.While contention remains regarding the effect of long-term culture on the adipogenic potential of MSCs, our data suggest that culturing MSCs could potentially impair adipogenesis after just one passage 15,17,29 .Furthermore, genes classified as negative regulators of chondrocyte differentiation were downregulated in cultured MSCs, indicating a possible differentiation bias towards the chondrocyte lineage.However, others have reported reduced chondrogenic potential in MSCs following long-term culture, indicating that any bias towards the chondrocyte lineage may only be short-lived 29 .Future studies could validate the effect of cell culture on the differentiation potential of primary MSCs by performing tri-lineage differentiation assays.In addition, the self-renewal capacity of MSCs could also be investigated by comparing the colony-forming unit ability of primary MSCs and cultured MSCs.
MSCs are also capable of regulating the proliferation and activation of immune cells, thereby modulating the body's response to injury and illness 30,31 .Many studies are currently exploring the role of MSC immunomodulation in disease and for the treatment of immune disorders 32 .However, long-term culture can impair the immunosuppressive capabilities of MSCs 18 .Corroborating these findings, we identified that genes downregulated in cultured MSCs were enriched for biological processes including 'response to bacterium' , 'immune response' and 'complement activation' , indicating that impairment of the immunomodulatory ability of MSCs may be altered after just 10 days in culture.For example, a gene found to be downregulated in cultured MSCs was Enpp2, which encodes for autotaxin.Autotaxin is known to mediate lymphocyte migration, thus supporting the proposition that its downregulation may impair MSC immunomodulation 33 .Future studies could validate the immune modulatory properties of primary MSCs and cultured MSCs by priming these cells with cytokines such as IFN-γ, TNF-α, and IL-1β, and subsequently assess changes in the production/secretion of functional factors that regulate immune responses.
Interestingly, the impact of cell culture on the molecular signature of MSCs have also been demonstrated in several studies involving the use of MSCs derived from human bone marrow.For instance, genes involved in regulation of the complement cascade (SERPING1), antitumoral immunity (OAS2) and osteogenesis (ALPL) were found to be downregulated in cultured adherent human bone marrow-derived MSCs compared to primary human bone marrow-derived MSCs 34 .In addition, separate studies have shown that uncultured human bone marrow-derived CD45 low CD271 + MSCs demonstrate distinct transcriptional differences compared to cultured bone marrow-derived MSCs 35,36 .These data from primary human bone marrow MSCs corroborate our findings from murine bone-derived MSCs, supporting the impact of culture-induced molecular changes on MSCs.
Another area of interest is the impact of culture methodologies on MSC biology.Unlike human bone marrow, murine MSCs are not easily isolated from bone marrow aspirates by plastic adherence due to contaminating hematopietic cells.However, the discovery that MSCs are found lining the inner surface of compact bones has given rise to a new method where murine MSCs can be directly isolated from bones cleared of bone marrow, including one pioneered by Zhu et al.where MSCs were successfully isolated from collagenase II-digested bone fragments in vitro 37 .Although this method can remove contaminating hematopoietic cells, it is worth noting that this would involve multiple cell passages which could inadvertently affect MSC biology.Therefore, this necessitates further protocol refinements in the future.In addition, several studies have examined the influence of technical variables including culture medium, dynamic culture, 3-dimensional (3D) culture and oxygenation on MSC biology [38][39][40][41][42][43][44][45] .For instance, a previous study has shown that high levels of oxygen exposure can alter multilineage differentiation capacity of primary bone marrow-derived mouse MSCs, as well as inducing cellular stress, growth arrest and apoptosis via p53 activation 46 .Whilst our study did not specifically examine whether atmospheric oxygen can affect the physiology of long bone-derived MSCs and p53 activity, this remains an avenue for future investigation.Remarkably, dynamic culture systems, preconditioning with cytokines and the use of 3D scaffolds has been found to improve the osteogenic differentiation potential and/or immunomodulatory capabilities of cultured MSCs, which highlights the importance of optimizing culture methodology 41,[47][48][49] .It is also worth noting that there is a lack of consensus with regards to defining MSCs based on surface markers.For instance, some studies define populations of MSCs based on the expression of LEPR and Nestin, while we defined our bone-derived MSCs based on Sca-1 and PDGFRα 19,50,51 .Therefore, consensus regarding a standardized protocol for MSC definition and culture will be beneficial for stem cell research going forward.
A key limitation of our current study is the difficulty of obtaining high numbers of primary MSCs for bulk RNA sequencing via fluorescence-activated cell sorting (FACS) due to the rarity of this population.While we managed to obtain a total of four cultured MSC samples from culturing heterogenous bone cells prior to FACS, we sorted and pooled primary MSCs from four mice and six mice respectively, which only yielded two biological samples for the primary MSC group.Future studies could utilize next-generation sequencing techniques, such as single cell RNA sequencing to overcome this limitation.It is also important to note that our study can only delineate the molecular mechanisms, rather than biological properties using conventional assays due to the fundamental difference between in situ-derived primary MSCs and in vitro-derived cultured MSCs.These differences may also be reflected by the notion that MSCs reside in BM niches where crosstalk among MSCs and niche cells influence their biology, in contrast to mono-cultured MSCs which are deprived of this critical influence.Preclinical experiments which utilize cultured MSCs may therefore not accurately recapitulate in vivo phenotypes and thus consideration of this caveat should be made when interpreting results.
In conclusion, this study provides insight into the molecular changes of murine MSCs after a single passage in culture.Our findings provide a foundation from which we can optimize MSC culture methodology to maintain the characteristics of primary MSCs and investigate the impact of culture-induced changes on the therapeutic efficacy of MSCs in the clinic.

Experimental model and study participants details
Ten-week old female C57BL/6J mice were obtained from the Animal Research Centre, Perth and housed under pathogen-free conditions.Mice were housed at 21-22 °C with 12 h light/dark cycle.All animal work conducted in this study was approved by the Telethon Kids Institute Animal Ethics Committee (AEC), under the following approval numbers and titles: AEC#330: investigating the interaction between bone cells and leukaemia cells (approved 16/10/2017).AEC#P2176: investigate the crosstalk between bone cells and leukaemia cells (approved 01/09/2022).

Figure 2 .
Figure 2. RNA sequencing reveals that cultured MSCs and primary MSCs are transcriptionally distinct.(A) Schematic of the experimental process to obtain cultured MSCs and primary MSCs for RNA sequencing.Schematic was generated by BioRender.com.(B) Principal component analysis based on row z-score of FPKM values for cultured MSC (n = 4) and primary MSC (n = 2) samples.(C) Volcano plot demonstrating the presence of 2165 differentially expressed genes (DEGs) in cultured MSCs compared to primary MSCs (|Log2FC|≥ 2, q-value ≤ 0.01).The top 10 most significantly upregulated DEGs and 15 selected DEGs that are significantly downregulated (according to q-value) are listed.Green represents downregulated genes; red represents upregulated genes.(D) Unsupervised clustering of DEGs as a heatmap.Heatmap is of FPKM values standardized by row z-score.

Figure 3 .
Figure 3. Gene set enrichment analysis of cultured MSCs and primary MSCs.Gene set enrichment analysis (GSEA) using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database demonstrating (A) Gene sets upregulated in cultured MSCs and (B) Gene sets downregulated in cultured MSCs.For GSEA, a |normalized enrichment score (NES)|≥ 1, nominal (NOM) p-value ≤ 0.05 and false discovery rate (FDR) q-value ≤ 0.25 were used as threshold values.

Figure 4 .
Figure 4. Gene ontology enrichment analysis and RT-qPCR validation of the upregulated genes in cultured MSCs compared to primary MSCs.(A) Gene ontology (GO) analysis of biological processes enriched in the 952 genes upregulated in cultured MSCs.The top 20 significantly enriched biological processes are displayed (based on q-value ≤ 0.05).(B) GO analysis of cellular components enriched in the 952 genes upregulated in cultured MSCs.The top 20 significantly enriched cellular components are displayed.(C) Upregulation of Ccna2 and Anln genes as validated by RT-qPCR analysis.All datapoints are presented as biological replicates (n = 7 for primary MSCs and n = 4 for cultured MSCs).Data presented as mean ± SEM relative to primary MSCs.****p ≤ 0.0001.