The short-term and long-term effects of intranasal mesenchymal stem cell administration to noninflamed mice lung

Abstract

Mesenchymal stem cells (mesenchymal stromal cells; MSC)-based therapies remain a promising approach to treat degenerative and inflammatory diseases. Their beneficial effects were confirmed in numerous experimental models and clinical trials. However, safety issues concerning MSCs' stability and their long-term effects limit their implementation in clinical practice, including treatment of respiratory diseases such as asthma, chronic obstructive pulmonary disease, and COVID-19. Here, we aimed to investigate the safety of intranasal application of human adipose tissue-derived MSCs in a preclinical experimental mice model and elucidate their effects on the lungs. We assessed short-term (two days) and long-term (nine days) effects of MSCs administration on lung morphology, immune responses, epithelial barrier function, and transcriptomic profiles. We observed an increased frequency of IFNγ- producing T cells and a decrease in occludin and claudin 3 as a long-term effect of MSCs administration. We also found changes in the lung transcriptomic profiles, reflecting redox imbalance and hypoxia signaling pathway. Additionally, we found dysregulation in genes clustered in pattern recognition receptors, macrophage activation, oxidative stress, and phagocytosis. Our results suggest that i.n. MSCs administration to noninflamed healthy lungs induces, in the late stages, low-grade inflammatory responses aiming at the clearance of MSCs graft.

Keywords: epithelial barrier; mesenchymal stem cell; noninflamed lung; stem cell-based therapy; transcriptomic profiles.

Figure 1

Induction of low-grade inflammation is a long-term effect of intranasal administration of adipose-tissue-derived mesenchymal stem cells. (A) Experimental mice model used in the study. Female C57BL/6 mice were sacrificed directly after two days (short-term), and nine days (long-term) after mesenchymal stem cells (MSCs) intranasal (i.n.) administration. Saline (vehicle)-treated mice were used as control. (B) Representative Hematoxylin & Eosin (H&E) and Periodic Acid-Shiff (PAS) staining in the lung sections. (C) Summary of flow cytometry analyses of IFNγ, IL-4, IL-17, or IL-10 producing T cell frequency after MSCs administration in the long-term model. U Mann-Whitney test was used to evaluate differences between groups, *p < 0.05; ns, not significant; n=5. (D) Representative confocal staining of occludin, claudin 3, and zonula occludens-1 (ZO-1) in the alveoli after MSCs i.n. administration; DAPI – blue; green – positive signal for analyzed proteins. (E) Summary of transcriptomic profiles of the lung after MSCs i.n. administration. The heatmap shows all (n =853) differentially regulated genes (DEGs) among the analyzed groups. DEGs were identified based on |Log2FoldChange| > 0.5, and adjusted p value < 0.1. Cutoffs were applied with matched HGNC identifiers. Complete linkage clustering was applied. (F) Venn diagram of differentially and commonly regulated genes in the short- (marked red) and long-term model (marked blue). Arrows indicate the up- or down-regulated expression of DEGs.

Figure 2

Mesenchymal stem cell administration to non-inflamed lungs induces the expression of genes associated with immunological pathways. Summary of gene set enrichment analysis for gene ontology. Top 20 most significant GO terms are listed according to the adjusted p-value. Ontologies are presented using lollipop charts with normalized enrichment scores. The percentage values represent the coverage of DEGs in each group to the theoretical size of an analyzed term. Blue and yellow refer to biological processes (BP) and cellular components (CC) of gene ontology, respectively.

Figure 3

MSCs administration to noninflamed lungs causes changes in signaling pathways, and innate and adaptive immune gene clusters. (A) Changes in canonical and non-canonical signaling pathways after MSCs administration. A bar chart representing signaling pathways was generated using Ingenuity Pathway Analysis (IPA) Software. The top 15 most significant pathways in each group were presented. The gene cutoffs were adjusted on p-value<0.1, |Log2FoldChange|>0.5. Specific tissue filters restricting the analysis to pathways related to lungs were applied. Bars marked as red indicate pathway upregulation, while blue bars indicate downregulation. Grey bars refer to no activity pattern was available. White bars correspond to the pathways with a z-score = 0. (B) Heatmaps represent genes related to Th1-, Th2-, and Th17- driven immune response, differentiation of T regulatory cells, tight junction molecules, and mucins. The ratios considering significantly regulated genes to the total number of genes were equaled as follows: 11/67 for Th1- (Biological Process; GO:0042088), 8/62 for Th-2 (Biological Process; GO:0042092), 6/59 for Th17- driven immune response (Biological Process; GO:0072538), 9/42 for regulatory T cell differentiation (Biological Process; GO:0045066), 4/47 tight junctions (Tan et al. Allergy 2018), and 0/16 for mucins (Tan et al. Allergy 2018); S – short-term model; L – long-term model; L/S – long-term model vs short-term model; FDR<0.05; n=5; Wald test with Benjamini-Hochberg correction was used; *p<0.05; ** p<0.01; ***p<0.001.

Figure 4

Innate immune gene clusters are differentially regulated upon mesenchymal stem cell administration. The analysis was performed based on Mouse Genome Informatics (MGI v.6.17) Gene Ontology Browser; http://www.informatics.jax.org/vocab/gene_ontology/; access 29^th June 2022). Only significant genes in either of the groups were plotted (adjusted p-value < 0.05). Heatmaps represent the changes in gene expression related to (A) Pattern Recognition Receptor Signaling Pathway (Biological Process; GO:0002221). (B) Macrophage Activation (Biological Process; GO:0042116). (C) Cellular Response to Oxidative Stress genes (Biological Process; GO:0034599). (D) Phagocytosis (with 4932438a13rik gene excluded due to unidentified biological role); (Biological Process; GO:0006909). (E) Inflammation of Respiratory System (genes predicated by functional analysis in Ingenuity Pathway Analysis software). S – short-term model; L – long-term model, L/S long- vs short-term model. Wald test with Benjamini-Hochberg correction was used; *p<0.05; **p<0.01; ***p<0.001.

Figure 5

The relative expression of analyzed common genes changed longitudinally. (A) Genes-genes interaction networks corresponding to protein products were created using the String database. Nodes marked with the same color represent gene clusters. Solid lines show the connections within the individual cluster, whereas dashed lines refer to the interactions among the nodes. The line weight signifies the confidence of the relationship. (B) Circular bar plot indicated delta of genes expression with the altered z-scores at the time according to Ingenuity Pathway Analysis (IPA). The delta was obtained by subtraction between the expression of genes in the long-term and short-term models. The size and color of the bars represent the magnitude and the direction of change, respectively. A two-tailed Mann U Whitney test was performed to assess the statistical significance.

Figure 6

(A) TPM differences of genes altering the z-scores of pathways indicated by IPA. Dots represent the median (n=5) values of TPM in the group, short-term (S), long-term (L), or control (C). Vertical lines connected to the dots indicate the upper (Q3) and lower (Q1) quantiles. (B) Distribution of the commonly regulated genes in short-term and long-term models. Commonly regulated genes in short- and long-term models altering the z-scores of pathways in IPA were marked red. Yellow dots refer to unique DEGs, whereas dark blue indicates no cutoff criteria met. Dashed horizontal and vertical lines represent the cutoffs for adjusted p-value (BH) < 0.1 and absolute Log2FoldChange > 0.5, respectively. (C) The proposed hypothesis of MSCs fate in the non-inflammatory microenvironment in the lungs. In non-inflamed lungs, MSCs undergo apoptosis which induces low-grade inflammation. An increase in IFNγ producing CD3⁺CD4⁺ cells activates phagocytic M1 macrophages to clear apoptotic MSCs.