Amniotic fluid-derived mesenchymal stem cells as a therapeutic tool against cytokine storm: a comparison with umbilical cord counterparts

Abstract

Background: Several immunosuppressive therapies have been proposed as key treatment options for critically ill patients since the first appearance of severe acute respiratory syndrome coronavirus 2. Mesenchymal stem cells (MSCs) from different sources have been considered for their potential to attenuate the cytokine storm associated to COVID-19 and the consequent multi-organ failure, providing evidence for safe and efficacious treatments. Among them, administration of umbilical cord-derived MSCs (UC-MSCs) has demonstrated a significant increase in survival rates, largely due to their potent immunosuppressive properties.

Methods: We applied next-generation sequencing (NGS) analysis to compare the transcriptomic profiles of MSCs isolated from two gestational sources: amniotic fluid (AF) obtained during prenatal diagnosis and their clinically relevant umbilical cord counterparts, for which datasets were publicly available. A full meta-analysis was performed to identify suitable GEO and NGS datasets for comparison between AF- and UC-MSC samples.

Results: Transcriptome analysis revelaed significant differences between groups, despite both cell lines being strongly involved in the tissue development, crucial to achieve the complex task of wound healing. Significantly enriched hallmark genes suggest AF-MSC superior immunomodulatory features against signaling pathways actively involved in the cytokine storm (i.e., IL-2/STAT, TNF-a/NFkB, IL-2/STAT5, PI3K/AKT/mTOR).

Conclusions: The data presented here suggest that AF-MSCs hold significant promise for treating not only COVID-19-associated cytokine storms but also a variety of other inflammatory syndromes (i.e., those induced by bacterial infections, autoimmune disorders, and therapeutic interventions). Realizing the full potential of AF-MSCs as a comprehensive therapeutic approach in inflammatory disease management will require more extensive clinical trials and in-depth mechanistic studies.

Keywords: Amniotic fluid; COVID-19; Cytokine storm; GSEA; Immunosuppression; Mesenchymal stem cells; Regulatory moieties; Transcriptomic analysis; Umbilical cord.

Fig. 1

AF-MSC proliferation capacity, morphology, phenotype and immunosuppressive effect. Graph showing the growth of AF-MSCs (at P2) over a 10-day culture period A. Doubling time for AF-MSCs at different passages (from P2 to P5) B. AF-MSC morphology visualized by fluorescence microscopy for the identification of phalloidin (green) and the cell nucleus are stained using DAPI (blue) C. Images taken at 20 × magnification, scale bar: 10 µm. Quantitative PCR analysis for the expression of pluripotent (Nanog, Oct4) mesenchymal (cd73, cd90, cd44 and cd105) and hematopoietic (cd34 and cd45) markers in AF-MSCs at passage P2 D. Flow cytometric analysis revealing the percentage of AF-MSCs positive for MSC-associated markers (CD29, CD73, CD90 and CD44). Results are presented as percentage of marker-positive cells and as average of three biological replicates ± standard deviation E. Expression of immunosuppressive markers in AF-MSCs as revealed by qPCR following 24 and 48 h treatment with a cocktail of pro-inflammatory cytokines (TNF-α and IFN-g, 20 ng/ml) F. Data represent the average of fold as fold-change compared with the expression levels found in the untreated cells (* = p > 0.05; ** = p < 0.01). Immunosuppressive effect of AF-MSCs on monocyte proliferation (ThP-1 cell line) as shown by flow cytometry G. Three different culture conditions were set: 1) Cell-to-cell contact, 2) Conditioned Media (CM), and 3) Transwell system. Resting and PHA-activated ThP-1 cells (aThP-1) were included in the study as negative and positive controls, respectively. Data represent the percentage ± SD of three biological replicates (* = p > 0.05; ** = p < 0.01; *** = p < 0.001). H qPCR analysis of anti(Il-10, Tgf-b, Ccl22, Mrc1)- and pro(iNos, IL-6, Tnf-a)- inflammatory gene expression in undifferentiated macrophages (MF) or inflamed macrophages (MF1), respectively, exposed to CM from AF-MSC grown in standard conditions (AF-MSCs) or primed with pro-inflammatory cytokines (iAF-MSCs). Data are normalized to housekeeping gene expression (Gapdh) and presented as fold change relative to MF not exposed to CM for anti-inflammatory genes and to MF1 for pro-inflammatory markers (n = 3). Error bars represent the mean ± SD; ** = p > 0.05; ** = p < 0.01)

Fig. 2

Transcriptome analysis reveals gene enrichment heatmaps between AF- and UC-MSCs. A correlation heatmap of Euclidean distance scores of normalized genes counts on all three amniotic fluid samples and all four umbilical cord samples A. Dendrograms show hierarchical clustering results, and the ColorBrewer Blues scale show the similarities between AF-MSC and UC-MSC samples. A principal component analysis plot of all three amniotic fluid samples and all four umbilical cord samples B. The PCA plot shows the uniqueness in the sample groups. Volcano plot displaying the all differentially expressed genes between AF- and UC-MSCs C. Volcano plot shows the magnitude of change (log fold change) versus P adjusted < 0.05. Red markers indicate significantly upregulated genes and blue markers indicate significantly down-regulated genes (< 0.05). Grey markers indicate genes where Padj > 0.05. Leading edge analysis heatmap on pathways of interest D. Red boxes represent highly enriched genes per the respective pathway. Pink boxes represent weaker yet present enrichment per respective pathway

Fig. 3

Evaluation of differentially expressed genes and the enrichment analysis. Hierarchical clustering heatmap showing top differentially expressed genes between both samples A. 25 highest expressed genes among the sample groups on three different count matrices B. The first heatmap a shows the top 25 highest expressed genes shared among AF-MSC and UC-MSC sample groups, the second heatmap b shows the highest expressed genes in AF-MSCs compared to UC-MSCs and the third heatmap c provides the highest gene expression in UC-MSCs and compares it to AF-MSCs. All differentially expressed genes have a P adjusted < 0.05

Fig. 4

Gene set enrichment analysis of hallmark gene sets significantly enriched in AF- vs UC-MSCs. A Leading edge analysis showing enrichment scores (black line) and false discovery rates (orange line) of all 43 hallmarks pathways enriched in AF-MSCs. B MSCs related to immunomodulatory pathways enriched and their associated enrichment score plots for Hedgehog signaling pathway, Angiogenesis and NOTCH signaling. C SARS-CoV-2 associated cytokine storm pathways enriched and their associated enrichment score plots