Midkine Prevents Calcification of Aortic Valve Interstitial Cells via Intercellular Crosstalk

Abstract

Calcified aortic valve disease (CAVD), the most common valvular heart disease, lacks pharmaceutical treatment options because its pathogenesis remains unclear. This disease with a complex macroenvironment characterizes notable cellular heterogeneity. Therefore, a comprehensive understanding of cellular diversity and cell-to-cell communication are essential for elucidating the mechanisms driving CAVD progression and developing therapeutic targets. In this study, we used single-cell RNA sequencing (scRNA-seq) analysis to describe the comprehensive transcriptomic landscape and cell-to-cell interactions. The transitional valvular endothelial cells (tVECs), an intermediate state during the endothelial-to-mesenchymal transition (EndMT), could be a target to interfere with EndMT progression. Moreover, matrix valvular interstitial cells (mVICs) with high expression of midkine (MDK) interact with activated valvular interstitial cells (aVICs) and compliment-activated valvular interstitial cells (cVICs) through the MK pathway. Then, MDK inhibited calcification of VICs that calcification was validated by Alizarin Red S staining, real-time quantitative polymerase chain reaction (RT-qPCR), and Western blotting assays in vitro. Therefore, we speculated that mVICs secreted MDK to prevent VICs' calcification. Together, these findings delineate the aortic valve cells' heterogeneity, underlining the importance of intercellular cross talk and MDK, which may offer a potential therapeutic strategy as a novel inhibitor of CAVD.

Keywords: CAVD; VICs’ calcification; cell communication; midkine (MDK); scRNA-seq.

FIGURE 1

scRNA-seq analysis identifies major valve cell types. (A) Specific cell types were defined (left); t-SNE plot colored by disease groups (middle); every cluster contains cells from normal and calcified patients (right). (B) Cell numbers of each sample. (C) Canonical markers of VECs, VICs, and immune cells; the dot size represents the number of cell proportions. (D–F) Stromal score and immune score in all samples from scRNA-seq and bulk RNA-seq data (n = 5, per group). Box plots describe the median and interquartile range (IQR) of each score. The whiskers depict the 1.5 x IQR. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.001.

FIGURE 2

Immune cells’ profiles in the normal and calcified aortic valve. (A) Identification of macrophages, T cells, and DCs. (B) Colors display group of immune cells. (C) Marker genes: macrophages (CCL3, IL1B, and MMP9), T cells (CD3D, CD7, and IFITM1), and DCs (CCR7, HLA-DPB1, and IDO1). (D) Representative GO terms in each cluster. (E) Differential expression genes between normal-derived and calcific-derived T cells. (F) Gene enrichment of normal-derived and calcific-derived T cells in hallmark pathways.

FIGURE 3

VECs’ profiles in normal and calcified aortic valve. (A) Main cell types of VECs. (B) Colors display the group of VECs. (C) The expression of EDN1 between nVECs and cVECs. (D) GO pathways of upregulated genes in cVECs compared with nVECs. (E) The heatmap shows the activity of transcription factors in each cell population. (F) Pseudotime analysis of VECs’ subsets. (G) Mesenchymal cell markers in VECs’ subsets, color representing expression, and dot size representing cell numbers.

FIGURE 4

VICs heterogeneity. (A) VICs’ subsets colored by sample origin. (B) Markers expression of each cell subset. (C) Differential expression genes: aVICs (IGF1), cVICs (CFD), iVICs (MMP3), mVICs (COL1A1), liVICs (APOE), and sVICs (HSPB1). (D) GO terms of each subset.

FIGURE 5

Intercellular communication. (A) Inferred outgoing communication, and the thickness of flow indicates the contribution of the cell population. (B) Inferred incoming communication. (C,D) Relative contribution of each pair of ligand–receptor.

FIGURE 6

MDK inhibits VICs’ calcification. (A) The hierarchical plot shows the intercellular communication network of the MK pathway. The left portion describes the paracrine activity and the right portion describes autocrine activity; the solid and open circles represent the source and the target, respectively; the size of the circle represents the proportion of cell numbers in each cell subset; the width of the edge represents the communication probability; the edge color is the same as the source. (B) The relative importance of each subset with different roles. (C) The expression of MDK in VICs’ subsets. (D) Alizarin Red S staining of VICs with different treatments: control (normal culture medium), OM (osteogenic medium), MDK (normal culture medium-plus MDK treatment), and OM + MDK (osteogenic medium-plus MDK treatment). (E) The protein expression level of RUNX2 and ALP with different treatments (n = 5 per group). (F) The RNA expression level of osteogenesis-specific genes (RUNX2 and ALP) with different treatments (n = 6 per group). *p < 0.05, **p < 0.01.

FIGURE 7

Schematic illustration of intercellular cross talk among valve cells. Macrophages secreting IL1β and CCL3 to activate VICs’ subsets. cVECs secreting INHBA and tVECs secreting END1 to affect VICs. Calcified-derived T cells secreting IL11, IL6, and TNFRSF11B to activate VICs. mVICs secreting MDK to inhibit VICs’ calcification.