Transcriptome Analysis of Fibroblasts in Hypoxia-Induced Vascular Remodeling: Functional Roles of CD26/DPP4

Abstract

In hypoxic pulmonary hypertension (PH), pulmonary vascular remodeling is characterized by the emergence of activated adventitial fibroblasts, leading to medial smooth muscle hyperplasia. Previous studies have suggested that CD26/dipeptidyl peptidase-4 (DPP4) plays a crucial role in the pathobiological processes in lung diseases. However, its role in pulmonary fibroblasts in hypoxic PH remains unknown. Therefore, we aimed to clarify the mechanistic role of CD26/DPP4 in lung fibroblasts in hypoxic PH. Dpp4 knockout (Dpp4 KO) and wild-type (WT) mice were exposed to hypoxia for 4 weeks. The degree of PH severity and medial wall thickness was augmented in Dpp4 KO mice compared with that in WT mice, suggesting that CD26/DPP4 plays a suppressive role in the development of hypoxic PH. Transcriptome analysis of human lung fibroblasts cultured under hypoxic conditions revealed that TGFB2, TGFB3, and TGFA were all upregulated as differentially expressed genes after DPP4 knockdown with small interfering RNA treatment. These results suggest that CD26/DPP4 plays a suppressive role in TGFβ signal-regulated fibroblast activation under hypoxic conditions. Therefore, CD26/DPP4 may be a potential therapeutic target in patients with PH associated with chronic hypoxia.

Keywords: CD26; DPP4; TGFβ; dipeptidyl peptidase-4; fibroblast; hypoxia; vascular remodeling.

Figure 1

Pulmonary hemodynamic evaluation of hypoxia-induced pulmonary hypertension in wild-type (WT) and Dpp4 knockout (Dpp4 KO) mice. (a) After 4 weeks, right ventricular systolic pressure (RVSP) was significantly higher in WT/hypoxia mice than in WT/normoxia mice, and in Dpp4 KO/hypoxia mice than in WT/hypoxia mice. (b) Fulton’s index (the weight ratio of the right ventricle to the left ventricle plus the ventricular septum) was significantly higher in WT/hypoxia mice than in WT/normoxia mice, and Dpp4 KO/hypoxia mice tended to have a higher index than WT/hypoxia mice (p = 0.13). ns; not significant, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 2

Evaluation of pulmonary small vessel remodeling in hypoxia-induced pulmonary hypertension in WT and Dpp4 KO mice. (a) Representative Elastica van Gieson (EVG)-stained small pulmonary vessels in mice. Blue arrows indicate pulmonary small vessels. Scale bar, 50 μm. (b) The medial wall thickness of the small pulmonary vessels in mice was calculated as the average thickness of four medial walls divided by the average diameter of two perpendicular external elastic laminae. ns; not significant, **** p < 0.0001.

Figure 3

The number of constituent cells of the lung in Dpp4 KO or WT mice 4 weeks after normoxic or hypoxic exposure evaluated by flow cytometry (n = 5–9). (a) The number of CD45⁻/CD31⁺ endothelial cells was larger during chronic hypoxia, although Dpp4 KO did not affect this response. (b–d) No significant differences were observed in the numbers of CD45⁻/CD31⁻/CD326⁺ epithelial cells (b), CD45⁻/CD31⁻/CD326⁻ mesenchymal cells (c), and CD45⁺/CD31⁻ hematopoietic cells (d) regardless of Dpp4 KO under hypoxic conditions. ns; not significant, * p < 0.05.

Figure 4

Inflammatory responses during acute and subacute hypoxic exposure. (a) The number of CD45⁺/Gr-1⁺ neutrophils in bronchoalveolar lavage fluid (BALF) from the mice was not different between WT/hypoxic and Dpp4 KO/hypoxic mice. (b) BALF protein levels did not differ between WT/hypoxic and Dpp4 KO/hypoxic mice. ns; not significant.

Figure 5

Transcriptome analysis of cultured human lung fibroblasts (HLFs) treated with hypoxia and DPP4-small interfering RNA (siRNA). Cultured HLFs were treated as follows (each group, n = 4): Control/normoxia (treated with negative control siRNA followed by exposure to normoxic conditions), control/hypoxia (treated with negative control siRNA followed by exposure to hypoxic conditions), DPP4 KD/normoxia (HLFs treated with DPP4-siRNA followed by exposure to normoxic conditions),and DPP4 KD/hypoxia (HLFs treated with DPP4-siRNA followed by exposure to hypoxic conditions). (a–c) Comparisons between control/normoxia and control/hypoxia: (a) principal component analysis (PCA) revealed that the two groups could be distinguished; (b) volcano plot of the distribution of the log2 fold changes and p-values, with blue dots representing downregulated differentially expressed genes (DEGs) and red dots representing upregulated DEGs; (c) heat map of the DEGs. (d–f) Comparisons between control/normoxia and DPP4 KD/normoxia: (d) PCA revealed that the two groups could be distinguished; (e) volcano plot of the distribution of the log2 fold changes and p-values, with blue dots representing downregulated DEGs and red dots representing upregulated DEGs; (f) heat map of the DEGs. (g–i) Comparisons between control/hypoxia and DPP4 KD/hypoxia: (g) PCA revealed that the two groups could be distinguished; (h) volcano plot of the distribution of the log2 fold changes and p-values, with blue dots representing downregulated DEGs and red dots representing upregulated DEGs; (i) heat map of the DEGs.