Proceedings of The Physiological Society

Europhysiology 2018 (London, UK) (2018) Proc Physiol Soc 41, C096

Oral Communications

Deficiency of prolyl-4-hydroxylase domain 2 (PHD2) in neurons enhances microvessel formation in the postnatal mouse brain through HIF-dependent and -independent mechanisms

E. Nasyrov1, H. H. Marti1, R. Kunze1

1. Institute of Physiology and Pathophysiology, Heidelberg University, Heidelberg, Germany, Heidelberg, Germany.


In contrast to embryonic vascularization of the CNS, mechanisms underlying postnatal angiogenesis and remodeling of quiescent adult vasculature are still poorly understood. Regarding forebrain vascularization, the roles of neurons and neuronal oxygen-dependent mechanisms (PHD-HIF axis) remain completely elusive. The Cre-loxP technique was used in mice to induce deficiency of PHD2 (P2), HIF-1α/HIF-2α (H1/H2) and PHD2/HIF-1α/HIF-2α (P2/H1/H2) in postnatal neurons. Brains were harvested at 0, 1 (n=5) and 12 weeks (n=6), respectively. Co-detection of CD31 and intravenously applied EvansBlue by fluorescence microscopy was used to determine the density of functional vessels. Endothelial cell proliferation was evaluated by Ki67/CD31 staining. Immunoblotting, real-time RT-PCR, and RNAscope® in situ hybridization were performed to analyze gene expression. Values are means ± SD, compared by two-way ANOVA. The density of functional vessels in 12 week-old P2 knockout (KO) mice was increased by 55% (255±7 vs 396±7 vessels/µm2, p<0.001) as compared to wildtype control (WT). Ablation of H1/H2 resulted in 20% less perfused vessels (261±17 vs 210±9, p<0.001). Interestingly, P2/H1/H2 KO showed no significant difference as compared to WT, but elicited 17% more functional vessels in comparison to H1/H2 KO (246±8 vs 210±9, p<0.01). At day of birth, no significant difference in vascular density was found in any KO. Earliest alterations occurred at 1 week: +50% (p<0.05) in P2 KO, -20% (p<0.05) in H1/H2 KO, and no difference in P2/H1/H2 KO. Accordingly, endothelial cell proliferation was elevated at 1 week in P2 KO (38±6 vs 60±1 Ki67+ endothelial cells/µm2, p<0.001) and P2/H1/H2 KO (25±5 vs 39±3, p<0.05). Gene expression analysis revealed an up-regulation of the key angiogenic factor Vegf (1.5-fold, p<0.05) in P2 KO, and enhanced transcript levels of angiogenic Spp (1.9-fold, p<0.05) and Adm (1.5-fold, p<0.01). HIF-1α protein levels were elevated in P2 KO by 2.9-fold (p<0.001). Despite the loss of HIF-α in neurons, Vegf transcript levels were increased by 1.7-fold (p<0.05) in juvenile P2/H1/H2 KO as compared to WT. Although HIF-1α protein abundance in P2/H1/H2 KO was reduced to 0.7-fold (p<0.05) of WT levels, it was still elevated as compared to H1/H2 KO (0.7 vs 0.3, p<0.001). These findings suggest that lack of PHD2 in neurons leads to HIF-dependent up-regulation of Vegf in non-neuronal cells. In situ hybridization confirmed astrocytic Vegf up-regulation in P2 KO (18±0.4 vs 24±0.5 transcripts/astrocyte, p<0.01) and in P2/H1/H2 KO (18±0.4 vs 21±0.1, p<0.01), while neurons only minorly contributed to Vegf expression in P2 KO, but not in P2/H1/H2 KO. Collectively, our findings clearly indicate that the PHD-HIF axis in neurons controls microvasculature formation during early postnatal CNS development.

Where applicable, experiments conform with Society ethical requirements