Synergistic effect of sildenafil combined with controlled hypothermia to alleviate microglial activation after neonatal hypoxia-ischemia in rats

Abstract

Background and purpose: The only validated treatment to prevent brain damage associated with hypoxia-ischemia (HI) encephalopathy of the newborn is controlled hypothermia with limited benefits. Additional putative neuroprotective drug candidates include sildenafil citrate, a phosphodiesterase-type 5 inhibitor. The main objective of this preclinical study is to assess its ability to reduce HI-induced neuroinflammation, in particular through its potential effect on microglial activation.

Methods: HI was induced in P10 Sprague-Dawley rats by unilateral carotid permanent artery occlusion and hypoxia (HI) and treated by either hypothermia (HT) alone, Sildenafil (Sild) alone or combined treatment (SildHT). Lesion size and glial activation were analyzed by immunohistochemistry, qRT-PCR, and proteomic analyses performed at P13.

Results: None of the treatments was associated with a significant early reduction in lesion size 72h after HI, despite significant changes in tissue loss distribution. Significant reductions in both Iba1 + (within the ipsilateral hemisphere) and GFAP + cells (within the ipsilateral hippocampus) were observed in SildHT group, but not in the other treatment groups. In microglia-sorted cells, pro-inflammatory markers, i.e. Il1b, Il6, Nos2, and CD86 were significantly downregulated in SildHT treatment group only. These changes were restricted to the ipsilateral hemisphere, were not evidenced in sorted astrocytes, and were not sex dependent. Proteomic analyses in sorted microglia refined the pro-inflammatory effect of HI and confirmed a biologically relevant impact of SildHT on specific molecular pathways including genes related to neutrophilic functions.

Conclusions: Our findings suggest that Sildenafil combined with controlled hypothermia produces maximum effect in mitigating microglial activation induced by HI through complex proteomic regulation. The reduction of neuroinflammation induced by Sildenafil may represent an interesting therapeutic strategy for neonatal neuroprotection.

Fig. 1

Study design (upper panel) and temperature recordings during controlled hypothermia (n = 45 for the control group, n = 57 for the hypothermia group)

Fig. 2

Histological brain lesions in P13 rats subjected to HI at P11 and kept either on normothermia (HI, n = 32) compared to sham (n = 9), or treated by hypothermia alone (HT, n = 17), Sildenafil ip alone (Sild, n = 23), and the combined treatment (Sild + HT, n = 23). Ipsilateral tissue loss (A, B) and hippocampal surface (C) in right hemisphere were normalized to contralateral hemisphere in Nissl staining. Mean % (SD) of tissue loss (A, C) or mean (SD) hippocampal surface (B) were measured by assessing the entire brain from Bregma + 2.52 mm to − 7.20 mm. In A and C, sham vs untreated (NT) HI animals were first compared using a non-parametric Mann Whitney t-test (****p < 0.0001). Then, each treatment was compared to untreated animals using a Kruskal–Wallis test (ns: non-significant). In B, distribution of lesion size groups was analyzed in each treatment group compared to HI using Chi-square for trend test (***p < 0.001; **p < 0.01)

Fig. 3

Iba1-positive microglia density in P13 rats subjected to HI at P11 and kept either on normothermia (HI, n = 32) compared to sham (n = 9), or treated by hypothermia alone (HT, n = 17), Sildenafil ip alone (Sild, n = 23), and the combined treatment (Sild + HT, n = 23), in various brain areas. Ipsilateral optical density (OD) of immunoreactive cells was normalized to contralateral hemisphere at Bregma − 3.36 mm. Sham vs untreated (NT) HI animals were first compared using a non-parametric Mann Whitney t-test (***p < 0.001; ****p < 0.0001). Then, each treatment was compared to untreated animals (HI) using a Kruskal–Wallis test followed by a Dunn’s multiple comparison test when appropriate (^#p < 0.05)

Fig. 4

GFAP-positive astrocytes density in P13 rats subjected to HI at P11 and kept either on normothermia (HI, n = 32) compared to sham (n = 9), or treated by hypothermia alone (HT, n = 17), Sildenafil ip alone (Sild, n = 23), and the combined treatment (Sild + HT, n = 23), in various brain areas. Ipsilateral optical density (OD) of immunoreactive cells was normalized to contralateral hemisphere at Bregma − 3.36 mm. Sham vs untreated (NT) HI animals were first compared using a non-parametric Mann Whitney t-test (*p < 0.05; **p < 0.01; ****p < 0.0001). Then, each treatment was compared to untreated animals (HI) using a Kruskal–Wallis test followed by a Dunn’s multiple comparison test when appropriate (^#p < 0.05)

Fig. 5

Gene expression polarization in microglial or astroglial cells sorted from P13 rats subjected to HI at P11 and kept either on normothermia (HI, n = 41) compared to sham (n = 12), or treated by hypothermia alone (HT, n = 19), Sildenafil ip alone (Sild, n = 20), and the combined treatment (Sild + HT, n = 18). Data were summarized as changes observed in each category of markers, either in ipsi or contralateral hemisphere. Mann–Whitney test was performed for Sham vs HI untreated groups comparison and a Kruskal–Wallis test followed by a Dunn’s multiple comparison test to compare HI untreated group and each of the HI treated groups (Sild, HT and SildHT). Number of arrows corresponds to direction and statistical difference magnitude (one arrow: p < 0.05; 2 arrows: p < 0.01; 3 arrows: p < 0.001; 4 arrows: p < 0.0001; ns: non-significant)

Fig. 6

Proteomic analyses of microglial cells sorted from the ipsilateral hemisphere of P13 rats subjected to HI at P11 without and with various neuroprotective treatments (n = 5 in each experimental group). Principal component analysis of the experimental groups (A), comparisons between Sham and HI-injured animals and effect of hypothermia alone (HT), Sildenafil ip alone (Sild), and the combined treatment (Sild + HT) (B–D). For group comparison (Heat maps and volcano-plots), 2-side unpaired t-test were done and a p-value < 0.01 threshold were used to consider protein differentially significantly expressed

Fig. 7

Over-representation analysis (ORA) against the Reactome of the proteome of microglial cells sorted from the ipsilateral hemisphere of P13 rats subjected to HI at P11 without and with neuroprotective treatment Sild + HT. This ORA analysis was performed to identify the enriched Reactome pathways inversely regulated by HI and in HI-injured animals treated by Sild + HT (A). Reactome pathways defined by logF_C > 1,3, p-value < 0.05 and FDR < 0.25 were considered significantly up or downregulated. Three reactome pathways were identified (B–D) among them common proteins were also inversely deregulated (highlighted in orange). A sensitivity analysis using ORA geneontology/Biological Process noredundant was also performed. Overall proteins inversely deregulated in each analysis were summarized in E and a network of these proteins was obtained using String (http://string-db.org/) with a confidence threshold fixed at 0.4 (medium) (F). This network shows significantly more relations than expected by chance (5 nodes, 7 edges while none was expected, p-value = 2.6^–11)