Fig 1.
Hypoxia-inducible factors (HIFs) are transcription factors that respond to changes in environmental oxygen.
Three HIF prolyl-4-hydroxylases (PHDs) and one HIF asparaginyl hydroxylase (factor inhibiting HIF; FIH) act as oxygen (O2) sensors by regulating the hydroxylation of HIF-α, which controls the stability and transcriptional activity of the HIF system. In this schematic, under Normoxia hydroxylases utilize oxygen as a co-substrate to catalyzing hydroxylation of a specific proline (P via PHDs) and a specific asparaginyl (N; via FIH) amino acid residues on HIF-α. Proline hydroxylation by PHD marks HIF for degradation (i.e., specifically ubiquitination (UB) occurs by VHLE3 ubiquitin ligase (VHL) binding and then rapid degradation by the ubiquitin–proteasome pathway occurs). Hydroxylation at a conserved asparaginyl residue in the HIF-α carboxy-terminal activation domain also blocks interaction with the CBP/p300 transcriptional co-activator required for binding to HIF-α and to DNA. Thus, in normoxia HIF-α is degraded and “Transcriptional Repression” of the HIF system occurs. Under Hypoxia PHDs and FIH are inhibited. Without the binding of VHL to non-hydroxylated HIF-α, HIF-α is not degraded. The coactivator CBP/p300 does bind to non-hydroxylated HIF-α that then binds to HIF-β and then to DNA resulting in HIF system “Transcriptional Activation”, which further up-regulates the expression of a large array of target genes. These increased target genes increase protein expressions and cellular processes that result in red blood cell production (e.g., erythropoietin; EPO), angiogenesis (e.g., vascular endothelial growth factor; VEGF), free radical scavenging and stem cell homing and differentiation. GSK360A mimics hypoxia by inhibiting PHDs and producing HIF transcriptional activation and the upregulation of EPO and VEGF target genes and EPO and VEGF protein synthesis. This schematic was modified from Schofield and Ratcliffe, 2004 [71].
Fig 2.
Overall schematic of basic experimental studies and designs.
Two-pretreatments of GSK360A (30mg/kg) at 18 hours and 1 hour prior the experiment was administrated orally. (A) Pilot pharmacokinetic and pharmacodynamic tests including plasma, kidney and brain GSK360A levels and blood EPO and VEGF levels were measured 5 and 24 hours after and data were presented in Figs 3 and 4. (B) Ischemic stroke was induced by tMCAO and GSK360A at 18 and 5 hours prior to Stroke that was produced in rats at time 0. The effects of GSK360A on body weight, sensory-motor neurological deficits, cognitive function, biochemical changes on HIF related protein/molecular and brain infarcts were evaluated (Figs 5–10).
Fig 3.
The blood plasma, kidney, and brain concentration levels (ng/ml) of GSK360A presented for 5 and 24 hours after GSK360A (30 mg/kg) was administered orally by gavage.
Adequate plasma and peripheral end-organ levels were achieved as also cross-validated by our previous plasma data (37). N = 2 rats per measurement; corroborated by previous data [37].
Fig 4.
(A) The blood plasma levels of EPO 5 hours and 24 hours after 30 mg/kg, p.o. GSK360A administration is presented. The GSK360A treated group had significantly higher levels of EPO compared to vehicle group even at 24 hours after administration (e.g., >80-fold over the vehicle group for both time points). (B) The blood plasma levels of VEGF 5 hours and 24 hours after GSK360A administration is presented. GSK360A treated group had significantly higher levels of VEGF at both time points compared to vehicle group. N = 5 rats per group. Two-way ANOVA test followed by post hoc analysis using the Bonferroni procedure for multiple comparisons; *p < 0.05 and **p < 0.001 when compared with vehicle group.
Fig 5.
Body weight was measured over the course of the study.
Thus, 24 hours prior to tMCAO Stroke body weight was measured and represents the “Pre” measurement. This was followed by the oral (gavage) administration of 30mg/kg GSK360A at 18 and 5 hours prior to Stroke that was produced in rats at time 0. Then body weight was measured at 5 hours, 1 day, 1 week and 3 weeks after stroke. No significant differences in body weight were observed between the vehicle and the GSK360A treated groups over the course of study. N = 8 rats per group. Two-way ANOVA test followed by post hoc analysis using the Bonferroni procedure for multiple comparisons. There was no difference in body weight over time (p = 0.7064) between vehicle and GSK360A groups.
Fig 6.
The effects of GSK360A on (A) mNSS, (B) beam balance, (C) foot fault and (D) hind limb performance at 5 hours, 1 day, 1 week and 3 weeks after stroke. GSK360A treatment significantly reduced post-stroke deficits in all tests (ANOVA group factor for all tests p < 0.01). Additional analysis following up on the significant (p < 0.01) group by trial interaction indicated that mNSS was significantly decreased compared to vehicle treatment at 1 and 3 weeks post-tMCAO stroke. Beam Balance performance, a component of the mNSS was similarly improved by GSK360A treatment. GSK 360A also produced a significant overall improvement in foot fault performance. Finally, GSK360A treatment produced a significant overall improvement in hind limb deficit, and following up on the significant (p < 0.01) group by time interaction resulted in identifying hind limb deficit score reduced compared to vehicle at 3 weeks after stroke. N = 8 rats per group. Two-Way ANOVA, followed by post hoc analysis using the Bonferroni procedure for multiple comparisons; *p < 0.05 when compared with vehicle groups.
Fig 7.
Effect of GSK360A on APA learning performance 3 weeks after tMCAO stroke.
(A) The Two-Way ANOVA indicated that the group and group x trial interaction effects were significant (p < 0.01). Thus, looking at the interaction of groups over trials, the GSK360A treated group received significantly less shocks (i.e., made less errors in avoiding the shock quadrant) over trials (p < 0.01). Specifically, the GSK360A treated group received significantly less shocks in trial 5 and trial 7. (B) The effect of GSK360A was not due to differences in rat movements during the trials, as there was no significant difference in distance traveled per trial between the two groups. N = 8 rats per group. Two-Way ANOVA, followed by post hoc analysis using the Bonferroni procedure for multiple comparisons; *p < 0.05 and **p < 0.01 when compared with vehicle groups.
Fig 8.
GSK360A treatment significantly increased both (A) plasma EPO and (B) plasma VEGF levels 24 hours after stroke. The increase in EPO 24 hours after stroke was greater than that produced by the same dose without stroke suggesting a more prolonged increase than without stroke as shown in Fig 4. N = 8–10 rats per group. t-test; **p < 0.01 when compared with vehicle groups.
Fig 9.
GSK360A increased kidney EPO and brain VEGF mRNA after stroke.
(A) EPO mRNA in the kidney was significantly increased by GSK360A at 5 hours post-stroke compared to vehicle group. Levels of EPO mRNA returned to vehicle treated levels by 24 hours post-stroke.(B) VEGF mRNA in the brain ischemic hemisphere (above) and in the non-ischemic hemisphere (not shown) was increased by GSK360A at 5 hours and 24 hours post-stroke compared to vehicle group (p = 0.002). N = 5 rats per group. Two-Way ANOVA test, followed by post hoc analyses using the Bonferroni procedure for multiple comparisons; **p < 0.01 when compared with vehicle groups.
Fig 10.
TTC staining of forebrain 2 mm thick sections conducted 4 weeks after tMCAO stroke (1 week after APA testing).
GSK360A treatment resulted in a 30% decrease in hemispheric loss reflecting the decreased infarct size compared to vehicle treatment. N = 10 rats per group. t- test; *p < 0.05 when compared with vehicle group.