Erythropoietin Ameliorates Neonatal Hypoxia-Ischemia-Induced Neurobehavioral Deficits, Neuroinflammation, and Hippocampal Injury in the Juvenile Rat

Abstract

The hematopoietic growth factor erythropoietin (EPO) has been shown to be neuroprotective against hypoxia-ischemia (HI) in Postnatal Day 7 (P7)-P10 or adult animal models. The current study was aimed to determine whether EPO also provides long-lasting neuroprotection against HI in P5 rats, which is relevant to immature human infants. Sprague-Dawley rats at P5 were subjected to right common carotid artery ligation followed by an exposure to 6% oxygen with balanced nitrogen for 1.5 h. Human recombinant EPO (rEPO, at a dose of 5 units/g) was administered intraperitoneally one hour before or immediately after insult, followed by additional injections at 24 and 48 h post-insult. The control rats were injected with normal saline following HI. Neurobehavioral tests were performed on P8 and P20, and brain injury was examined on P21. HI insult significantly impaired neurobehavioral performance including sensorimotor, locomotor activity and cognitive ability on the P8 and P20 rats. HI insult also resulted in brain inflammation (as indicated by microglia activation) and neuronal death (as indicated by Jade B positive staining) in the white matter, striatum, cortex, and hippocampal areas of the P21 rat. Both pre- and post-treatment with rEPO significantly improved neurobehavioral performance and protected against the HI-induced neuronal death, microglia activation (OX42+) as well as loss of mature oligodendrocytes (APC-CC1+) and hippocampal neurons (Nissl+). The long-lasting protective effects of rEPO in the neonatal rat HI model suggest that to exert neurotrophic activity in the brain might be an effective approach for therapeutic treatment of neonatal brain injury induced by hypoxia-ischemia.

Keywords: erythropoietin; hypoxia-ischemia; neurobehavioral performance; neuronal death.

Figure 1

Recombinant EPO (rEPO) reduced HI-induced deficits in sensorimotor behaviors of P8 rats: (A) righting reflex; (B) negative geotaxis; (C) wire hanging maneuver; and (D) hind-limb suspension test. The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI, EPO+HI, or HI+EPO group compared with the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 2

rEPO attenuated HI-induced reduction in mature oligodendrocytes in cingulum area (A–E) and myelination in the subcortical white matter (F–J) of P21 rats. Representative photomicrographs of APC-CC1+ cells (red color) (A–D), and MBP+ staining (brown color) (F–I): (A,F) sham group; (B,G) HI group; (C,H) EPO+HI group; and (D,I) HI+EPO group. (E) Cell density of the APC-CC1 positive mature oligodendrocytes and (J) MBP positive myelination were quantified in both the ipsilateral and contralateral cingulum areas of the P21 rat brain, and HI-induced injury to oligodendrocytes or myelination was determined as a ratio of the ipsilateral cell density to the contralateral cell density. The results from both APC-CC1 and MBP staining are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI group compared with the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 3

rEPO attenuated HI-induced neuronal death, as assessed by Fluoro-Jade B staining (A–E) in cingulum area of P21 rats. Representative photomicrographs of Jade B+ (green color) cells (A–D): (A) sham group; (B) HI group; (C) EPO+HI group; and (D) HI+EPO group. (E) Quantitation of the Fluoro-Jade B labeled degenerating neurons in the cingulum area. The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI, or HI+EPO group compared with the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 4

rEPO attenuated HI-induced growth retardation (A) and deficits in locomotor activity, as determined in the open field test (B–D) of P20 rats: (A) body weight; (B) crossing distance; (C) total rearing events; and (D) stereotyped behaviors. The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI, EPO+HI, or HI+EPO group compared with the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 5

rEPO attenuated HI-induced decrease in the successful rate of placing in the contralesional limb in the vibrissa-elicited forelimb-placing test of P20 rats: (A) ipsilateral foot and (B) contralateral foot. The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI group compared to the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 6

rEPO reduced HI-induced neuronal death, as assessed by Fluoro-Jade B staining in striatum (A–E) and cortical area (F–J) of P21 rats. Representative photomicrographs of Jade B+ cells (green color) in striatum (A–D) and cortical area (F–I): (A,F) sham group; (B,G) HI group; (C,H) EPO+HI group; and (D,I) HI+EPO group. Quantitation of the Fluoro-Jade B labeled degenerating neurons in the striatum (F), or cortical area (J). The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI, or HI+EPO group compared to the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared to the HI group.

Figure 7

rEPO attenuated HI-induced learning deficit, as determined by the passive avoidance, 15 days (P20) after hypoxia-ischemia. The number of electric foot shocks needed to retain the rat on the safe board was recorded as an index of acquisition of passive avoidance on P20 rats (A); and the retention latency was studied the next day as an index of retention of passive avoidance on P21 rats (B). The results are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI group as compare to the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 8

rEPO attenuated HI-induced reduction in Nissl+ neurons (A–E) and HI-induced neuronal death, as assessed by Fluoro-Jade B staining (F–J) in CA1 region of hippocampal area of P21 rats. Representative photomicrographs of Nissl+ (violet color) (A–D), and Jade B+ (green color) cells (F–I): (A,F) sham group; (B,G) HI group; (C,H) EPO+HI group; and (D,I) HI+EPO group. (E) Cell density of the Nissl positive neurons was counted at both the ipsilateral and contralateral cingulum areas in P21 rat brain, and HI-induced injury to neurons was determined as a ratio of the ipsilateral cell density to the contralateral cell density. (J) Quantitation of the Fluoro-Jade B labeled degenerating neurons in the CA1 region of hippocampal area. The results from both Nissl and Fluoro-Jabe B staining are expressed as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 represents significant difference for the HI, or HI+EPO group compared with the sham group; ^# p < 0.05 represents significant difference for the EPO+HI, or HI+EPO group compared with the HI group.

Figure 9

rEPO reduced HI-induced microglia activation, as assessed by CD11b (OX42) staining in hippocampal CA1 region (A–E) and white matter cingulum area (F–J) of P21 rats. Representative photomicrographs of OX42+ cells (red color) in hippocampal CA1 region (A–D), and white matter cingulum area (F–I): (A,F) sham group—most microglia were at a resting status with a ramified shape (arrows indicated in (A,F)) in the rat brain of the sham group (A,F); EPO+HI group (C,H); and HI+EPO group (D,I). Numerous activated microglia with an enlarged cell bodies and blunt processes were observed in the HI group (arrows indicated in (B,G)). Quantitation of the OX42 labeled activated microglia in the hippocampal CA1 region (F), or white matter cingulum area (J). The results are represented as the mean ± SEM of six animals in each group and analyzed by one-way ANOVA, followed by the Student–Newman–Keuls test. * p < 0.05 expresses significant difference for the HI, or HI+EPO group as compared to the sham group; ^# p < 0.05 expresses significant difference for the EPO+HI, or HI+EPO group as compared to the HI group.