Perinatal iron deficiency predisposes the developing rat hippocampus to greater injury from mild to moderate hypoxia-ischemia

Abstract

The hippocampus is injured in both hypoxia-ischemia (HI) and perinatal iron deficiency that are co-morbidities in infants of diabetic mothers and intrauterine growth restricted infants. We hypothesized that preexisting perinatal iron deficiency predisposes the hippocampus to greater injury when exposed to a relatively mild HI injury. Iron-sufficient and iron-deficient rats (hematocrit 40% lower and brain iron concentration 55% lower) were subjected to unilateral HI injury of 15, 30, or 45 mins (n=12 to 13/HI duration) on postnatal day 14. Sixteen metabolite concentrations were measured from an 11 microL volume on the ipsilateral (HI) and contralateral (control) hippocampi 1 week later using in vivo 1H NMR spectroscopy. The concentrations of creatine, glutamate, myo-inositol, and N-acetylaspartate were lower on the control side in the iron-deficient group (P<0.02, each). Magnetic resonance imaging showed hippocampal injury in the majority of the iron-deficient rats (58% versus 11%, P<0.0001) with worsening severity with increasing durations of HI (P=0.0001). Glucose, glutamate, N-acetylaspartate, and taurine concentrations were decreased and glutamine, lactate and myo-inositol concentrations, and glutamine/glutamate ratio were increased on the HI side in the iron-deficient group (P<0.01, each), mainly in the 30 and 45 mins HI subgroups (P<0.02, each). These neurochemical changes likely reflect the histochemically detected neuronal injury and reactive astrocytosis in the iron-deficient group and suggest that perinatal iron deficiency predisposes the hippocampus to greater injury from exposure to a relatively mild HI insult.

Figure 1

Post-hypoxic–ischemic (HI) hippocampal injury. Magnetic resonance imaging and brain sections from three iron-deficient rats obtained 1 week after a unilateral HI injury of 15, 30, or 45 mins on postnatal day 14 showed no hippocampal injury (A, D, and G), neuronal injury and demyelination limited to CA1 hippocampal subarea (arrows in B, E, and H) and injury involving the whole hippocampus (black arrows in C, F, and I) and the adjacent parietal cortex (white arrow in C) on the HI side, respectively. The distribution of rats with these injuries in the two dietary groups is given in Table 1. (15 μm coronal brain sections, Nissl histochemistry, and immunohistochemistry for degraded myelin basic protein on adjacent brain sections. See text for details of RARE image acquisition).

Figure 2

¹H NMR spectra of the rat hippocampus post-HI injury. ¹H NMR spectra from the hippocampi on the HI (top) and control (bottom) hemispheres from an iron-deficient rat were obtained one week after an acute HI injury of 30 mins on postnatal day 14 (whose MRI is shown in Figure 1b). Arrows indicate the direction of changes in certain neurochemicals on the HI side when compared with the control side. Additional peaks due to lipid signals are also present on the HI side. See text for NMR spectroscopy details. Abbreviations: Ala, alanine; Cr, creatine; Glu, glutamate; Gln, glutamine; GPC, glycerophosphorylcholine; Lac, lactate; Ins, myo-inositol; NAA, N-acetylaspartate; PCr, phosphocreatine; PCho, phosphorylcho-line; PE, phosphorylethanolamine; and Tau, taurine.

Figure 3

Post-HI neurochemical profile of the rat hippocampus on the control side. Values = mean±s.e.m. neurochemical concentration or ratio on the unligated left side one week after a unilateral HI injury of 15, 30, or 45 mins on postnatal day 14 in the iron-sufficient (A) and the iron-deficient (B) groups, N as in Table 1. *Significant differences between the iron-sufficient and iron-deficient groups (P≤0.02, each; ANOVA). ^§Significant differences between the dietary groups in these HI subgroups (P < 0.02, each; Bonferroni-adjusted t tests). None of the neurochemicals or ratios differed among HI subgroups within each dietary group. Abbreviations: Ala, alanine; Asp, aspartate; Cr, creatine; GABA, γ-aminobutyric acid; Glc, glucose; Glu, glutamate; Gln, glutamine; GSH, glutathione; GPC, glycerophosphorylcholine; Lac, lactate; Ins, myo-inositol; NAA, N-acetylaspartate; NAAG, N-acetylaspartylglutamate; PCr, phosphocreatine; PCho, phosphorylcholine; PE, phosphorylethanolamine; and Tau, taurine.

Figure 4

Post HI neurochemical changes in the rat hippocampus on the HI side. Values = mean±s.e.m. changes in neurochemical concentrations and glutamate/glutamine ratio on the HI side as a percentage of those on the control side 1 week after a unilateral HI injury of 15, 30, or 45 mins on postnatal day 14, N as in Table 1. There was a main effect of iron deficiency for each neurochemical and glutamate/gluta-mine ratio (P < 0.01, each; ANOVA). *Significant differences between iron-sufficient and iron-deficient groups in these HI subgroups (P < 0.02, each; Bonferroni-adjusted Student's t-tests). Abbreviation: Glu:Gln ratio, glutamate/glutamine ratio.

Figure 5

Post-HI reactive astrocytosis in the hippocampal CA1 subarea. Brain sections on the control side (A and C) and HI side (B and D) are from the same iron-deficient rats shown in Figure 1, and were obtained 1 week after an acute HI injury of 30 mins (A and B) or 45 mins (C and D). Arrows point to reactive astrocytes amidst degenerating neurons in CA1 subarea of the hippocampus on the HI side (B and D) and extending to the control side in the animal with more extensive injury (C). (E) The number of reactive astrocytes on the control and HI sides in the two dietary groups. ^§Significant difference between the iron-sufficient and iron-deficient groups (P < 0.001; Wilcoxon Signed-Rank test). *Significant differences between the control and HI sides in the iron-deficient group (P < 0.01 each; paired t-tests). (A–D: 15 μm coronal brain sections, GFAP immunohistochemistry for astrocytes and Gill's hematoxylin histochemistry for neurons. Values in (E) = mean±s.e.m. GFAP positive cells within a 290 μm × 235 μm grid placed on identical positions on CA1 hippocampal subarea bilaterally, N as in Table 2).