Minocycline-Suppression of Early Peripheral Inflammation Reduces Hypoxia-Induced Neonatal Brain Injury

Abstract

While extensive studies report that neonatal hypoxia-ischemia (HI) induces long-term cognitive impairment via inflammatory responses in the brain, little is known about the role of early peripheral inflammation response in HI injury. Here we used a neonatal hypoxia rodent model by subjecting postnatal day 0 (P0d) rat pups to systemic hypoxia (3.5 h), a condition that is commonly seen in clinic neonates, Then, an initial dose of minocycline (45 mg/kg) was injected intraperitoneally (i.p.) 2 h after the hypoxia exposure ended, followed by half dosage (22.5 mg/kg) minocycline treatment for next 6 consecutive days daily. Saline was injected as vehicle control. To examine how early peripheral inflammation responded to hypoxia and whether this peripheral inflammation response was associated to cognitive deficits. We found that neonatal hypoxia significantly increased leukocytes not only in blood, but also increased the monocytes in central nervous system (CNS), indicated by presence of C-C chemokine receptor type 2 (CCR2⁺)/CD11b⁺CD45⁺ positive cells and CCR2 protein expression level. The early onset of peripheral inflammation response was followed by a late onset of brain inflammation that was demonstrated by level of cytokine IL-1β and ionized calcium binding adapter molecule 1(Iba-1; activated microglial cell marker). Interrupted blood-brain barrier (BBB), hypomyelination and learning and memory deficits were seen after hypoxia. Interestingly, the cognitive function was highly correlated with hypoxia-induced leukocyte response. Notably, administration of minocycline even after the onset of hypoxia significantly suppressed leukocyte-mediated inflammation as well as brain inflammation, demonstrating neuroprotection in systemic hypoxia-induced brain damage. Our data provided new insights that systemic hypoxia induces cognitive dysfunction, which involves the leukocyte-mediated peripheral inflammation response.

Keywords: hypomyelination; inflammation; leukocyte; minocycline; neonatal hypoxia.

Figure 1

The time line of interventions in this study: represents the time point to perform Y-shape electric maze test, represents the time point to perform Morris Water maze test, represents the time point to perform 7-T MRI DTI scanning, represents the time point to detect circulatory total leukocyte number count, represents the time point to perform immunoblotting assay, represents the time point to perform ELISA, represents the time point to perform flow cytometry, represents the time point to perform LFB staining, represents the time point to perform immunofluorescence/immunochemistry. Mino represents minocycline.

Figure 2

Learning and memory alterations after hypoxia with/without minocycline treatment. (A) In P30d rats, the average number of days that animals of indicated groups needed to reach RLCDs was obtained from Y-shape electronic maze test. ^***p < 0.001 vs. the normal group (NG); ^##p < 0.01 vs. the hypoxia group (Hy). n = 14–16 animals per group. (B) The swimming distance from day1 to day 6 in each group, ^**p < 0.01 and ^***p < 0.001 vs. NG group; ^#p < 0.05 vs. Hy group. n = 16–20 animals per group. (C) Escape latencies in hidden-platform task during training in indicated groups. ^***p < 0.001 and ^****p < 0.0001 vs. NG group; ^#p < 0.05 and ^###p < 0.001 vs. Hy group. n = 16–20 animals per group. (D,E) In the probe test of the Morris water maze test, the time spent in the target quadrant (D) and the number of times crossing the platform region (E) of indicated groups were measured. ^*p < 0.05 and ^***p < 0.001 vs. NG group; ^#p < 0.05 vs. Hy group. n = 14–16 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment.

Figure 3

7T MRI/DTI images and the quantification of FA value in the targeted region of interest (ROI). (a–i) Representative T2, T2-C, and DTI images were obtained from the 13rd scanning section (with a clear CCM structure) of P60d rats in indicated groups. Arrows in T2 images indicated expanded septum and lateral ventricles. Arrowheads in T2-C images indicated the regions with reduced myelin sheath direction and destroyed myelin structure. (j) Illustration of the CC and CCM areas where the FA values were calculated. (k,l) The average of FA value of corpus callosum (CC) (k) and middle of CC (l) were measured and analyzed among indicated groups. ^*p < 0.05 and ^**p < 0.01 vs. NG group. ^#p < 0.05 and ^##p < 0.01 vs. Hy group. n = 5 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment. Scale bar = 2.5 mm.

Figure 4

Immunostaining for detection of white matter changes in P60d rat brain after hypoxia. (a–i) The representative images of DAPI (a,d,g), MBP immunofluorescence staining (b,e,h) and merged ones (c,f,i) in corpus callosum in indicated groups and (j) the quantification of MBP fluorescence intensity. Scale bar = 100 μm. n = 3 animals per group. Data expressed as means ± SEM. (a1–f1) The representative images of MBP immunohistochemistry staining in corpus callosum of indicated groups. Scale bar = 50 μm in the top panel and scale bar = 20 μm for the bottom panel. (a2–f2) The representative images of MBP immunohistochemistry staining in cingulum in indicated groups. Scale bar = 50 μm in the top panel and scale bar = 20 μm for the bottom panel. NG represents normal control rats, Hy represents hypoxia rats, Hy M represents hypoxia rats with minocycline treatment.

Figure 5

LFB staining for detection of white matter changes 60 days after hypoxia. The representative images of LFB staining in corpus callosum (a–f) and in cingulum (g–l) in indicated groups. Scale bar = 50 μm in the top and scale bar = 20 μm for the bottom. NG represents normal control rats, Hy represents hypoxia rats, Hy M represents hypoxia rats with minocycline treatment.

Figure 6

The circulatory leukocyte alteration, its relationship to cognitive dysfunction and inflammatory cytokines in peripheral blood after hypoxia. (A) The total number of leukocytes in blood was counted 4 h after hypoxia exposure for 1, 2, or 3.5 h. ^**p < 0.01 and ^****p < 0.0001 νs. 0 h group; ^####p < 0.0001 vs. 1 h group; ^ΔΔp < 0.01 vs. 2.5 h group. n = 20–84 animals per group. (B) The total number of leukocytes in blood was determined at the indicated time points (4 h, 3 d, 7 d, 14 d) after 3.5 h hypoxia exposure. ^**p < 0.01 vs. NG group. n = 26–30 animals per group. (C) Two hours after hypoxia (3.5 h), a single dose of minocycline (45 mg/kg) or vehicle was administered, and after another 2 h the blood samples were taken from indicated groups and the leukocytes were counted. ^****p < 0.0001 vs. NG group. ^####p < 0.0001 vs. Hy group. n = 25–26 animals per group. (D) The correlation between the total number of leukocytes accessed 4 h after hypoxia (3.5 h) with the behavior RLCDs score obtained 30 days after 3.5 h hypoxia exposure was analyzed, using the Spearman's rank correlation analysis. n = 20–22 animals per group. (E,F) ELISA was performed to determine the levels of IL-1β and TNF-α in serum at the indicated time points (4 h and 3 d) after 3.5 h hypoxia exposure. ^*p < 0.05 and ^**p < 0.01 vs. NG group. ^#p < 0.05 vs. Hy group. n = 6–10 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment.

Figure 7

The percentage of CCR2⁺/CD11b⁺CD45⁺ after hypoxia. (A–C) Representative percentage (A) and histogram (B) of CCR2⁺ fluorescence intensity in CD11b⁺CD45⁺ cells after 3.5 h hypoxia and quantitation (c) of CCR2⁺/CD11b⁺CD45⁺ after hypoxia in indicated groups are illustrated. ^*p < 0.05 vs. NG group, ^#p < 0.05 vs. Hy group. n = 7 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment.

Figure 8

Expression levels of CCR2, Iba-1, IL-1β and TNF-α and in the periventricular zone after 3.5 h hypoxia. (A) The representative images and quantification of CCR2 expression that was normalized to GAPDH at indicated time points after hypoxia for 3.5 h were appeared. (B) The representative images and quantification of Iba-1 expression that was normalized to β-actin at indicated time points after 3.5 h hypoxia exposure were shown. (C) The representative images and quantification of IL-1β that was normalized to β-actin expression at indicated time points after hypoxia (3.5 h) were illustrated. (D) The representative images and quantification of TNF-α expression (normalized to β-actin) at indicated time points after hypoxia for 3.5 h were indicated. ^*p < 0.05 vs. NG group. ^#p < 0.05 and ^##p < 0.01 vs. Hy group. n = 5–8 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment.

Figure 9

Effect of minocycline on Iba-1, IL-1β, TNF-α, and TGF-β1 in the Hippocampus after 3.5 h hypoxia. (A) The representative images and quantification of Iba-1expression that was normalized to β-actin at indicated time points after 3.5 h hypoxia exposure were shown. (B) The representative images and quantification of IL-1β that was normalized to β-actin expression at indicated time points after hypoxia (3.5 h) were illustrated. (C) The representative images and quantification of TNF-α expression (normalized to β-actin) at indicated time points after hypoxia for 3.5 h were indicated. (D) The representative images and quantification of TGF-β1 expression that was normalized to β-actin at indicated time points after hypoxia for 3.5 h were appeared. ^*p < 0.05 vs. NG group. ^#p < 0.05 vs. Hy group. n = 7–9 animals per group. Data expressed as means ± SEM. NG represents normal control rats; Hy represents hypoxia rats; Hy M represents hypoxia rats with minocycline treatment.