The duration of hypothermia affects short-term neuroprotection in a mouse model of neonatal hypoxic ischaemic injury

Abstract

Neonatal hypoxic-ischaemic encephalopathy (HIE) is major cause of neonatal mortality and morbidity. Therapeutic hypothermia is standard clinical care for moderate hypoxic-ischaemic (HI) brain injury, however it reduces the risk of death and disability only by 11% and 40% of the treated infants still develop disabilities. Thus it is necessary to develop supplementary therapies to complement therapeutic hypothermia in the treatment of neonatal HIE. The modified Rice-Vannucci model of HI in the neonatal mouse is well developed and widely applied with different periods of hypothermia used as neuroprotective strategy in combination with other agents. However, different studies use different periods, time of initiation and duration of hypothermia following HI, with subsequent varying degrees of neuroprotection. So far most rodent data is obtained using exposure to 5-6h of therapeutic hypothermia. Our aim was to compare the effect of exposure to three different short periods of hypothermia (1h, 1.5h and 2h) following HI insult in the postnatal day 7 C57/Bl6 mouse, and to determine the shortest period providing neuroprotection. Our data suggests that 1h and 1.5h of hypothermia delayed by 20min following a 60min exposure to 8%O2 do not prove neuroprotective. However, 2h of hypothermia significantly reduced tissue loss, TUNEL+ cell death and microglia and astroglia activation. We also observed improved functional outcome 7 days after HI. We suggest that the minimal period of cooling necessary to provide moderate short term neuroprotection and appropriate for the development and testing of combined treatment is 2h.

Fig 1

Summary diagram of the average temperature during (A) 1h, (B) 1.5h and (C) 2h of hypothermia/normothermia induced at 20min after neonatal HI (individual values and mean±SEM). The temperature for the hypothermia/normothermia groups was calculated as an average of the oral temperature for 3 different animals per group per time point, measured with a neonatal digital pacifier thermometer. The graphs show the mean ambient temperature for the hypothermia (HT) and normothermia (NT) chambers. The target temperature for normothermic animals was 36°C and for the hypothermic 33°C (A, B, C, HT n = 3 per time point, NT, n = 3 per time,).

Fig 2. Exposure to 1h hypothermia/normothermia does not provide neuroprotection following neonatal HI.

A) Ipsilateral brain tissue volume loss was not significantly affected in any of the examined brain regions in the hypothermia treated animals compared to normothermia treated controls. B) The number of TUNEL+ dying cells (per 20x eye-field) at 48h following HI, was not decreased in the hypothermia treated group compared to normothermia treated littermate controls. C) Ipsilateral microglial activation score (individual values and median ± interquartile bars) based on the αM integrin immunoreactivity was not decreased in any of the examined regions and overall was not significantly affected in the hypothermia treated group (n = 11) compared to the normothermia treated littermate controls (n = 10). D) Ipsilateral reactive astrogliosis (GFAP immunoreactivity in OLV) was not significantly affected in any of the examined brain regions in the hypothermia treated group compared to normothermia treated littermate controls. Abbreviations: CTX–cerebral isocortex, PYR–pyriform cortex, HIP–hippocampus, STR–striatum, Thal–thalamus, EC–external capsule, IR–immunoreactivity.

Fig 3. Exposure to 1.5h hypothermia/normothermia reduces ipsilateral brain tissue volume loss following neonatal HI.

A) 1.5h of hypothermia significantly reduced ipsilateral brain tissue volume loss (individual values and median ± interquartile bars) compared to normothermia treated controls, with significant individual decrease (t-test) in external capsule (p = 0.0125) and consistent lower tissue loss values in the rest of the examined regions. Mixed Linear Model treating region as a repeated measure revealed p = 0.033. B) The number of TUNEL+ dying cells (per 20x eye-field) at 48h following HI, was not decreased in the 1.5h hypothermia treated group compared to normothermia treated littermate controls. C) Ipsilateral microglial activation score (individual values and median ± interquartile bars) based on the αM integrin immunoreactivity was overall not significantly affected despite the consistent lower average score in all of the examined regions in the 1.5h hypothermia treated group (n = 15) compared to the normothermia treated littermate controls (n = 12). D) Ipsilateral reactive astrogliosis (GFAP immunoreactivity in OLV) was not significantly affected in any of the examined brain regions in the 1.5h hypothermia treated group compared to normothermia treated littermate controls. Abbreviations: CTX–cerebral isocortex, PYR–pyriform cortex, HIP–hippocampus, STR–striatum, Thal–thalamus, EC–external capsule, IR–immunoreactivity.

Fig 4. 2h hypothermia/normothermia reduces microglial activation, reactive astrogliosis, TUNEL+ cell death and brain tissue loss following neonatal HI.

A-C: Ipsilateral Nissl staining (Cresyl-Violet, at rostral parietal level) of 2h hypothermia (A) and normothermia (B) animals and ipsilateral volume loss quantification (C) 48h post-HI. 2h hypothermia reduced volume loss (individual values and median ± interquartile bars) compared to normothermia controls, with significant, individual decrease (t-test) in cortex (p = 0.020), striatum (p = 0.049), external capsule (p = 0.005) and total volume loss (p = 0.040). MLM p = 0.038. D-F: TUNEL+ staining at 48h post-HI–Ipsilateral overview in 2h hypothermia (D) and normothermia (E) animals and quantification (F) (number of TUNEL+ cells per 20x eye-field, individual values and median ± interquartile bars). The normothermia group showed typical pyknotic nuclei of the TUNEL+ cells (E-insert, hippocampus) while the hypothermia group was lacking such cells (D). 2h hypothermia significantly reduced TUNEL+ cell death with significant individual decrease (t-test) in cortex (p = 0.032), hippocampus (p = 0.031), striatum (p = 0.022), external capsule (p = 0.039) and total TUNEL+ cell death (p = 0.024). MLM p = 0.024. G-I: αM+ microglia–Ipsilateral overview in hypothermia (G) and normothermia (H) animals and ipsilateral αM microglial activation score (I, individual values and median ± interquartile bars). Note the phagocytic morphology of the cells in the normothermia group (H-insert, hippocampus), compared to the ramified phenotype in 2h hypothermia treated brains (G-insert). 2h hypothermia reduced αM+ microglial activation across all 6 examined regions, with significant, individual decrease (t-test) in isocortex (p = 0.004), pyriform cortex (p = 0.007), hippocampus (p = 0.003), striatum (p = 0.001), thalamus (p = 0.004), external capsule (p = 0.006) and total microglial activation (p = 0.002). Mixed Linear Model treating region as a repeated measure (MLM) p = 0.002. Hypothermia (n = 11) and Normothermia (n = 11) in all assessments. J-L: GFAP immunoreactivity at 48h - Ipsilateral overview in 2h hypothermia (J) and normothermia (K) animals and ipsilateral quantification (L) in optical luminosity values (OLV, individual values and median ± interquartile bars). J and K inserts: higher magnifications of rostro-parietal isocortex. 2h hypothermia reduced astrogliosis, with significant, individual decrease (t-test) in cortex (p = 0.006), pyriform cortex (p = 0.016), hippocampus (p = 0.015), striatum (p = 0.018), thalamus (p = 0.010) external capsule (p = 0.047) and total GFAP immunoreactivity (p = 0.013). MLM p = 0.009. Abbreviations: CTX–cerebral isocortex, PYR–pyriform cortex, HIP–hippocampus, STR–striatum, Thal–thalamus, EC–external capsule, IR–immunoreactivity. *p<0.05 Scale bars: A, B, D, E, G, H = 600um; inserts = 30um; J, K = 700um.

Fig 5. Exposure to 2h hypothermia/normothermia reduces the time required to change orientation in negative geotaxis following neonatal HI, but has no effect in slipping test.

A) Negative geotaxis at 48h post-HI (postnatal day 9) and 7 days (postnatal day 14). No differences were observed at postnatal day 9 between naïve, normothermia and 2h hypothermia treated animals (Kruskal Wallis test with Dunn’s multiple comparison test). At 7 days post-HI normothermia treatment significantly increased the time needed for change of orientation when compared to naïve controls (p = 0.0004). 2h hypothermia treatment significantly reduced the time necessary for change of orientation compared to normothermia treated littermates (p = 0.005; one–way ANOVA with Sidak’s multiple comparison test). Hypothermia (n = 5), Naïve (n = 4) and Normothermia (n = 4) in all assessments, B) 2h of hypothermia treatment had no effect on the number of missed steps (slipping test) at 14 days (postnatal day 21) post-HI (Mann-Whitney test).