Multi-parametric photoacoustic microscopy (PAM) of hemodynamic and oxygen-metabolic responses of uninjured neonatal mouse brains to hypothermia or hypoxia.

(A) Schematic of the head-restrained multi-parametric PAM system. PA, photoacoustic; PD, photodiode; HWP, half-wave plate; EOM, electro-optical modulator; PBS, polarizing beam splitter; NDF, neutral-density filter; PM-SMF, polarization-maintaining single-mode fiber; BPF, band-pass filter; DBS, dichroic beam splitter; BS, beam sampler; SMF, single-mode fiber; DL, doublet; CL, correction lens; UT, ring-shaped ultrasonic transducer; WT, water tank. (B) Photograph of the placement of an awake 10-day-old (P10) mouse wearing the head plate in the PAM system. Note that the water tank is removed to better show the mouse and related parts for head-restrained awake-brain imaging. (C) Illustration of the imaging field (5 × 3 mm2) that covers both hemicortices between the Bregma and Lambda on the skull of mouse neonates. (D) Multi-parametric PAM images of the oxygen saturation of hemoglobin (sO2) and blood flow speed in an awake P10 mouse, with the skull temperature set at 37, 32, and 29 °C, respectively. Scale bar: 500 μm. (E) Hemodynamic and oxygen-metabolic responses of the neonatal mouse cortex to different skull temperatures, including cerebral blood flow (CBF), oxygen extraction fraction (OEF), and the cerebral metabolic rate of oxygen (CMRO2). Gray bars: 37 °C; light blue bars: 32 °C; dark blue bars: 29 °C. One-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 4). ns, no significance; *, p < 0.05; **, p < 0.01. (F) Multi-parametric PAM images of sO2 and blood flow speed in an anesthetized P10 mouse under normoxia (inhaled oxygen concentration: 21%) versus hypoxia (inhaled oxygen concentration: 10%). Scale bar: 500 μm. (G) Hemodynamic and oxygen-metabolic responses of the neonatal mouse cortex to normoxia vs. hypoxia, including CBF, OEF, and CMRO2. Gray bars: normoxia; orange bars: hypoxia. Student t-test was performed, and data are presented as mean ± standard deviation (n = 5). *, p < 0.05; **, p < 0.01.

Effects of combined hypoxia-ischemia (HI) on cerebral hemodynamics, oxygen metabolism, and mitochondrial bioenergetics in mouse neonates under normothermia.

(A) Time-lapse PAM images of the oxygen saturation of hemoglobin (sO2) and blood flow speed in an awake P10 mouse during unilateral CCA-ligation, combined HI, as well as 0–40, 40–80, and 80–120 minutes post-HI. Scale bar: 500 μm. (B–D) Hemodynamic and oxygen-metabolic responses of the contralateral (green line and symbols) and ipsilateral (red line and symbols) cortices to the Vannucci HI Model under normothermia, including (B) cerebral blood flow (CBF), (C) oxygen extraction fraction (OEF), and (D) the cerebral metabolic rate of oxygen (CMRO2). Green and red asterisks indicate the statistical significance, if any, over the baseline (i.e., CCA-ligation alone) values measured in the same animal, while black asterisks indicate the statistical significance, if any, between the measurements in the two hemicortices at the same time point. Two-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 5). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. (E) Electroencephalography (EEG) recording of the ipsilateral cortex in a P10 mouse right after the unilateral CCA-ligation, 60-minute hypoxia (inhaled oxygen concentration: 10%), and 60-minute normoxia (inhaled oxygen concentration: 21%). Zoom-in views of the boxed regions show characteristic EEG patterns, including 1. pre-HI baseline, 2. Suppression, 3. burst-suppression correlated with seizure behaviors (highlighted by yellow asterisks), and 4. post-HI suppression. N = 6. (F–H) Comparison of the mitochondrial parameters acquired in the uninjured (UN, gray bars), contralateral (CL, green bars), and ipsilateral (IL, red bars) cortices at 2 or 5 hours post-HI, including (F) oxygen consumption rate (OCR), (G) MitoSox Red, and (H) ΔRH-123 fluorescence. One-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 7). ns, no significance; *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Effects of hypoxia-ischemia (HI) on cerebral hemodynamics, oxygen metabolism, and mitochondrial bioenergetics in mouse neonates under hypothermia vs. normothermia.

(A) Time-lapse PAM images of the oxygen saturation of hemoglobin (sO2) and blood flow speed in an awake P10 mouse during unilateral CCA-ligation, combined HI, as well as 0–40, 40–80, and 80–120 minutes post-HI. Scale bar: 500 μm. (B–D) Hemodynamic and oxygen-metabolic responses of the contralateral (green line and symbols) and ipsilateral (red line and symbols) cortices to the Vannucci HI Model under hypothermia, including (B) cerebral blood flow (CBF), (C) oxygen extraction fraction (OEF), and (D) the cerebral metabolic rate of oxygen (CMRO2). Green and red asterisks indicate the statistical significance, if any, over the baseline (i.e., CCA-ligation alone) values measured in the same animal, while black asterisks indicate the statistical significance, if any, between the measurements in the two hemicortices at the same time point. (E) Comparison of CBF, OEF, and CMRO2 responses of the ipsilateral cortex to the HI insult under normothermia (37 °C, gray bars) vs hypothermia (32 °C, light blue bars). For (B–E), two-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 5). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. (F–I) Comparison of the mitochondrial parameters acquired in the contralateral (CL, green bars) and ipsilateral (IL, red bars) cortices at 5 hours post-HI under normothermia (37 °C, gray bars) vs hypothermia (32 °C, light blue bars), including (F) oxygen consumption rate (OCR), (G) MitoSox Red, (H) H2O2 emission rate, and (I) ΔRH-123 fluorescence. Two-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 4 or 5 for normothermia or hypothermia treatment group, respectively). ns, no significance; *, p < 0.05. (J) Comparison of the ATP concentrations measured in the CL and IL cortices at 6 hours post-HI under normothermia (37 °C, gray bars) vs hypothermia (32 °C, light blue bars). Two-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 3). ns, no significance; *, p < 0.05. For (F–J), the in-vitro analyses were performed in mitochondria isolated from HI-injured mice with or without a 4-hour hypothermia treatment, followed by another (F–I) one hour or (J) two hours recovery in normothermia.

Correlation of CMRO2 and post-HI brain infarction in mouse neonates.

(A–C) Multi-parametric PAM images of (A) the oxygen saturation of hemoglobin (sO2) and blood flow speed and (B) the cerebral metabolic rate of oxygen (CMRO2), as well as (C) the triphenyl-tetrazolium chloride (TTC) analysis of the dissected brain after completion of the in vivo PAM imaging at 24 hours post-HI. The top and bottom rows are data acquired in the neonate brain treated with normothermia (37 °C, upper row) or hypothermia (32 °C, lower row) for 4 hours, respectively. CL, contralateral; IL, ipsilateral. Scale bar: 500 μm. (D) Comparison of the infarct volume quantified based on the TTC analysis (n = 6 for the normothermia group and n = 10 for the hypothermia group). Student t-test was performed, and data are presented as mean ± standard deviation. **, p < 0.01. (E–G) Comparison of (E) cerebral blood flow (CBF), (F) oxygen extraction fraction (OEF), and (G) CMRO2 acquired at 24 hours post-HI in the neonate brain treated with normothermia (37 °C, gray bars) or hypothermia (32 °C, light blue bars) for 4 hours. Two-way ANOVA was performed, and data are presented as mean ± standard deviation (n = 4). ns, no significance; *, p < 0.05; **, p < 0.01. (H–I) Characteristic proton-HRMAS spectra of the snap-frozen tissue from the contralateral and ipsilateral cortex extracted at 24 hours post-HI from HI-injured mice treated with (H) normothermia or (I) hypothermia. The spectra are scaled to the same peak height for myo-inositol. Cr, creatine and phosphocreatine; Cho, choline; Glx, glutamate and glutamine; NAA, N-acetyl aspartate. Two-way ANOVA was performed (n = 5 for normothermia and n = 3 for hypothermia). *, p < 0.05; **, p < 0.01.

Schematic conclusion and open questions.

(A) Combined hypoxia-ischemia (HI) initially suppresses cerebral oxygen metabolism, but sparks a rapid rebound and overshoot of post-HI CMRO2 caused by uncoupled OXPHOS with greater ROS emission, which leads to the demise of mitochondria and the onset of secondary energy failure (SEF). In contrast, recovery at hypothermia attenuates the post-HI surge of mitochondrial respiration and produces more enduring cerebral oxygen metabolism. The threshold of CMRO2-reduction that causes brain infarction in neonates is expected to exist but yet to be established. (B) The combination of energy deficit and tissue acidosis accumulated during HI plus increased glucose-uptake and glycolysis acutely after HI stimulates uncoupled OPXHOS in the HI-injured neonatal brain, leading to excessive ROS emission and rapid mitochondrial injury. According to the literature (Berntman et al., 1981; Hägerdal et al., 1975) and our results, hypothermia may interrupt this pathological process by inhibiting glycolysis and the TCA cycle.

Relationship between the water tank temperature and the mouse skull temperature.

Oxygen consumption of mitochondria isolated from the uninjured (UN), contralateral (CL), and ipsilateral (IL) cortex at (A) 2 hours or (B) 5 hours post-HI. HI, hypoxia-ischemia; mito, mitochondria protein; ADP, adenosine diphosphate; DNP, 2’-4’ Dinitrophenol; OCR, oxygen consumption rate.

1H-HRMAS MRS analysis of the effects of normothermia vs. hypothermia on brain metabolites at 24 hours post-HI. NAA, N-acetyl aspartate; Cr, creatine; PCr, phosphocreatine; Cho, choline; Glx, glutamate. Two-way ANOVA was performed (n = 5 for normothermia and n = 3 for hypothermia). *, p < 0.05; **, p < 0.01.

Potential mechanisms that contribute to the uncoupling of OXPHOS after HI and the effects of therapeutic hypothermia. OMM, outer mitochondrial membrane; IMM, inner mitochondrial membrane; IMS, mitochondrial intermembrane space; ROS, reactive oxygen species; NADH, nicotinamide adenine dinucleotide hydrogen; FADH2, flavin adenine dinucleotide; CoQ, Coenzyme Q; cyt c, cytochrome complex; NO, nitric oxide; CCO, cytochrome c oxidase; ΔΨm, mitochondrial membrane potential; MCU, mitochondrial calcium uniporter; OXPHOS, oxidative phosphorylation.

CBF, OEF, and CMRO2 values measured at 37 °C in the contralateral (green) and ipsilateral (red) cortex in the indicated period that correspond to those illustrated in Fig. 2A.

Data are shown in mean ± SD and compared to the “CCA-ligation” on its own side, unless indicated with a bracket. The p-values were determined by two-way ANOVA (n=5; *: p<0.05; **: p<0.01).

CBF, OEF, and CMRO2 values with post-HI hypothermia treatment, measured at 32 °C in the contralateral (green) and ipsilateral (red) cortex in the indicated period that correspond to those illustrated in Fig. 3A.

Data are shown in mean ± SD and compared to the “CCA-ligation” on its own side, unless indicated with a bracket. The p-values were determined by two-way ANOVA (n=5; ***: p<0.001; ****: p<0.0001).