C57BL/6J male and female mice aged 8–10 weeks were purchased from Janvier Labs (Paris, France) and housed in a temperature-controlled and pathogen-free environment with a 12:12-h light-dark cycle. The pups were generated by crossing female and male mice. P5 mouse pups were anesthetized with isoflurane (5% for induction, 1.5–2.0% for maintenance) in a mixture of air and oxygen (1:1), and the right common carotid artery was cut between double ligatures. After the surgical operation, pups were returned to their dams for 1 h. Later, pups were placed in a chamber perfused with a humidified gas mixture (10% oxygen in nitrogen) at 36°C for 45 min. After hypoxic exposure, the pups were returned to their dams until sacrifice or until weaning at P21. Pups that died or bled profusely during the surgical operation were excluded from study. A total of 12 pups were excluded, and a total of 170 pups were used in all experiments. The pups were arbitrarily assigned to each group stratified by sex. Pups for behavior tests were assigned to normal, vehicle-treated HI, and AAT-treated HI groups, and pups for other experiments were assigned to vehicle-treated HI and AAT-treated HI groups. The sample size was calculated based on experiences in our previous studies (Li et al., 2019; Rodriguez et al., 2020). Mice were maintained in the Laboratory for Experimental Biomedicine of Gothenburg University, and all of the experiments were performed in accordance with Swedish national guidelines established by the Swedish Board of Agriculture (SJVFS 2019: 9) and were approved by the Gothenburg Animal Ethics Committee (2200/2019).

The powdered AAT (Sigma, Cat# 9041-92-3) was dissolved in saline at 1 mg to 20 μl for the stock solution and then diluted five times in saline for the working solution. We performed a comparison between intranasal and intraperitoneal administration of AAT treatment and found that the concentration of AAT in the brain tissue was higher after intraperitoneal administration compared to intranasal administration (Supplementary Figure 1). Thus, AAT was administered intraperitoneally in our study. The treatment dose of AAT was chosen based on previous research found in the literature (Janciauskiene et al., 2011; Toldo et al., 2011, 2016; Lewis, 2012; Moldthan et al., 2014; Mauro et al., 2017) and our preliminary experiments. AAT treatment (50 mg/kg) was given twice, with the first dose being given immediately after HI at P5 and the second dose given 72 h later at P8. Vehicle control pups received the same volume of saline intraperitoneally. The whole experimental procedure is illustrated in Figure 1.

Pups were deeply anesthetized with an overdose of sodium pentobarbital and perfused intracardially with PBS. The mouse brains were fixed in 5% buffered formaldehyde (Histofix; Histolab, Gothenburg, Sweden) at 4°C overnight. After dehydration with graded ethanol and xylene, brain samples were paraffin-embedded and cut into 5 μm coronal sections. Brain sections were deparaffinized in xylene and rehydrated in ethanol. Antigen retrieval was performed by boiling sections in sodium citrate buffer, and nonspecific binding was blocked with 4% donkey serum in PBS. The primary antibodies were mouse anti-microtubule-associated protein 2 (MAP2; 1:1,000 dilution, clone HM-2, Sigma, M4403), mouse anti-myelin basic protein (MBP; 1:500 dilution, clone SMI94, BioLegend, 836504), rabbit anti-cleaved caspase-3 (1:200 dilution, ASP175, Cell Signaling Technology, Beverly, Cat# 9661), goat anti-albumin (1:6,000 dilution, Abcam, Cambridge, UK, Cat# ab19194), and rat anti-galectin-3 (1:200 dilution, Invitrogen, Carlsbad, CA, USA, Cat# 14-5301-82). After primary antibody incubation, the appropriate biotinylated secondary antibodies (1:200 dilutions; all from Vector Laboratories, Burlingame, CA, USA) were added for 60 min at room temperature. After blocking endogenous peroxidase activity with 3% H₂O₂, sections were visualized with Vectastain ABC Elite (Vector Laboratories) and 0.5 mg/ml 3,3’-diaminobenzidine enhanced with ammonium nickel sulfate, β-D glucose, ammonium chloride, and β-glucose oxidase.

After deparaffinization, sections were incubated with freshly prepared 0.06% potassium permanganate (KMnO₄) for 15 min and rinsed in distilled water. Slides were then incubated with 0.0004% Fluoro-Jade B (Merck Millipore, Burlington, USA, Cat# AG310-30MG) in 0.09% acetic acid for 30 min in a dark container. After rinsing with distilled water, slides were dehydrated and covered with mounting medium.

Paraffin-embedded brain samples were cut into 5 μm thick coronal sections, and eight consecutive coronal sections with an interval of 500 μm between each section were measured for each brain. The gray-matter area was determined by measuring the MAP2 immunoreactive area, and the subcortical white matter area was determined by measuring the MBP immunoreactive area. The MAP2-positive and MBP-positive areas in each section were measured in both hemispheres, and the volume was calculated using the following formula: V = ΣA × P × T, where V = the total volume, ΣA = the sum of area measurements, P = the inverse of the sampling fraction, and T = the section thickness. The total tissue loss volume ratio of gray matter and subcortical white matter was calculated using the following formula: [(contralateral hemisphere MAP-2 or MBP-positive volume − ipsilateral hemisphere MAP-2 or MBP-positive volume) / contralateral hemisphere MAP-2 or MBP-positive volume]. All of the evaluations were performed using Image J software by investigators blinded to group assignment.

Area contours with fixed locations were drawn and measured in every 50th section. The gelactin-3-positive and caspase-3-positive cells were counted within a defined area (one visual field) of the cortex (100×), striatum (100×), CA1 (100×), and habenular nuclei (200×) for each section. The Fluoro-Jade B-positive cells were counted within the same area of the cortex (100×), striatum (100×), CA1 (200×), and habenular nuclei (200×) for each section. All the evaluations were performed using Image J software by investigators blinded to group assignment.

Brain tissue from the parietal cortex in both hemispheres was dissected out and homogenized immediately on ice with a Dounce tissue homogenizer (Sigma, D8938) in tissue lysis buffer [15 mM Tris-HCl, pH 7.6, 320 mM sucrose, 1 mM dithiothreitol, 1 mM MgCl₂, 3 mM EDTA-K, and 0.5% protease inhibitor cocktail (Sigma, P8340)]. The homogenate was centrifuged at 4°C and 9,200× g for 15 min, and the protein concentration of the supernatant was measured using the bicinchoninic acid method. Individual samples of 20 μg protein were loaded and run on 4%–12% NuPAGE Bis-Tris gels (Invitrogen, Cat# NP0336BOX) then transferred to reinforced nitrocellulose membranes (Bio-Rad, Cat# 162-0112). Membranes were incubated with mouse anti-fodrin (1:1,000 dilution, Enzo Life Sciences, Cat# BML-FG6090-0500) and rabbit anti-actin (1:200 dilution, Sigma, Cat# A2066) overnight at 4°C. After washing, the membranes were incubated with peroxidase-labeled goat anti-rabbit IgG antibody (1:2,000 dilution, Vector, Cat# PI-1000) or peroxidase-labeled horse anti-mouse IgG antibody (1:4,000 dilution, Vector, Cat# PI-2000). Immunoreactive species were visualized using the Super Signal West Pico PLUS Chemiluminescent Substrate (ThermoFisher Scientific, Cat# 34580) and a LAS 3000 cooled CCD camera (Fujifilm, Tokyo, Japan).

A total of 25 μl homogenate sample was mixed with 75 μl extraction buffer containing 50 mM Tris-HCl (pH 7.3), 100 mM NaCl, 5 mM EDTA, 1 mM EGTA, 1 mM PMSF, and 1% protease inhibitor cocktail on a microtiter plate. After incubation for 15 min at room temperature, 25 μM caspase-3-substrate (Ac-DEVD-AMC, Peptide Institute, 670613) in 100 μl assay buffer was added. Caspase-3 activity was measured using a Spectramax Gemini microplate fluorometer (excitation/emission wavelength 380/460 nm every 2 min for 1 h at 37°C) and expressed as pmol AMC/mg protein per minute.

T2-weighted imaging was performed on a preclinical MR scanner 4.7 T (MR Solution, United Kingdom) at P60. Fast spin echo sequence was used for T2-weighted imaging, and the scanning parameters were set as follows: repetition time = 5,000 ms, echo time = 51 ms, reversal angle = 180°, field of view = 22 mm × 22 mm, matrix = 256 × 256, number of slices = 18, slice thickness = 1.0 mm. Preclinical Scan 1.2 software was used for image acquisition, and the volume and statistics functions in the ITK-SNAP 3.8.0 software was used for tissue volume quantification. The total tissue loss volume was calculated as the contralateral hemisphere volume minus the ipsilateral hemisphere volume.

We first wanted to examine whether AAT treatment affected HI-induced brain damage. Brain injury was evaluated by measuring the total tissue loss volume ratio at 7 days after HI based on MAP2 immunochemistry staining (Figure 2A). The average value in AAT-treated mice was 35.2% lower compared to vehicle-treated littermates (Figure 2B). However, a significant difference between AAT-treated and vehicle-treated mice was seen in males but not in females (Figure 2C). The average value was reduced by 42.5% in AAT-treated male mice compared to vehicle-treated male littermates, while the average value was only 22.8% lower in AAT-treated female mice compared to vehicle-treated female littermates (Figure 2C).

For evaluating the effect of AAT treatment on white matter, MBP immunochemistry staining was performed at 7 days after HI (Figure 2D). The white matter injury was calculated as the MBP-positive tissue loss volume ratio in the subcortical white-matter area. The average value in AAT-treated mice was decreased by 31.2% compared to vehicle-treated littermates (Figure 2E). Analyzing the data according to sex, we found again that AAT treatment significantly reduced the degree of white-matter injury in males but not in females, and the average value was 42.9% lower in AAT-treated male mice compared to vehicle-treated male littermates, but there was no significant difference in females between the AAT and vehicle groups (Figure 2F).

We also evaluated the long-term effect of AAT treatment on brain injury in neonatal mice by magnetic resonance imaging (Figure 2G). According to the results of T2-weighted imaging analysis at P60 (55 days after HI), there was a significant 47.9% reduction in the total tissue loss volume in the AAT-treated group compared to the vehicle-treated group (Figure 2H), but no sex difference was observed (Figure 2I).

To determine whether AAT treatment affected HI-induced motor function deficiency and anxiety-like behavior in the mice, a battery of neurobehavior tests were performed at P60 (55 days after HI). Gait analysis by CatWalk XT showed that the paw parameter values of print area, maximum contact area, and mean intensity of all four paws were significantly reduced in the vehicle-treated group compared to normal mice, which indicated that motor development in neonatal mice was markedly impaired after HI injury. However, no significant differences in these parameters between vehicle-treated and AAT-treated groups were observed (Supplementary Figure 2). Stride length is defined as the distance between the placement of a paw and the subsequent placement of the same paw (Figure 3A), and the stride length of all four paws was significantly lower in vehicle-treated mice compared to normal mice, and this deficiency was rescued by AAT treatment (Figure 3B). We also analyzed changes in the stride length of all four paws between the AAT-treated and vehicle-treated groups by sex, and the protective effect of AAT administration was evident in both males and females (Figures 3C,D).

Anxiety-like behavior and locomotor activity were examined by the open field test (Figure 3E). The open field test of locomotor activity requires normal motor skills and is suitable for the evaluation of anxiety level and the response to a novel environment. Some outcomes, such as time spent in the center and distance traveled in the center, likely gauge some aspects of emotionality, including anxiety. We found that the total distance traveled in the open field of vehicle-treated HI mice was reduced compared to normal mice (Figure 3F). In addition, normal mice would walk along the periphery in a new environment, but this phenomenon was not observed in vehicle-treated HI mice, and the distance traveled in the center area by vehicle-treated HI mice was increased compared to normal mice (Figure 3G). AAT treatment did not have an effect on either the total distance traveled in the open field (Figure 3F) or the distance traveled in the center area (Figure 3G). Analyzing the data according to sex, AAT administration had no apparent effect on the total distance traveled in the open field in either male or female HI mice (Figure 3H), but an effect of AAT treatment on the distance traveled in the center area was found in female HI mice (Figure 3I).

Extravagated albumin can be detected with immunohistochemical methods and is a straightforward way to demonstrate the presence of BBB leakage. We observed albumin staining in the parenchyma of the cortex, hippocampus, and part of the thalamic region in the ipsilateral hemispheres at 7 days after HI (Figure 4A). The albumin-positive area in the injured hemisphere was significantly reduced in AAT-treated mice compared to vehicle-treated mice (Figure 4B), but no sex difference was observed between the two groups (Figure 4C).

Galectin-3 immunochemistry staining was performed to detect the activation of microglia in brain tissue at 24 h after HI (Figure 4D). We measured the density of activated microglia cells in four brain regions (the cortex, CA1, habenular nuclei, and striatum), but only the density of activated microglia cells in the striatum was statistically reduced in AAT-treated vs. vehicle-treated mice (Figure 4E). Furthermore, the effects of AAT on microglial activation in the striatum did not exhibit any sex differences (Figure 4F).

The quantification of neuronal cell death based on the number of Fluoro Jade B-positive cells demonstrated substantial overall neuronal cell loss at 24 h after HI (Figure 5A). Regional analysis showed that the numbers of dead or dying neuronal cells in the habenular nuclei and striatum were significantly reduced in AAT-treated compared to vehicle-treated mice (Figure 5B), but no sex difference was observed (Figures 5C,D).

Caspase-dependent apoptotic cell death was investigated by measuring the active form of caspase-3 in different brain regions at 24 h after HI (Figure 6A). In the analyzed brain regions, caspase-3-positive cells were clearly increased in the cortex, habenular nuclei, and striatum in vehicle-treated mice compared to AAT-treated mice, but no significant changes were seen in the CA1 area between the two groups (Figure 6B). Caspase-3-positive cells were significantly decreased in AAT-treated male mice compared to vehicle-treated male mice in both the cortex and habenular nuclei (Figures 6C,D). However, in the striatum caspase-3 activation was significantly inhibited in AAT-treated female mice compared to vehicle-treated female mice (Figure 6E). Additionally, the caspase-3 activity assay clearly showed that AAT treatment significantly reduced caspase-3 activation after HI (Figure 6F), which was more pronounced in females (Figure 6G).

The 280 kDa non-erythroid fodrin protein is a widely studied substrate for calpain. The identification of fodrin cleavage fragments is a method to detect the activation of calpain and caspase-3 (Wang et al., 2001), and calpain-mediated proteolysis of fodrin results in 145 kDa and 150 kDa fragments while caspase-3-mediated cleavage results in a 120 kDa fragment (Figure 6H). From the Western blot quantification results, it was observed that the expression of both the 145/150 kDa fragments and 120 kDa fragment were significantly decreased in AAT-treated mice compared to vehicle-treated mice, which indicated that AAT treatment reduced the activation of both calpain and caspase-3 in the neonatal brain after HI injury (Figures 6I,J). The expression of proteolytically generated fodrin fragments between the two groups was also analyzed by sex, and the results showed that the inhibition of calpain activation by AAT administration was more pronounced in males (Figures 6K,L).