Maternal Alcoholism and Neonatal Hypoxia-Ischemia: Neuroprotection by Stilbenoid Polyphenols
Abstract
Growing attention is being directed toward understanding how maternal nutrition influences fetal neurodevelopment and neonatal brain protection. While a balanced maternal diet can support healthy brain development in the fetus, certain dietary habits may have detrimental effects. One major concern is prenatal exposure to alcohol, which is known to impair overall neonatal development and may heighten susceptibility to hypoxic-ischemic encephalopathy—an acute and often severe condition that contributes significantly to neonatal mortality and long-term neurological disability.
Despite public health initiatives aimed at discouraging alcohol use during pregnancy and lactation, approximately 10% of women in the general population continue to consume alcohol during these critical periods. This study was developed in response to this troubling statistic. The objective was to assess whether supplementing maternal diets with specific polyphenols during pregnancy and breastfeeding could mitigate brain injury and associated neurological impairments in neonates subjected to hypoxic-ischemic insult, particularly in the context of moderate maternal alcohol consumption.
The investigation focused on two stilbenoid polyphenols: trans-resveratrol (RSV) and its hydroxylated analog, trans-piceatannol (PIC), both administered at 0.15 mg/kg/day via drinking water, with or without the inclusion of alcohol at 0.5 g/kg/day. Using a well-established postnatal day seven rat model of hypoxia–ischemia (HI), the study examined brain lesion volumes through magnetic resonance imaging and evaluated the resulting motor and cognitive deficits.
The results indicated that moderate levels of maternal alcohol exposure did not exacerbate the extent of visible brain lesions as measured by imaging, but did significantly worsen motor function outcomes. Maternal supplementation with RSV was unable to counter these effects. In contrast, supplementation with PIC led to a complete recovery of both motor and cognitive functions in the affected pups. Remarkably, the level of neuroprotection achieved by PIC was attained using a dose equivalent to the amount naturally present in a single passion fruit consumed daily by a pregnant woman. These findings suggest that PIC may offer a simple yet effective dietary strategy to enhance neonatal brain resilience in scenarios complicated by maternal alcohol intake
Introduction
Neonatal hypoxia–ischemia (HI) remains a significant global health issue, being one of the primary causes of morbidity and mortality in newborns. Affecting approximately one to six out of every 1,000 births, this condition is associated with a high risk of chronic motor impairments, cognitive dysfunction, and neurological disabilities. The core pathological mechanism of HI involves a critical reduction in cerebral oxygen and glucose supply, leading to neuronal death, which in turn results in long-term sensorimotor deficits and learning impairments. While therapeutic hypothermia remains the only clinical intervention currently available, its efficacy is limited. Despite lowering brain metabolism and slowing down cell death, up to 53% of treated infants still succumb to death or experience moderate to severe neurological impairments. This reality emphasizes the urgent need for novel, more effective treatment options to combat the devastating consequences of neonatal HI.
In parallel, maternal alcohol consumption during pregnancy and breastfeeding continues to be a serious concern, especially in Western countries. Despite public health campaigns, approximately 10% of pregnant women consume alcohol, with statistics in the United States revealing that one in every twelve expectant mothers partakes in alcohol use. Maternal alcohol exposure is a well-documented risk factor for impaired fetal brain development and is the leading cause of intellectual disability in newborns. This condition, widely recognized as fetal alcohol syndrome, occurs in approximately one out of every 500 births. Experimental studies on primary neuronal cultures have demonstrated that even moderate alcohol exposure, when combined with hypoxic conditions, significantly increases neuronal death, suggesting a synergistic exacerbation of brain injury. This highlights the necessity to explore how maternal alcohol intake may influence the severity of neonatal HI and to identify potential protective strategies.
To investigate this issue, the well-established Rice-Vannucci rat model was utilized. This model involves permanent unilateral occlusion of the common carotid artery followed by a controlled hypoxic episode in seven-day-old rat pups, an age developmentally equivalent to term human infants. The model reliably produces substantial damage in the striatum, hippocampus, and cortex, leading to observable behavioral and cognitive impairments. Recent research into the influence of maternal diet on neonatal outcomes has garnered increasing interest. For instance, supplementation with polyphenol-rich pomegranate juice or extracts during pregnancy has demonstrated significant neuroprotective effects in similar animal models. Furthermore, the polyphenol trans-resveratrol (RSV), found in grape skins and certain nuts, has been shown to reduce cellular markers of brain damage when administered directly to pups.
RSV has been extensively studied and is considered safe for maternal consumption. However, its clinical application is hindered by its low bioavailability, which is less than one percent. Interestingly, its structurally similar analog, trans-piceatannol (PIC), derived from passion fruit seeds, exhibits approximately twice the bioavailability. Despite this advantage, PIC remains underexplored, particularly in the context of neonatal HI. Only a limited number of studies have examined its potential, with preliminary data suggesting promising results.
This study was designed to further explore and clarify the protective potential of maternal PIC supplementation during pregnancy and lactation in a model where offspring are subjected to neonatal HI in the context of moderate maternal alcohol exposure. Pregnant and nursing rats were given either RSV or PIC, alone or in combination with ethanol, and the subsequent effects on their pups were assessed. The extent of brain injury in the neonates was evaluated using magnetic resonance imaging (MRI), alongside cellular analysis and behavioral testing, to determine the influence of these compounds on motor function and cognitive performance.
Results
Piceatannol, but Not Resveratrol, Mitigates Neonatal Brain Injuries Following a Hypoxic-Ischemic Episode in a Context of Maternal Alcohol Exposure
Analysis of neonatal brains using Diffusion Weighted Imaging (DWI) revealed distinct patterns of brain injury following HI. While no abnormalities were observed in the sham group, which served as the control, all other experimental groups displayed signs of cerebral damage. These injuries primarily affected the cortex, hippocampus, and striatum. At postnatal day seven (P7), there was no significant difference in brain lesion volumes among the HI-affected groups. However, by postnatal day nine (P9), pups born to dams supplemented with PIC in combination with ethanol showed significantly reduced lesion volumes compared to those in the HI-only and ethanol-only maternal groups. Specifically, lesion volumes in the PIC plus ethanol group were approximately 8%, compared to 19% and 18% in the control HI and ethanol-only groups, respectively.
The apparent diffusion coefficient (ADC), a measurement derived from DWI, offers insight into the severity of cytotoxic brain edema, with lower values indicating more severe injury. After HI, ADC values dropped in all affected groups, consistent with the onset of cytotoxic edema. However, the degree of edema in the cortex and hippocampus was notably less severe in pups from ethanol and PIC-supplemented mothers, suggesting a protective effect.
Neuronal death was further evaluated using Nissl staining in the cortex and striatum. In the cortical region, pups in the control HI group exhibited the highest levels of neuronal death. No substantial difference was observed between the ethanol-only and RSV plus ethanol groups. However, the PIC plus ethanol group demonstrated significantly less neuronal death, indicating a neuroprotective effect of PIC. The trend was even more pronounced in the striatal region, where pups in the PIC plus ethanol group displayed the lowest levels of neuronal loss, while both the ethanol-only and RSV plus ethanol groups showed the highest levels of cell death.
These findings suggest that while RSV did not significantly counteract the effects of maternal alcohol exposure on neonatal HI injury, PIC provided substantial neuroprotection. The administered dose of PIC, equivalent to the piceatannol content found in a single passion fruit, was sufficient to confer these beneficial effects. This result holds promise for future nutritional or pharmacological interventions aimed at improving neonatal outcomes in populations where maternal alcohol consumption cannot be fully avoided.
Maternal Polyphenol Supplementation Allows Recovery of Neonatal HI Brain Lesions Detected by MRI Despite Moderate Maternal Alcoholism
At postnatal day 30, which corresponds to 23 days after the hypoxic-ischemic (HI) insult, magnetic resonance imaging (MRI) continued to reveal persistent brain lesions in all pups belonging to the HIC (hypoxic-ischemic control) and Eth (ethanol-only) groups. However, in the RSV/Eth (resveratrol plus ethanol) group, brain lesions were no longer detectable in 20% of the pups, while 50% of the pups in the PIC/Eth (piceatannol plus ethanol) group showed complete lesion resolution. Pups from the HIC, Eth, and RSV/Eth groups exhibited significantly larger lesion volumes compared to those in the PIC/Eth group. Although maternal supplementation with RSV appeared to reduce the number of pups with visible damage, it did not significantly lower the actual lesion volume. Quantitatively, lesion volumes were recorded as 31% ± 5% in the HIC group, 27% ± 3% in the Eth group, 24% ± 7% in the RSV/Eth group, and 12% ± 6% in the PIC/Eth group, underscoring the notably stronger neuroprotective effect of PIC.
Piceatannol Counteracts the Deleterious Sensorimotor and Cognitive Effects Induced by Neonatal HI, Despite Moderate Maternal Consumption of Ethanol
The hypoxic-ischemic injury in this experimental model caused significant damage to the cortex and striatum, both of which are critical for regulating automatic motor functions. To evaluate the resulting sensorimotor impairments, a series of behavioral tests were conducted.
The righting reflex test was administered on postnatal days 8, 10, and 12 to assess neurological reflexes and motor coordination in pups exposed to various maternal supplementation regimens. Across most groups, performance improved over time, which likely reflects both developmental maturation and learning. However, pups in the Eth group failed to show such improvement, indicating lasting motor deficits. Pups from the sham and PIC/Eth groups performed best, showing no significant difference between them. The HIC and RSV/Eth groups demonstrated intermediate levels of recovery, while the Eth group performed the worst.
Similar trends were observed in the negative geotaxis test, a measure of postural control and coordination. Pups in the Eth group again had the poorest outcomes, whereas those in the PIC/Eth group performed at the same level as the sham controls throughout the testing period. The consistent similarity between sham and PIC/Eth groups across multiple assessments highlights the potent protective effect of maternal piceatannol supplementation.
Sensorimotor asymmetries, resulting from unilateral brain lesions induced by the HI model, were evaluated using the adhesive removal test. This test is designed to measure tactile response and motor planning. All groups, except for the Eth group, showed progressive improvement in the time taken to remove the adhesive patches, suggesting functional recovery. Pups from the RSV/Eth group exhibited intermediate capabilities, while those in the HIC and PIC/Eth groups performed comparably to the sham controls, further reinforcing the neuroprotective benefits of piceatannol.
To quantify the overall degree of sensorimotor impairment, the modified neurological severity score (mNSS) was used at postnatal day 24. Pups in the sham and PIC/Eth groups had significantly lower scores, indicating minimal impairment. In contrast, the HIC, Eth, and RSV/Eth groups recorded much higher scores, consistent with more severe neurological deficits. Specifically, the average mNSS scores were 0.38 ± 0.18 for the sham group, 0.83 ± 0.31 for the PIC/Eth group, 4.86 ± 0.88 for the HIC group, 5.0 ± 0.47 for the Eth group, and 4.83 ± 1.24 for the RSV/Eth group.
Given that the hippocampus is often affected by HI in this model, a long-term cognitive assessment was conducted using the novel object recognition test. This test, administered at postnatal day 45 with a 24-hour retention interval, evaluated hippocampal-dependent memory. Rats in the HIC and Eth groups displayed marked deficits in object recognition, indicative of impaired long-term memory. In contrast, rats in the PIC/Eth and sham groups were able to successfully discriminate between familiar and novel objects. The RSV/Eth group showed partial recovery with intermediate performance. Discrimination index scores were as follows: 0.77 ± 0.04 for the sham group, 0.47 ± 0.09 for the HIC group, 0.41 ± 0.04 for the Eth group, 0.58 ± 0.09 for the RSV/Eth group, and 0.66 ± 0.02 for the PIC/Eth group.
Together, these results demonstrate that while maternal alcohol consumption exacerbates the consequences of neonatal HI, maternal supplementation with piceatannol offers significant protection. It not only reduces the extent of brain injury but also preserves motor function and cognitive ability, effectively reversing many of the negative effects of HI in the presence of maternal ethanol exposure.
Discussion
Neonatal hypoxia-ischemia, resulting from restricted oxygen and glucose supply, initiates a cascade of harmful events starting with energy failure and a significant drop in ATP levels. This energy depletion impairs the sodium–potassium ATPase pump, leading to the loss of membrane potential and cellular depolarization. As a consequence, neurons release excessive amounts of glutamate while simultaneously allowing an influx of calcium ions. This calcium entry is known to activate several detrimental enzymes, including proteases, endonucleases, and nitric oxide synthase, which together contribute to widespread neuronal damage. The initial insult, caused by oxygen and glucose deficiency, triggers excitotoxicity, ionic imbalance, and calcium dysregulation, which all have detrimental effects on mitochondrial function. Following the initial injury, the process of reperfusion introduces an additional burden as reactive oxygen species and free radicals are generated in high amounts, leading to cellular swelling and programmed cell death. Neurons, being highly metabolically active, are particularly susceptible to this combined metabolic and oxidative stress. Their loss often manifests as long-term neurological deficits, including motor impairments, cognitive dysfunction, and developmental delays. The limited success of hypothermia as the only currently approved therapy underscores the urgent need for alternative or complementary treatment strategies.
A growing body of research has highlighted the potential neuroprotective benefits of resveratrol in neonatal hypoxic-ischemic conditions. This polyphenolic compound has demonstrated the ability to reduce the extent of brain infarcts, minimize neuronal death, and alleviate associated cognitive deficits in various experimental models. However, these neuroprotective effects have been observed primarily in studies where resveratrol was administered directly to neonatal subjects through intraperitoneal injections, a route that introduces an inflammatory component. Additionally, prior investigations employed high pharmacological doses of resveratrol, ranging from 20 to 100 mg/kg. In contrast, our study uniquely utilized a low-dose, nutritionally relevant amount of resveratrol, equivalent to 0.15 mg/kg per day, delivered through maternal supplementation. For comparison, this dose approximates an intake of 10.5 mg per day for a pregnant woman weighing 70 kilograms. Resveratrol is naturally present in many foods, such as rhubarb, peanuts, cocoa, and notably grapes, which can contain up to 486 mg per kilogram. Thus, our supplementation regimen simulates the daily consumption of approximately 22 grams of grapes, translating to around thirty grape berries.
In a similar vein, piceid, a glycosylated form of resveratrol found in passion fruit seeds, was included in our study. The content of piceid in passion fruit seeds has been reported to range between 10 and 20 mg per fruit, depending on the size and seed weight. Therefore, the dosage of piceid administered in our rat model corresponds to what a human might receive from eating a single passion fruit each day during pregnancy. Regarding ethanol, the dosage selected to simulate maternal alcohol consumption was moderate, reflecting real-world exposure. Female rats received ethanol at 0.5 g/kg/day, which translates to an intake of approximately 35 grams per day for a pregnant woman—roughly equivalent to two standard glasses of wine. This model mirrors the reality that about 10 percent of pregnant women continue to consume alcohol despite public health recommendations to abstain. While general prevalence estimates are available, precise data on the quantities of alcohol typically consumed during pregnancy remain limited. Nevertheless, historical data suggest that regular consumption of even one glass of wine per day during pregnancy may result in fetal alcohol syndrome in approximately 10 percent of cases.
The primary objective of our study was to evaluate whether maternal supplementation with either resveratrol or piceid could mitigate the effects of hypoxic-ischemic brain damage in neonates, especially under conditions of moderate maternal alcohol consumption. In our investigation, lesion volume was assessed alongside apparent diffusion coefficient measurements, which provided early insight into the severity of cytotoxic edema. Cytotoxic edema is characterized by cellular swelling that restricts extracellular water diffusion, resulting in lower ADC values. Among the different groups studied, pups from mothers receiving piceid supplementation exhibited not only smaller brain lesions but also milder cytotoxic edema, as reflected by higher ADC values in the cortex and striatum. This finding indicates a protective effect of piceid at the early stages of injury. Histological analyses further supported this observation, showing reduced neuronal death in these brain regions. Interestingly, ethanol exposure alone also led to slightly improved ADC values, though this did not correlate with reduced lesion size or diminished neuronal damage. This suggests that ADC measurements alone, particularly shortly after injury, may not reliably predict long-term outcomes in affected brain regions.
We also evaluated sensorimotor function using established behavioral tests. Pups that experienced hypoxia-ischemia displayed significant impairments in the righting reflex and negative geotaxis tasks when compared to control animals. These findings are consistent with earlier studies demonstrating the vulnerability of early reflexes to hypoxic-ischemic injury in rodents. In the ethanol-exposed group, performance did not improve over time, highlighting the compounding negative effects of prenatal alcohol exposure. While maternal resveratrol supplementation yielded modest improvements, it was piceid that produced a striking restoration of function. Pups in the piceid plus ethanol group performed at levels comparable to uninjured control animals, suggesting that maternal piceid intake can effectively counteract early functional impairments resulting from neonatal brain injury, even under conditions of prenatal alcohol exposure. These results align with previous reports that neonatal rodents can exhibit significant recovery capacity following hypoxia-ischemia and that certain treatments may enhance this recovery. Moreover, resveratrol has been shown to mitigate some ethanol-induced neurodevelopmental damage in the hippocampus of young mice.
Later-stage sensorimotor performance was assessed using the adhesive removal test, which is typically used to detect lasting somatosensory deficits following stroke. This test, adapted for use in juvenile rodents, confirmed the earlier findings. All groups, with the exception of the ethanol-only group, showed gradual improvement over time. While the resveratrol plus ethanol group displayed modest gains compared to the ethanol group, the piceid plus ethanol group again performed indistinguishably from the uninjured controls. These outcomes strongly support the idea that maternal piceid supplementation can prevent long-term somatosensory deficits, even in the presence of prenatal alcohol exposure. Additional behavioral testing further confirmed this protective effect.
It is well documented that hypoxia-ischemia in human infants can lead to hippocampal injury and long-term memory deficits. Similarly, animal models of neonatal brain injury often show impairments in both short-term and long-term cognitive performance. In our study, long-term memory was assessed using the novel object recognition test. Here again, pups from mothers supplemented with piceid performed at the same level as the control group, suggesting complete protection against both the hypoxic-ischemic insult and the effects of prenatal ethanol exposure. These results echo previous findings demonstrating that resveratrol can prevent cognitive decline when administered directly to pups, although that work did not examine the effects of alcohol. The protective action of resveratrol has been attributed to its ability to preserve key brain regions involved in memory and learning, such as the hippocampus, cortex, and striatum. However, the group exposed to ethanol alone showed the most severe memory deficits, indicating that even moderate maternal alcohol consumption can significantly impair cognitive development. While resveratrol supplementation provided some benefit, it was not as effective as piceid, which fully prevented memory loss.
The neuroprotective potential of resveratrol has often been linked to its antioxidant capabilities, although these effects typically require high concentrations. In the context of neonatal brain injury, resveratrol may act through multiple mechanisms. It is known to activate SIRT1, a key regulator of cell survival and metabolism, and it may also influence energy pathways. Hypoxic-ischemic injury is known to impair glutamate uptake by astrocytes, leading to excitotoxicity. Resveratrol has been shown to reverse this effect and to stimulate glycolysis, thereby increasing lactate production. Lactate, once considered a waste product, is now recognized as a vital energy substrate for the brain. Astrocytes convert glucose to lactate, which is then shuttled to neurons and used for oxidative metabolism, a process known as the astrocyte-neuron lactate shuttle. Providing lactate in this manner has been shown to be neuroprotective, in part by allowing glucose to be used in the pentose phosphate pathway, which is essential for maintaining cellular antioxidant defenses. Resveratrol has been demonstrated to elevate levels of glutathione, a key antioxidant, in both neuronal stem cells and hippocampal tissue under oxidative stress. It also reduces reactive oxygen species in astrocytes. Thus, by promoting glycolysis and lactate production, resveratrol may enhance neuronal survival and preserve brain function.
There are some limitations to our study. One is the absence of a dose-response analysis. However, the use of a nutritionally relevant dose enhances the applicability of our findings to real-world dietary scenarios. Another consideration is that our model examined the effects of chronic moderate alcohol exposure. Yet, binge drinking, particularly during early unrecognized pregnancy, remains prevalent among young adults. Future research should investigate the impact of such acute alcohol episodes on neonatal brain injury outcomes.
In conclusion, while resveratrol supplementation did not fully protect against the brain lesions caused by neonatal hypoxia-ischemia in the context of moderate maternal alcohol consumption, piceid supplementation provided a much more robust effect. Despite the structural similarity between the two compounds, a single additional hydroxyl group gives piceid a significantly higher bioavailability, which may account for its enhanced efficacy. Given the limited research available on piceid compared to resveratrol, our findings strongly support the need for further studies on this promising compound. Taken together, our data indicate that maternal dietary supplementation with piceid, at levels achievable through the daily consumption of one passion fruit, can not only reduce the extent of brain injury but also completely prevent the resulting motor and cognitive deficits in neonates. This suggests that piceid could represent a valuable preventive strategy for at-risk pregnancies, particularly in cases where maternal alcohol consumption may heighten vulnerability to hypoxic-ischemic events.
Experimental Procedures
Study Design
The purpose of this study was to investigate how maternal nutritional supplementation with specific polyphenolic compounds influences the extent of brain injury caused by hypoxic-ischemic (HI) events in neonatal offspring, particularly in a context of moderate maternal alcohol consumption. All experiments involving animals were carried out in strict accordance with the European Communities Council Directive on the protection of animals used for scientific purposes, dated November 24, 1986 (86/609/EEC). The experimental protocols adhered to the ethical standards established by the French Ministry of Agriculture and Forests and were reviewed and approved by the relevant ethical committees under authorization number 9476. The sample size used in the study was determined based on precedent from earlier research to ensure statistical reliability and reproducibility of the results.
Animals
Pregnant Wistar RJ-HAN rats were obtained from Janvier Laboratories located in Le Genest Saint Isle, France. These animals were housed under a 12-hour light and 12-hour dark cycle in a controlled environment, with unrestricted access to standard laboratory chow (A03) and water. The rats were acclimatized for one week prior to delivery to ensure stability and minimize stress.
Model of Hypoxic-Ischemic Brain Injury
To model cerebral HI in neonatal rats, the procedure established by Rice and colleagues and modified in subsequent research was employed. On postnatal day 7 (P7), pups of both sexes and weighing between 14 and 20 grams were anesthetized using isoflurane (4% for induction and 1.5% for maintenance). A midline incision was made under local anesthesia using lidocaine (0.5%) to expose the left common carotid artery, which was then permanently ligated using a 7–0 silk suture. Following surgery, the pups were placed on a heated mattress for a recovery period of 30 minutes to maintain core body temperature at 36 ± 1°C. After recovery, the pups were subjected to a hypoxic environment by being placed in a humidified chamber containing 8% oxygen and 92% nitrogen at 33°C for two hours. Sham-operated control pups were maintained in a similar heated environment but were not subjected to hypoxia. Diffusion magnetic resonance imaging (MRI) was performed at three time points—three hours after ligation (P7), forty-eight hours post-injury (P9), and at postnatal day thirty (P30).
Experimental Groups
The experimental framework consisted of five primary groups. Dams received nutritional supplementation beginning seven days before delivery and continuing through nine days post-delivery. In the ethanol-only (Eth) group, maternal drinking water was supplemented with ethanol at a dose of 0.5 g/kg/day, equivalent to the intake of approximately two glasses of wine per day for a pregnant woman. For the RSV/Eth group, dams received ethanol at the same dose along with resveratrol (RSV) at 0.15 mg/kg/day. In the PIC/Eth group, dams were given ethanol in combination with piceid (PIC) at the same dose of 0.15 mg/kg/day. Control animals in the sham and HI-only (HIC) groups were given maternal drinking water that included 4 µl of absolute ethanol, the amount used to solubilize the polyphenols, resulting in a final ethanol concentration of 0.002%. Supplementary control experiments confirmed that maternal RSV or PIC supplementation alone did not produce adverse outcomes in pups, as confirmed by MRI and behavioral analysis. In these control groups, dams were administered RSV or PIC individually at 0.15 mg/kg/day over the same supplementation period.
Acquisition of In Vivo Longitudinal MRI
Longitudinal MRI scans were acquired at three distinct time points—P7, P9, and P30—using a 4.7 Tesla Biospec 47/50 horizontal system manufactured by Bruker in Ettlingen, Germany. The system was equipped with a 6 cm BG6 gradient coil capable of 1000 mT/m strength. Anesthesia was induced in pups with 4% isoflurane and maintained at a concentration of 1.5 ± 1% throughout the imaging session. Respiratory rates were monitored using a ventral pressure sensor, and body temperature was carefully maintained at 36 ± 1°C using a water-heated MRI bed. T2-weighted anatomical images were acquired using a Rapid Acquisition with Relaxation Enhancement (RARE) sequence, consisting of 20 axial slices of 0.7 mm each, with a TE of 50 ms and a TR of 3000 ms, for a total scan duration of approximately 4 minutes and 48 seconds. Brain lesions were visualized through Diffusion Weighted Imaging (DWI), which included 20 axial slices of 0.7 mm thickness with 30 directions and a b-value of 1347 s/mm². The diffusion-weighted scans had a TE of 24 ms and a TR of 2 seconds, with a total scan duration of approximately 17 minutes and 4 seconds. All imaging data were analyzed using Paravision 6.0.1 software from Bruker BioSpin.
MRI Analysis
Quantitative analysis of lesion volumes was conducted using apparent diffusion coefficient (ADC) maps obtained at P7 and P9, as well as T2-weighted images at P30. The volume of injury was expressed as a percentage of the total brain volume, derived from 20 serial slices per animal. Two regions of interest (ROI) were manually delineated on each slice: one encompassing the injury and the other representing the total brain area. The severity of brain edema was inferred from ADC values measured in affected brain regions at P7. ROIs were drawn within the lesion areas, specifically targeting the cortex and hippocampus, while a circular ROI was applied to the central region of the striatum, located in a separate slice. Analysis was conducted blindly by two independent investigators to minimize bias, with intra-operator variability below 1% and inter-operator variability below 5%.
Behavioral Study
Behavioral tests were conducted to assess functional outcomes following HI. For the righting reflex, carried out on P8, P10, and P12, each pup was placed on its back and timed for how quickly it could return to an upright position. Each animal was subjected to three trials per day, and the mean time was calculated. Somatosensory and motor asymmetries were evaluated from P20 to P24 using the adhesive removal test. After a brief 60-second habituation in an open field, adhesive tapes were placed on the wrists of both forelimbs. Time to contact and remove the adhesive were recorded, with a maximum observation window of 120 seconds per trial. Neurological deficits were quantified using the modified neurological severity score (mNSS) on P24, encompassing motor function, sensory response, beam balance, and reflex tests. Scoring ranged from no impairment (score <1) to severe impairment (score 13–18). To assess long-term cognitive function, the novel object recognition test was performed on P45. Following an initial habituation phase, pups were exposed to two identical objects for two consecutive days, followed by a test phase on day four in which one object was replaced by a novel item. Time spent exploring each object was recorded, and the discrimination index was calculated as the difference in time spent on the new and familiar objects divided by the total exploration time.
Nissl Staining
Histological analysis was conducted on P9 using Nissl staining to identify neuronal death within the penumbra region. Pups were deeply anesthetized and perfused transcardially with phosphate-buffered saline (PBS), followed by paraformaldehyde (PFA 4%) for fixation. Brains were post-fixed overnight in PFA, cryoprotected using a sucrose gradient, and then snap-frozen. Serial sections of 16 µm thickness were stained with 0.5% cresyl violet for five minutes, rinsed, and analyzed using ImageJ software. A minimum of eight sections from at least three brains were analyzed per condition. Entire images were evaluated using the particle analysis function with a minimum threshold of 500 pixels.
Statistical Analysis
All statistical evaluations and graphical representations were conducted using GraphPad Prism version 7.00. Results are presented as mean values with their corresponding standard error of the mean (SEM). Statistical significance across multiple groups was assessed using one-way analysis of variance (ANOVA) followed by Fisher’s Least Significant Difference (LSD) post hoc tests. Two-way ANOVA was employed for repeated-measure analyses such as the righting reflex, novel object recognition, and adhesive removal tests, followed by Fisher’s LSD post hoc analysis and area under the curve (AUC) calculations. Significance was established at p < 0.05. The number of pups per group and the statistical test used were stated in association with each result.
Declaration of Competing Interest
The authors declare that there are no financial or personal conflicts of interest that could have influenced the conduct or reporting of this research.
Acknowledgment
The authors extend their appreciation to Macha Nikolski, head of the Bordeaux Bioinformatics Center, for valuable input and assistance with statistical analyses and interpretation.
Sources of Funding
This study was supported by grants numbered 2016/01 and 2018/01 from the Fondation de Recherche en Alcoologie (FRA), as well as funding from the French State’s “Investments for the Future” initiative through the IdEx program (ANR-10-IDEX), the LabEx TRAIL project (ANR-10-LABX-57), and a French-Swiss collaborative grant (ANR-15-CE37-0012).