Developmental Brain Injury and Social Determinants of Health: Opportunities to Combine Preclinical Models for Mechanistic Insights into Recovery

Epidemiological studies show that social determinants of health are among the strongest factors associated with developmental outcomes after prenatal and perinatal brain injuries, even when controlling for the severity of the initial injury. Elevated socioeconomic status and a higher level of parental education correlate with improved neurologic function after premature birth. Conversely, children experiencing early life adversity have worse outcomes after developmental brain injuries. Animal models have provided vital insight into mechanisms perturbed by developmental brain injuries, which have indicated directions for novel therapeutics or interventions. Animal models have also been used to learn how social environments affect brain maturation through enriched environments and early adverse conditions. We recognize animal models cannot fully recapitulate human social circumstances. However, we posit that mechanistic studies combining models of developmental brain injuries and early life social environments will provide insight into pathways important for recovery. Some studies combining enriched environments with neonatal hypoxic injury models have shown improvements in developmental outcomes, but further studies are needed to understand the mechanisms underlying these improvements. By contrast, there have been more limited studies of the effects of adverse conditions on developmental brain injury extent and recovery. Uncovering the biological underpinnings for early life social experiences has translational relevance, enabling the development of novel strategies to improve outcomes through lifelong treatment. With the emergence of new technologies to analyze subtle molecular and behavioral phenotypes, here we discuss the opportunities for combining animal models of developmental brain injury with social construct models to deconvolute the complex interactions between injury, recovery, and social inequity.

social circumstances.However, we posit that mechanistic studies combining models of developmental brain injuries and early life social environments will provide insight into pathways important for recovery.Some studies combining enriched environments with neonatal hypoxic injury models have shown improvements in developmental outcomes, but further studies are needed to understand the mechanisms underlying these improvements.By contrast, there have been more limited studies of the effects of adverse conditions on developmental brain injury extent and recovery.Uncovering the biological underpinnings for early life social experiences has translational relevance, enabling the development of novel strategies to improve outcomes through lifelong treatment.With the emergence of new technologies to analyze subtle molecular and behavioral phenotypes, here we discuss the opportunities for combining animal models of developmental brain injury with social construct models to deconvolute the complex interactions between injury, recovery, and social inequity.

Introduction
Worldwide, millions of children each year are born with an acquired prenatal or perinatal brain injury, making these insults the most common causes of lifelong neurodevelopmental morbidity and mortality [1,2].The causes of these injuries are variable, including hypoxia, inflammation, hydrocephalus, ischemic stroke, and hemorrhage [3].In addition, the spectrum of outcomes in affected children from these injuries is broad, ranging from normal developmental outcomes to severe neurodevelopmental disabilities [3][4][5].We have limited interventions at the time of birth, such as therapeutic hypothermia in children with acute hypoxic-ischemic encephalopathy (HIE) [6][7][8].However, hypothermia is not efficacious in low and middle-income countries (HELIX Trial), and even in high-income countries, a third of children still have moderate or severe impairment after hypothermia [8,9].After the neonatal period, we lack targeted therapies to improve neurodevelopmental outcomes.
A child's neurodevelopmental outcome from an early life brain injury is determined by a complex intersection of parameters that each have a range of variation: (1) normal brain development, (2) the extent of the insult, (3) the susceptibility to the insult, and (4) the recovery process (Fig. 1).Fundamental biological factors affecting all of these parameters include the type of insult, gestational age at the time of insult, sex, and gene-environment interactions.For instance, individuals vary markedly for brain growth throughout their lifespan, with regionspecific growth being associated with sex [10].In addition, individuals will have different brain regional susceptibility to the same injury associated with age, sex, and genetics [3,11,12].Biological factors are readily amenable to mechanistic studies in preclinical models either through altering the timing of the insult or using genetic models, especially in mice.As extensively reviewed elsewhere, several important mechanisms regulating outcomes from prenatal injuries have been elucidated by focusing on these biological determinants of outcomes [3,[13][14][15].
However, epidemiological studies have shown that non-biological factors, such as social determinants of health (SDH), are the most significant drivers that dictate outcomes after acquired developmental brain injury [16][17][18][19][20].The Center for Disease Control in the USA broadly defines SDH as the "conditions in the places where people live, learn, work, and play that affect a wide range of health and quality-of-life risks and outcomes" [21].SDH include socioeconomic status, parental education levels, race, and parental presence (Fig. 1).While factors like race and ethnicity are social rather than biological constructs, they can have adverse biological effects via structural and institutionalized racism and inequities.Unfortunately, there is a gap in our understanding of the biological mechanisms by which early social experiences impact recovery from developmental brain injury.Elucidating these mechanisms will enable us to develop therapeutic interventions that can be deployed throughout the lifespan to improve neurodevelopmental outcomes from brain injury.
An enormous body of work has studied the impact of early life environments on rodent and primate behavior models.Animal models allow for some control of genetic and environmental variability and are tractable for mechanistic studies of both social environment and developmental brain injury.Armed with this understanding, the field is now poised to combine models of early enrichment or early life adversity with models of developmental brain injuries to elucidate the molecular mechanisms through which social environments modulate outcomes.In this review, we (1) briefly summarize the current understanding of the biological consequences of early life environment on developmental outcomes, (2) provide a sampling of preclinical models for studying developmental brain injury and early life adversity, (3) offer examples of recent studies combining environmental manipulations with developmental brain injury, and (4) highlight opportunities and challenges for studying the effects of social environment on acquired prenatal and perinatal brain injury.We believe that future mechanistic studies combining developmental brain injury models with early social environment paradigms will uncover novel pathways that can be readily translated to therapies that improve neurodevelopmental outcomes in children.

Evidence for Biological Effects of SDH in Children
SDH are inextricably linked to complex wealth, access, and race issues.The effect of SDH on overall health is evident for almost every medical condition and is unmistakably associated with differences in developmental outcomes in children.For example, a family's socioeconomic status strongly correlates to a child's literacy rates and IQ scores [22].Most strikingly, the gap between children in different socioeconomic groups starts early and widens over time, supporting an additive and continuous effect of the environment as children grow [23,24].The consequences of socioeconomic status have structural and functional effects on the developing brain, presumably due to the sequelae of the social circumstances on neuroplasticity.Lower parental education is associated with higher hair cortisol values in children and decreased volume of hippocampal CA3 and dentate gyrus [25].Children in lower socioeconomic tiers are most severely affected.In fact, relatively small differences in resources in these families have a more significant difference in brain volumes compared to similar slight differences in children at higher socioeconomic tiers [26].Moreover, memory and EEG power in young children are correlated with socioeconomic status, even though there are no differences in EEG parameters at birth [27].
The effects of SDH are potentially even more consequential in children with a developmental brain injury.For instance, lower socioeconomic status and racial identification as "non-white," due to institutionalized racism, increase the risk for pregnancy complications such as prematurity [28][29][30].Race impacts language scores after extreme prematurity, with white children scoring higher than black or Hispanic children, even after adjusting for medical and psychosocial factors [31].Parental education level has one of the strongest associations with long-term developmental outcomes in very preterm neonates, matching the importance of the presence or absence of brain injury [16,18].The outcomes after a severe brain injury are attenuated by high parental education, suggesting environmental interventions exist that can universally improve outcomes for children [16,18].While these improvements are likely mediated through mechanisms such as better access to medical appointments and environmental enrichments, the direct effects of parental education may also play a role.There is ongoing research as to whether education for parents on interventions focused on improving childhood attachment can ameliorate some effects of childhood toxic stress [32].
SDH have many potential biological targets, with a common theme that these changes are associated with increased stress levels and disruption of normal DNA methylation patterns (more comprehensively reviewed in [33][34][35]).For example, poverty's chronic stress correlates with global metabolic dysregulation [36,37].The extensive literature further suggests that institutionalized racism is associated with elevated cortisol levels [38,39].Poverty and race have also been associated with increased levels of global DNA methylation consistent with advanced epigenetic age [40][41][42].Advanced epigenetic age Prenatal or perinatal insults lead to a wide spectrum of neurodevelopmental outcomes.Starting in the prenatal period, outcomes from these injuries are modulated by many biological and social determinants of health (SDH).Many preclinical studies have extensively probed mechanisms underlying the biological determinants of outcomes through use of different insult paradigms, studying specific cell types, and exploring gene-environment interactions.However, human epidemiological data show that SDH have some of the most significant effects on neurodevelopmental outcomes.Some of the pathways implicated in both biological and SDH are epigenetics, metabolism, neuroplasticity, and the stress response through the hypothalamic-pituitary axis (HPA).There is a paucity of preclinical studies that address the interplay between social environment and early, acquired developmental brain injury.Created with BioRender.com.

Intersection of Social Determinants of
Health and Developmental Brain Injury is associated with increased morbidity and mortality risk for obesity and cardiovascular disease, suggesting it is a relevant biomarker of health [43,44].Outstanding work to understand transcriptional changes includes the finding that early life stress confers risk to adult depressionlike behaviors in mice [45].These changes can be reversed or accentuated through the respective overexpression or knockdown of the transcription factor Otx2 in the ventral tegmental area [45].However, how these findings link to epigenetic aging findings or the increased cortisol levels present in early stress remains unclear.

A Sampling of Animal Models for Developmental Brain Injury
The many excellent models of developmental brain injury have been thoughtfully reviewed elsewhere [46][47][48][49][50].Here we highlight a few animal models of perinatal brain injury, focusing primarily on models of hypoxia or hypoxia-ischemia.Our goal for this section is to provide a brief overview of the types of models and discuss relevant considerations for selecting models to be combined with the preclinical early life social environment models.
An important consideration is the timing of an insult.Neuroanatomically, the rodent brain is considered "human late gestation" to "term" equivalent at postnatal day 7-10 (P7 or P10) [51].Thus, one of the most common models of early life brain injury is a postnatal model of HIE developed 40 years ago, known as the Vannucci model [52].Initially developed in the P7 rat, it has been adapted and widely used in both rats and mice.Although there have been many variations, the core of this model involves ligating one common carotid artery under brief anesthesia followed by exposure to hypoxia [52,53].The timing of the insult has been changed to P9 or P10 to approximate neuroanatomic injury to the "term" equivalent brain [54,55].Seizures occur after hypoxia-ischemia in the Vannucci model, similar to what is seen in children after HIE; thus, the Vannucci model is also a valuable model for studying the onset of epilepsy [50].
After injury in the Vannucci model, animals demonstrate sex-dichotomous outcomes, including lesion size, response to therapeutic hypothermia, and behavioral outcomes [56][57][58].This model is known for producing lesions of variable size.However, the variation can actually be harnessed to enable an analysis of results across multiple dependent measures for individual animals instead of solely comparing means on each measure.With sufficient group sizes, this model can adequately assess the effects of interventions [58].
While the modified Vannucci model is intended to represent term gestation perinatal HIE, variations on this timing can be used to model other developmental brain injuries.A similar procedure at an earlier age can model brain injuries of prematurity, including studying oligodendrocyte precursor injury in periventricular leukomalacia [49].These models may involve both ischemia and hypoxia at P3 or a more severe hypoxia exposure without ischemia; these have been used in rodents and sheep [59,60].Combining hypoxia-ischemia with inflammation can yield different injury patterns in myelination, axon integrity, and gait deficits [61,62].
In addition to hypoxia-ischemia models, there are also models that alter oxygenation levels alone to study longterm deficits.For example, there are sex-dichotomous motor deficits in rats at P7-8 after severe acute hypoxia (12 min to 0% inspired oxygen) [63].Models of chronic hypoxia rearing can be used to study encephalopathy of prematurity [47,64].Conversely, there are models of continuous hyperoxia and alternating hyperoxia and hypoxia starting in the newborn period that have demonstrated brain injury and cognitive deficits [65,66].
Prenatal models of brain injury are also important to consider.First, postnatal models cannot study the contributions of the maternal-placental unit to neonatal brain injuries, which limits our understanding of the factors regulating the initial response to injury.Multiple antenatal conditions increase risk for prenatal brain injury, including obesity or smoking [67,68].Placental pathology associated with inflammation and vascular insufficiency correlates to prematurity and HIE [69,70].Additionally, functional networks do not correlate with rodent neuroanatomical maturation [71].For example, mice do not have the respiratory or feeding dyscoordination at birth that is characteristic of children born before 34 weeks of gestation, and by P10, mice can walk, which is a developmental milestone acquired in infants at about 12 months of age [71][72][73].Lastly, as discussed below, several rodent models of neonatal stress are performed in the first week of life, so combining prenatal injury models with SDH models may be more technically feasible.
There are multiple models of prenatal hypoxia.Chronic hypoxia can be studied by having dams live in mild hypoxia (~10% inspired oxygen) throughout gestation [74,75].Transient, relatively mild insults in sheep and mice have also been studied [76][77][78].In sheep, transient prenatal hypoxia leads to long-term structural and functional injury despite the absence of cell death, including deficits in neuron dendritic complexity and action potential propagation [79,80].In mice, transient late gestation hypoxia (5% inspired oxygen at embryonic day 17.5 [E17.5])does not lead to increased cell death [78].However, in adult mice exposed to transient late gestation hypoxia, there is a decreased seizure threshold and sex-dichotomous behavior deficits in anxiety and social interactions [78].In addition, there are deficits in neuronal migration in a model of early preterm injury by transient intrauterine artery ligation at E16.5 [81].
Finally, almost all toxins can be studied in prenatal injury.Maternal immune activation or in utero lipopolysaccharide causes persistent behavioral deficits in adult animals [82][83][84].While not toxins per se, intrauterine exposure to substances with abuse liability (alcohol, nicotine, opioids) is readily modeled in rodents and can be combined with postnatal models [85][86][87].Clinically, perinatal exposure of infants to opioids (including methadone) can have profound and lasting negative effects on the central nervous system via altered inflammatory and neuroinflammatory trajectories [88,89].This negative neuroinflammatory impact is also seen in rodent models of perinatal opioid exposure [90][91][92].In fact, given that brain injury itself leads to sustained peripheral immune hyperactivity, opioid-induced dysregulation of the neuroimmune and neurodevelopmental trajectories is a major consideration in the context of early life adversity, as discussed further later in this piece [93].

Effects of an Enriched Environment on Developmental Brain Injury
Enriched environment paradigms in preclinical animal models consist of an increased cage size with several different toys that rotate every few days to stimulate curiosity of the animals [94].Animals exposed to an enriched environment have improved learning, memory, and decreased anxiety levels even without an initial neonatal injury [95][96][97].There have been more extensive studies on the effects of an enriched environment on perinatal brain injury than on the effects of adverse social environments.For instance, glucose metabolism was attenuated in a mouse model of hypoxia-ischemia at P3 by positron emission tomography [98].Environmental enrichment during gestation before the injury or during lactation after weaning prevented severe structural and metabolic injury [98].These exciting findings suggest that an enriched environment could both lead to a more resilient brain that can better tolerate insults as well as significantly enhance the recovery process.In a P10 model for HIE, an enriched environment after the injury improved novel object recognition and increased hippocampal volumes [99].Notably, while enriched environments have been shown to upregulate brain-derived neurotrophic factor (Bdnf) in the absence of injury, an enriched environment did not increase Bdnf in the setting of postnatal hypoxia-ischemia in rodents [99,100].
In a model of chronic perinatal hypoxia (P3-P11), environmental enrichment also improves motor outcomes in adult animals [101].In addition, enriched environments increase neurogenesis after injury [101,102].Some of these improvements were thought to be due to mitigating white matter injury from chronic hypoxia [101].In chronic hypoxia, the loss of astroglial progenitors is abrogated by an enriched environment [102].Notably, improvements in neurogenesis, white matter injury, and behavior required a sustained enriched environment from P15-45 for maximizing the impact of an enriched environment on neuronal structural and behavior changes [101].These findings support the idea that enriched environments improve outcomes but suggest that any interventions to improve injury must be sustained in the postnatal period, possibly through adulthood.
In summary, these studies have demonstrated that an enriched environment can mitigate the risks of structural and functional brain injury from various perinatal insults, but several questions remain.First, if enriched environments improve structure and function without increasing Bdnf levels, what alternate pathways are activated by the enriched environments that mitigate injury?Second, the lasting effects of an enriched environment after hypoxia require sustained treatment starting early after the injury.However, environmental enrichment starting 4 weeks postnatally also improved deficits in neuronal hippocampal spine density and autism-related behaviors from prenatal treatment with valproic acid at E12.5, supporting that this intervention can still successfully improve outcomes even after remote injury [103].These studies raise questions as to why effects might not be sustained after the initial injury and whether there are windows of neuroplasticity during development that should be the focus of interventions.These different responses to environmental enrichment after different types and timings of insults also raise the question of whether enriched environments have unique effects on various developmental injuries.Consistent with the enriched environment improving outcomes, exercise in juvenile animals has also been shown to improve outcomes [104].Early life exercise may be an alternative intervention that can be more thoughtfully implemented in children.Thus, further mechanistic studies into the molecular mechanisms underlying these changes from enriched environments or exercise in rodents may be translatable to novel interventions in children affected by acquired developmental brain injury that can be implemented throughout their lifespan.

Models of Early Life Adversity
Several animal models can be used to study the conditions of early life adversity (Table 1).While no animal model can replicate the complexity of human experience, these models have good ecological validity for representing the conditions they are intended to study.
One well-known rodent model of early life adversity is maternal separation, which mirrors early caregiver deprivation.Initially developed in rats by Plotsky and Meaney, this model has now been adapted in many ways [105].Briefly, control "handled" rats undergo brief 15-min separation daily from P2-14.First, the dam is removed to an adjacent chamber, and then the pups are removed from the home cage and placed in an alternate chamber.More recent protocols call for pups to be placed in an incubator to ensure they stay warm [106].This brief period is within the typical length of time that the dam may be away from the nest during routine care [107].The "maternal separation" group experiences the same procedure, but the separation length is 180 min (about 3 h) instead of 15 min.After maternal separation, offspring have altered stress responses, abnormal behavioral responses to stressors, increased depression-like behavior, and decreased working memory [105,106,108].Notably, some work describes sex differences in the offspring response, with female subjects often being less sensitive to maternal separation stress than male subjects [109].
As an alternative to maternal separation, a recent paper modeled paternal deprivation compared to control mice receiving biparental care [110].They found that paternal deprivation reduced dentate gyrus volume in all offspring but altered microglial cell density only in the female, paternally deprived offspring.These results are thought provoking considering that many experiments across most laboratories rear rodent pups with the dam alone (i.e., paternal deprivation is the standard).While the parental separation models have been instrumental in elucidating some mechanisms by which early life adversity causes later neurodevelopmental impairment, these models utilize intermittent episodes of stress.Notably, some dams are known to lavish attention on their pups upon reunion after the separation period; this increased nurturing may mitigate some of the effects of the separation period.Thus, there is more to be learned from using chronic stress models.An excellent model that provides chronic constant stress is the limited bedding and nesting model, used as an analog for chronic poverty and neglect.Initially developed in rats, this model is also valid in mice [111,112].Briefly, from P2 to P9, dams and pups are placed in housing with a small amount of bedding below a metal grate and given about half of the standard nesting material.Dams with limited nesting material provide fragmented and inconsistent maternal nurturing behaviors such as licking, grooming, and nursing [113].These fragmented or unpredictable maternal behaviors lead to multiple outcomes for offspring: altered stress-response system, poorer cognitive performance, and altered hippocampal neurogenesis [112,[114][115][116]. Minimal nesting material from P8-12 still results in the dam spending more time away from the nest, disruption of nurturing behaviors, leading to increased abnormal vocalizations in the pups and depressive-like behavior in adolescent offspring [117].Finally, it yields decreased developmental competence as indicated by disrupted nipple attachment, distress vocalization when in physical contact with an anesthetized mother, and reduced preference for maternal odor with corresponding changes in brain activation [118].
These animal models of early caregiver deprivation or limited bedding are intended to model poverty [113,119].The degree of poverty is perhaps most applicable to understanding the biological consequences of family leave policies on outcomes for maternal and child health more than the utilization of non-parental caregivers (i.e., daycare or paid care) in infancy.By contrast to Canada and many countries in Europe, the USA only requires that employers provide 12 weeks of unpaid maternity leave for employees and there are no mandates about paternal leave [120].The lack of paid leave guidelines has fostered a form of institutionalized inequity that exacerbates the effects of structural racism [121].A recent study demonstrated that only 16% of birth parents have access to paid leave and about a quarter return to work within 10 days because of the inability to sustain living expenses without income [122]."No" and "early leave" were associated with poor maternal and child health outcomes, including decreased maternal-infant attachment and later childhood academic achievement [122].Of note, the NICHD Study of Early Child Care demonstrated that differences in childhood outcomes correlated to family socioeconomic factors and not whether the child was in care outside the home or by a non-parent caregiver, like in a daycare setting [123].
While the models described above use postnatal stressors, there are also animal models where the adverse event occurs during gestation.One such model involves repeated restraint stress: placing a pregnant mouse dam in a physically confined space for 2 h on each gestational day 10 through 16.Male offspring show decreased social behavior in adulthood, decreased cortical oxytocin receptor and serotonin metabolism in adulthood, and altered gut microbiome, while female offspring show increased anxiety-like behavior and alterations in cognition as well as decreased Bdnf in their amygdala as adults [124,125].It would be interesting to determine if studies combining early life stressors with developmental brain injury would have the same effects on Bndf levels since the parallel enriched environment experiments showed that impact on behavior outcomes after injury was independent of Bdnf [99,100].Additional models of early life adversity, including fetal undernutrition, are also useful prenatal models [126,127].
As mentioned above, clinical and preclinical research suggest perinatal exposure to drugs with abuse liability, such as alcohol, smoking, and opioids, can negatively impact neurodevelopmental trajectory in part via the dysregulation of neuroimmune parameters [128][129][130].Work has already shown that the stressor of gestational or early life exposure to opioids with early life brain injury results in a higher incidence of brain injury [131].As even opioids used clinically in critically ill neonates for sedation and pain have significant effects on brain growth and development, more work is needed to understand the extent of how drugs with abuse liability influence subsequent response to injury [132,133].

Discussion
The goal of this review is not to be comprehensive with all models of early life brain injury or adversity but rather to provide a sampling of the injury and early life Intersection of Social Determinants of Health and Developmental Brain Injury environment models to illustrate how they can be combined.Understanding the biological underpinnings/ mechanisms that modulate early life brain injury outcomes is vital to developing novel, targeted intervention strategies to improve neurodevelopmental outcomes.Critically, these pathways may be lifelong targets that allow us to expand treatment windows of neuroprotection beyond the neonatal period.
Importantly, there is already evidence that social interventions benefit neurologic function in healthy children, suggesting modulating the social environment may also help children with developmental brain injuries.For example, a compelling study provided a fixed sum of unrestricted funds to families with an infant at risk for poverty and assessed neurologic function [134].EEG power was similar to EEG patterns previously observed in children not in poverty [27,134].This study signals that there may be molecular pathways affected by social constructs that are dynamic and that the adverse effects of social inequity are reversible.
Many of the models of adverse social environment are employed in the first week of life.Therefore, studying the combination of environment and injury may be more feasible using prenatal rather than postnatal models of injury (Fig. 2).There is substantial variability independently in injury models and social paradigms, suggesting these experiments will require careful consideration of effect size to be appropriately powered.There are additional limitations to using animals to explore complex social environments.Combining models entails using more variables, and these are large studies comparing at least 4 groups.There is great importance in noting experimental details, including the sex of the experimenter, litter composition, breeding in-house versus shipping pregnant dams, within-litter analysis, sex differences, and paternal involvement in pup rearing [135][136][137][138][139][140][141].Experimenters can consider multilinear regression statistical approaches to compare results for individual animals across multiple outcomes rather than comparing the means of each outcome [142].Furthermore, depositing protocols into open-source platforms (platforms.io) may help with the reproducibility and rigor issues that plague preclinical modeling [143][144][145][146]. Additionally, novel experimental modalities, including single-cell sequencing or machine learning through deep neural networks on behavior, promise to allow us to deconvolute the complex effects of injury and social environment on specific cell types or subtle but important behavior changes [147][148][149][150].
With the growing societal recognition that social inequity dictates health outcomes, we have a responsibility as basic scientists to contribute to a better understanding of the biological effects of non-biological determinants of health.For instance, extremely low birth weight was associated with advanced epigenetic age in adults [151].As poverty and race are also associated with advanced epigenetic age, it is unclear what the biological consequences of combining the effects of injury and social adversity are on neurodevelopmental outcomes [40][41][42].Furthermore, the need to consider issues of inequity in developing therapies was highlighted with the HELIX trial, where therapeutic hypothermia did not improve outcomes in low-and middle-income countries [9].In retrospect, it is a provocative but essential question whether models of prenatal social adversity combined with hypoxic-ischemic models may have predicted that this intervention may have had a more limited effect in lower resource environments.Therefore, as basic and translational scientists, we cannot afford to sideline ourselves because of the complexity of these modelswe may be missing opportunities to develop and screen therapies that can make a significant lifelong impact on the most vulnerable children.

Fig. 1 .
Fig. 1.Many factors affect neurodevelopmental outcomes throughout the lifespan.Prenatal or perinatal insults lead to a wide spectrum of neurodevelopmental outcomes.Starting in the prenatal period, outcomes from these injuries are modulated by many biological and social determinants of health (SDH).Many preclinical studies have extensively probed mechanisms underlying the biological determinants of outcomes through use of different insult paradigms, studying specific cell types, and exploring gene-environment interactions.However, human epidemiological data show that SDH have some of the most significant effects on neurodevelopmental outcomes.Some of the pathways implicated in both biological and SDH are epigenetics, metabolism, neuroplasticity, and the stress response through the hypothalamic-pituitary axis (HPA).There is a paucity of preclinical studies that address the interplay between social environment and early, acquired developmental brain injury.Created with BioRender.com.

Fig. 2 .
Fig.2.Example schematic of experimental design for studying the effect of adverse or enriched environment on the course of developmental brain injury.We propose combining animal models of prenatal or perinatal insults with animal models of early life adversity or environmental enrichment to uncover mechanisms by which environments impact neurodevelopmental outcomes and recovery after early life brain injury.Prenatal insults, such as prenatal hypoxia or inflammation, are particularly amenable to studying the effects of different social experimental paradigms on outcomes because many models of social stress are implemented in the first days of life.This experimental design would allow the investigation of a wide range of long-term consequences, ranging from juvenile and adult behavioral testing to molecular mechanisms.Created with BioRender.com.

Table 1 .
Preclinical models of early life adversity: examples and outcomes from select studies