香京julia种子在线播放

Swiss-Webster mice were purchased from Charles River (Wilmington, MA) and bred at Ohio State University (OSU). Pups were weaned at postnatal day (PND) 21 and housed in standard mouse cages under a 14:10 light-dark cycle. Mice were provided with access to food and water ad libitum, and all experiments were performed in accordance with approval from the OSU Institutional Animal Care and Use Committee.

To measure organizational effects of sex steroids, neonatal pups underwent hormone manipulations. Female mice were injected subcutaneously with testosterone, 100mg (per pup) crystalline testosterone (Sigma Aldrich) dissolved in 100 μL olive oil (Thermo Scientific) or oil alone on PND 4 (Klein et al., 2002). Male mice were gonadectomized (GDX) under cryoanesthesia at PND 4 or underwent a sham operation (Anderson et al., 2005). Briefly, the skin was disinfected with alternating swabs of betadine and 70% ethanol, a vertical midline incision was made, and the testes were removed with forceps. The incision was sutured, and pups were returned to their home cage on a heating pad. The sham procedure included an incision but no manipulation of the gonads. To measure activational effects of sex steroids, mice underwent GDX or control procedures at PND 60 as previously reported (Aubrecht et al., 2015) producing the following groups: females (oil + control “control”, oil + GDX “GDX”, T + GDX), males (control, neonatal GDX, adult GDX). Note that to control for age differences at the time of surgery, all male mice underwent two surgical procedures (at PND 4 and PND 60), undergoing either two control surgeries, or one control and one GDX. All mice were randomly assigned to each group.

A mild closed head injury (or sham procedure) was performed on mice at PND 21. Mice were anesthetized with isoflurane at 3% for induction and 1.5% during the operation. The injury was performed using an Impact One device (Leica Biosystems, Richmond, IL) while the head was stabilized in the stereotaxic frame. Bupivacaine (1mg/kg) was injected SC at the incision site, the skull was exposed, and the device impactor was placed on its surface at –1 AP and 1 ML relative to Bregma. The 2mm diameter tip of the impactor was then depressed 1mm into the skull at 3m/sec with a 30msec dwell time. The process was the same for sham mice, excluding the depression of the plunger. Mice were then returned to their homecage, and additional endocrine manipulations were performed at PND 60 as discussed above. The following groups were generated: Females (control/sham n = 9, control/TBI n = 9, GDX/sham n = 8, GDX/TBI n = 8, T + GDX/sham n = 9, T + GDX/TBI n = 10), Males (control/sham n = 7, control/TBI n = 7, neonatal GDX/sham n = 7, neonatal GDX/TBI n = 7, adult GDX/sham n = 8, adult GDX/TBI n = 8). See Figure 1A for the timeline of hormonal manipulations, TBI (or sham surgery), and behavioral analyses.

The ethanol CPP protocol (adapted from O’Neill et al., 2013; Cunningham, 2014) beginning PND 80 as follows. A plastic container held two chambers of different tactile (thick and thin grid) and visual patterns (checkered and lined) that were separated by a barrier with a small passageway. On the first day, mice were injected intraperitoneally with saline (10mL/kg) and allowed to cross freely between the chambers for 5 min to allow the mice to habituate and determine baseline side bias. Treatment/cue pairings were counterbalanced and randomly assigned. For days 2, 4, 6, and 8, mice were injected with a 20% ethanol solution (2 g/kg), returned to their home cages for 5 min, then confined to an assigned side of the container for 5 min. On alternating days (3, 5, 7, and 9), mice were injected with saline and confined to the opposite-patterned chamber for 5 min. Finally, on the 10th day, mice were injected with saline and allowed to travel freely between the chambers. Conditioned place preference was determined by assessing the amount of time spent exploring the chamber associated with ethanol on day 10 relative to day 1.

Tissue was collected after transcardial perfusion with 4% paraformaldehyde after mice were overdosed with sodium pentobarbital. The forebrain was sectioned coronally into 40 μm slices. Immunohistochemistry was performed for the microglia-specific protein Iba1 using an antibody purchased from Wako (1:1000 anti-Iba1, rabbit) as previously reported (Karelina et al., 2017). Briefly, tissue was washed with 0.1M phosphate buffered saline, quenched with hydrogen peroxide, and incubated overnight with anti-Iba1 antibody. The next day tissue was incubated with a goat anti-rabbit biotinylated secondary antibody (1:500) and visualized using the ABC-DAB method. To assess axonal degeneration, a silver stain was performed with the FD NeuroSilver™ Kit II from FD Neurotechnologies following the manufacturer’s instructions (Columbia, MD).

Photomicrographs of Iba1 staining were obtained at a 20X magnification (Nikon E800 microscope) and cell counts were conducted using FIJI (Schindelin et al., 2012). Cell counts were obtained from defined regions of interest (ROI) (prefrontal cortex, ROI = 0.04 mm²; amygdala, ROI = 0.04 mm²; nucleus accumbens, ROI = 0.072 mm²), averaged across the hemispheres, and reported as cells per mm². Axonal degeneration was observed based on silver staining in the white matter tracts of the forebrain. The silver score was determined as previously reported qualitatively using a point-value system: such that 0 = little to no axonal degeneration, 1 = sparse silver staining limited to the corpus callosum, 2 = moderate silver staining in the corpus callosum and other white matter tracts, and 3 = very dense silver staining throughout multiple white matter tracts (Karelina et al., 2021). All analyses were performed blind to experimental conditions.

Statistical analysis was performed using SPSS Version 26 (IBM Corp., Armonk, NY, United States). CPP responses, silver staining, and microglial immunohistochemistry were assessed separately for males and females via a two-way ANOVA (injury x endocrine manipulation). Reproductive tissue and body masses were assessed via a two-way ANOVA (injury x endocrine manipulation). Significant overall ANOVA results were followed up by a one-way ANOVA (factor = injury) between specified groups for CPP, or a Tukey HSD post hoc test for reproductive tissue and body mass. Effect sizes are reported as partial eta squared (η_p²). Results are considered significant when p ≤ 0.05.

To determine the role of circulating sex steroids on alcohol-related behavior after brain injury, we manipulated gonadal steroid availability at critical developmental periods (perinatally and at adolescence). The effectiveness of our endocrine manipulations was verified by assessing reproductive tissue masses at necropsy (see Figure 1 for time line). Endocrine manipulations altered tissue masses, but injury did not (p > 0.5 for all sham vs. TBI comparisons). Uterine mass was reduced in GDX and T + GDX females compared to control mice (F_2,51 = 34.74, P < 0.00001, η_p² = 0.591), ovarian fat pad mass was also reduced but not significantly (F_2.50 = 20.154, P = 0.14, η_p² = 0.127). Body mass was also altered by endocrine manipulations (F_2,52 = 3.141, P = 0.05, η_p² = 0.114) such that GDX females were heavier than controls. For males, epididymides (F_{2, 43} = 13.874, P < 0.00001, η_p² = 0.41) and seminal vesicle masses (F_2,42 = 48.985, p < 0.00001, η_p² = 0.715) were reduced in NEO GDX and Adult GDX groups. Body mass was altered by endocrine manipulations (F_2,41 = 6.064, p = 0.005, η_p² = 0.228) such that NEO GDX male mice were lighter than intact mice.

An overall ANOVA including all of the sex steroid manipulations revealed no significant sex differences in CPP responses to ethanol (all p > 0.05), however this is not surprising given the sex-specific variability in both CPP after TBI, and the substantial difference in surgical procedures between male and female neonates (T/oil injections in females vs. GDX/Sham surgery in males). To confirm that we have replicated our previously reported findings (Weil et al., 2016b), we conducted an ANOVA to compare CPP responses among control (no sex steroid manipulations) male and female mice, and reveal that control female mice have significantly greater CPP responses to ethanol after TBI compared to intact males (F_1,28 = 6.022, p = 0.022, η_p² = 0.201). CPP behavior in males and females varied in response to gonadal manipulations and TBI. Among females there was no overall effect of injury (F_1,39 = 0.335, p > 0.05, η_p² = 0.009, Figure 2A) or of endocrine manipulations (F_2,39 = 0.574, p > 0.05, η_p² = 0.029). However, there was a significant interaction (F_2,39 = 4.19, p = 0.022, η_p² = 0.177) such that injury increased CPP responses among control (gonadal unmanipulated) females (F_1,13 = 5.756, p = 0.034, η_p² = 0.324) but had no effect in any other group (P > 0.05 in all cases). Injured females that were treated with T perinatally and then gonadectomized exhibited significantly lower CPP responses than did intact females (F_1,15 = 6.414, p = 0.025, η_p² = 0.330). In contrast, males exhibited no effects of injury (F_1,36 = 0.338, p > 0.05, η_p² = 0.009, Figure 2B), endocrine manipulation (F_2,36 = 1.371, p > 0.05, η_p² = 0.071), or interaction between the two variables (F_2,36 = 1.976, p > 0.05, η_p² = 0.099).

Silver staining was used as indicator of axonal degeneration such that a higher silver score indicated greater axonal degeneration. Injured females (F_1,43 = 33.16, p < 0.0001, η_p² = 0.435) and males (F_1,35 = 20.55, p < 0.0001, η_p² = 0.37) exhibited greater evidence of axonal degeneration, as assessed by silver staining than did sham-injured mice (Figure 3). But there was no overall significant sex difference or significant effect of endocrine manipulations among either sex (p > 0.05 in all cases) or interactions between the variables. We next examined microglial cell counts in regions of the mesolimbic pathway and prefrontal cortex (PFC), given the known relationship between activity in the amygdala, nucleus accumbens, and PFC and alcohol use disorder (Abernathy et al., 2010; Roberto et al., 2012; Seif et al., 2013; Grodin et al., 2018). Interestingly, microglial cell numbers in key forebrain structures was increased by TBI in injured males but not females (Figure 4). In the amygdala, nucleus accumbens, and PFC of females there were no injury or endocrine manipulation effects, nor were there interactions between the variables (P > 0.05 in all cases). In contrast, males, regardless of endocrine manipulation have increased microglial cell counts throughout the forebrain. For instance, in the amygdala (F_1,21 = 5.404, p = 0.35, η_p² = 0.265) and prefrontal cortex (F_1,21 = 5.101, p = 0.039, η_p² = 0.254) injured males exhibited significantly greater microglial cell numbers than did sham-injured mice. The nucleus accumbens did not exhibit this effect (F_1,21 = 1.752, p = 0.205, η_p² = 0.105). A direct comparison of microglial cell numbers in TBI mice between the sexes confirmed a sex difference such that males exhibited significantly greater numbers of microglia in the PFC (F_{1, 29} = 8.036, p = 0.009, η_p² = 0.229), and nucleus accumbens (F_1,35 = 6.162, p = 0.018, η_p² = 0.157).

We hypothesized that sex steroids were the mediators of this sex difference for two primary reasons. First, both neonatal testosterone treatment and adult gonadectomy in female rodents have been shown to affect their response to ethanol. For instance, ovariectomized females drink significantly less than other females (Becker et al., 1985) and estradiol administration increases drinking in ovariectomized mice (Ford et al., 2002). Furthermore, defeminization of female rats through neonatal sex steroid exposure reduces alcohol consumption (Almeida et al., 1998). Secondly, the epidemiological data suggest that women are most at risk of exhibiting later alcohol misuse when a TBI is incurred during puberty (Corrigan et al., 2020).

Sexual differentiation of brain and behavior results from the contribution of appropriately timed steroid hormone exposure (or the lack thereof) and differential gene expression associated with sex chromosome complement (Breedlove and Hampson, 2002). In the current, study we treated female mice with androgens during the early postnatal period. The exposure to androgens, which are aromatized into estrogens in the brain, permanently prevents the development of ovarian cyclicity and other female-typic behavioral patterns. We utilized the CPP paradigm as it specifically assesses the hedonic effects of alcohol that we have previously demonstrated were altered by TBI (Weil et al., 2016b). The current study replicates previous findings: intact female mice injured early in life developed a conditioned place preference to ethanol, while the uninjured females did not (Weil et al., 2016b). Treatment with testosterone neonatally and adult gonadectomy abolished the enhanced CPP response to alcohol after TBI. Thus, both female typical sex steroid exposure perinatally and in adulthood are necessary for the TBI-induced potentiation of alcohol reward. In contrast, neither sex steroid manipulations (perinatal or in adulthood), nor TBI was sufficient to enhance alcohol CPP in male mice. Thus, the enhanced CPP response among brain injured females appears to require normal ovarian cyclicity and cannot be replicated by neonatal castration of genetic males.

A separate possibility underlying the disparity in alcohol response between males and females is a difference in injury severity. For instance, some data from clinical populations suggest that women fare worse than men after TBI and exhibit greater delays in recovery (Farace and Alves, 2000). However, given that axonal degeneration is not significantly different between unmanipulated males and females after a TBI, we assert that this is not the cause of the sex difference represented in our model. Moreover, axonal degeneration was affected by injury but not significantly altered by neonatal testosterone treatment or gonadectomy. Since the axon injury scores do not mirror the alcohol-related behavior, these behavioral responses appear to occur independently of injury severity.

We also considered that the variations in alcohol response could be due to differential neuroinflammation, as there is an established link between neuroinflammation and substance abuse (Kohno et al., 2019). Moreover, we previously showed that treatment with an inhibitor of microglial activation reduced post-TBI drinking in male mice (Karelina et al., 2018). There are both sex differences in neuroimmune physiology and effects of steroid hormones on immune function (Villa et al., 2016). We had predicted that greater microglial activity in brain regions related to reward processing might correlate with CPP responses among injured mice. Traumatic brain injuries and other early life perturbations can produce a persistent state of increased reactivity to immune challenges (Fenn et al., 2014). Most drugs of abuse, including alcohol can drive inflammatory signaling (Alfonso-Loeches et al., 2010; Lacagnina et al., 2017). There is a substantial literature linking early life stress and adversity with both alterations in microglial physiology and increased susceptibility to substance abuse in both preclinical and clinical studies (Enoch, 2011; Crews et al., 2017; Pascual et al., 2017; Johnson and Kaffman, 2018). Taken together, we had reasoned that TBI would prime microglial immune responses such that the administration of alcohol would produce exaggerated inflammatory responses. Sex differences in this relationship could therefore mediate the sex differences in CPP responses to alcohol among injured mice gonadectomy, such that microglial count was greater in the PFC and amygdala of males that underwent a TBI. This is in line with existing evidence that the neuroinflammatory response is greater in males after a TBI (Villapol et al., 2017). One limitation of these experiments is that the length of time between injury and tissue collection (2 + months); thus, testosterone treatment or gonadectomy promote acute changes in neuroinflammation that may resolve over time. Additionally, examining microglial cell numbers only provides a relatively limited amount of information as to the reactivity, gene expression, and secreted factors. This finding suggests that although injury promotes a neuroinflammatory response in males, this occurs independently of sex hormones. Furthermore, this finding reduces the possibility that neuroinflammation is directly responsible for the increase in alcohol consumption after a TBI. neuroinflammatory response in males, this occurs independently of sex hormones. Furthermore, this finding reduces the possibility that neuroinflammation is directly responsible for the increase in alcohol consumption after a TBI.

These data have important implications for case management of individuals with a history of childhood traumatic brain injury. Substance abuse prevention and treatment have the potential to produce significantly better outcomes in this patient population because (1) substance abuse impairs rehabilitation and produces poorer overall outcomes in those with a history of brain injuries and (2) alcohol intoxication is a major risk factor for subsequent TBIs which can have devastating permanent consequences (Weil et al., 2016a,2019; Corrigan et al., 2020). Moreover, there is also epidemiological evidence that sex differences exist in the risk of developing substance abuse disorders after brain injury. Humans differ from rodents in that males tend to consume more alcohol whereas the opposite is true among laboratory rodents (Barker et al., 2010). Thus, from a public policy standpoint it will be beneficial to invest in strategies that reduce the risk of developing substance abuse disorders for individuals of both sexes that have experienced TBI. However, these data should highlight the importance of considering the role of sex and sex steroids as potential contributors to substance-abuse related outcomes after brain injuries.