Sex-Specific Alterations in Inflammatory MicroRNAs in Mouse Brain and Bone Marrow CD11b+ Cells Following Traumatic Brain Injury

Sex is a key biological variable in traumatic brain injury (TBI) and plays a significant role in neuroinflammatory responses. However, the molecular mechanisms contributing to this sexually dimorphic neuroinflammatory response remain elusive. Here we describe a significant and previously unreported tissue enrichment and sex-specific alteration of a set of inflammatory microRNAs (miRNAs) in CD11b+ cells of brain and bone marrow isolated from naïve mice as well as mice subjected to TBI. Our data from naïve mice demonstrated that expression levels of miR-146a-5p and miR-150-5p were relatively higher in brain CD11b+ cells, and that miR-155-5p and miR-223-3p were highly enriched in bone marrow CD11b+ cells. Furthermore, while miR-150-5p and miR-155-5p levels were higher in male brain CD11b+ cells, no significant sexual difference was observed for miR-146a-5p and miR-223-3p. However, TBI resulted in sex-specific differential responses of these miRNAs in brain CD11b+ cells. Specifically, miR-223-3p levels in brain CD11b+ cells were markedly elevated in both sexes in response to TBI at 3 and 24 h, with levels in females being significantly higher than males at 24 h. We then focused on analyzing several miR-223-3p targets and inflammation-related marker genes following injury. Corresponding to the greater elevation of miR-223-3p in females, the miR-223-3p targets, TRAF6 and FBXW7 were significantly reduced in females compared to males. Interestingly, anti-inflammatory genes ARG1 and IL4 were higher in females after TBI than in males. These observations suggest miR-223-3p and other inflammatory responsive miRNAs may play a key role in sex-specific neuroinflammatory response following TBI.


Introduction
Neuroinflammation is recognized as a critical factor to acute and chronic secondary injury following traumatic brain injury (TBI). While the acute inflammatory stage is essential for removal of cellular debris and promoting cellular repair, dysregulation of this response leads to chronic microglia activation and pro-inflammatory signaling impacting long-term neurological function (Johnson et al. 2013;Loane et al. 2014;Pierce et al. 1998). This is characterized by the excessive release of pro-inflammatory factors by resident and recruited myeloid cells that differentiate along a continuum from cytotoxic to reparative phenotypes (Jin et al. 2012;Kumar et al. 2016;Morganti et al. 2016). This neuroinflammatory response also is sex-specific and characterized by a differential myeloid cell response in females relative to males (Acaz-Fonseca et al. 2015;Doran et al. 2019;Villapol et al. 2017).
MicroRNAs (miRNAs) are non-coding RNAs that regulate gene expression post-transcriptionally in almost all cellular events, including those controlling sex-specific responses (Kodama et al. 2020;Liu and Xu 2011;Sharma and Eghbali 2014). Numerous studies including ours have reported a strong correlation between altered miRNA expression levels and secondary brain injury events (Liu 1 3 et al. 2014;Redell et al. 2009;Wang et al. 2015). Among these altered miRNAs, miR-146a, miR-223-3p, miR-155-5p, and miR-150 play key roles in inflammatory signaling (Cardoso et al. 2012;Zhao et al. 2011;Johnnidis et al. 2008;Shakerian et al. 2018). However, a mechanistic role for these inflammatory miRNAs in TBI-induced neuroinflammation is poorly understood. Moreover, no studies have revealed the involvement of these miRNAs to sex-specific neuroinflammatory responses.
Cluster of differentiation molecule 11b (CD11b, or Integrin alpha M (ITGAM)) is a leukocyte surface marker highly expressed on monocytes and macrophages of bone marrow as well as on resident CNS microglia (Gordon and Taylor 2005). Following injury, brain CD11b positive (CD11b+) cells represent a specific cellular population consisting of activated resident microglia and infiltrating macrophages. CD11b+ cells play a key role in neuroinflammation and their dynamic changes over time affect the complex inflammatory responses that are beneficial or detrimental to functional recovery (Kumar et al. 2016(Kumar et al. , 2017Truettner et al. 2017). However, the miRNA responses in CD11b+ cells following TBI are not known. Here we report a differential enrichment of these miRNAs in mouse CD11b+ cells from bone marrow and brain, and a time-dependent alteration of these miRNAs following TBI. Our study suggests that the enrichment and alteration of miRNAs in CD11b+ cells of a particular tissue origin may reflect their specific immunomodulatory role in response to TBI.

Animal Studies: Controlled Cortical Impact Injury (CCI)
All animal methodology was in full compliance with the National Institutes of Health (NIH) Guide and with the US Public Health Service Policy on Humane Care and Use of Laboratory Animals and was approved by the University of Kentucky's Institutional Animal Care and Use Committee (protocol #2019-3223). All procedures were conducted in a manner to minimize pain and discomfort of the animals. Young adult (9-10 week old) male and female C57BL/6J mice (RRID:IMSR_JAX:000664, Jackson Laboratories) were group-housed for one week prior to experimentation and kept in a 12/12-h light/dark cycle in a temperaturecontrolled vivarium room with free access to food and water. For the surgical procedures, animals (females: n = 5 mice/sham or CCI group; males: n = 6 mice/sham group, n = 8 mice/CCI group) were randomly assigned to either sham-surgery or controlled cortical impact (TBI) groups. All surgical procedures were performed using sterile techniques published previously including post-surgical recovery (Mbye et al. 2008). Briefly, animals are anesthetized with isoflurane, their head shaved, and an incision made along the rostral-caudal aspect to expose the skull. A 4 mm craniotomy was created just lateral to the midline and equidistant between bregma and lambda. The severe CCI to the cortex was performed using a PSI model TBI0300 pneumatic device with a 3-mm flat impactor tip. A velocity of 3.5-mm/sec, cortical depth of 1.0 mm, and dwell time of 500 ms produces a severe injury (Osier and Dixon 2016). Sham animals received the identical surgical procedures but no injury.

Isolation of CD11b+ Cells from Bone Marrow and Brain
Animals were sacrificed at 3 h, 24 h, 7 days and 14 days following TBI. CD11b positive (CD11b+) cells were isolated from bone marrow and brain using EasySep™ Mouse CD11b Positive Selection Kit II (Catalog # 18970, STEM-CELL Technologies) following the manufacturer's protocol. Briefly, bone marrow cells were harvested by flushing femurs and tibias with DMEM/F-12 with 10% FBS and 1% penicillin/streptomycin (Invitrogen Life Technologies) using syringe with a 25-gauge needle. A single cell suspension was prepared by passing harvested bone marrow cells through 18-gauge needle 4-6 times. After centrifugation at 300×g for 5 min, red blood cell (RBC) depletion was performed by adding 3 ml RBC lysis buffer (0.15 M NH 4 Cl, 10 M KHCO 3 , and 0.1 M Na 2 EDTA, pH 7. 4) on ice for 3 min. RBC lysis was stopped by adding 15 ml of EasySep™ Buffer (Catalog #20144) followed by centrifugation at 300×g for 5 min. Cells were resuspended in 0.5 ml EasySep™ Buffer and subjected to CD11b+ cell isolation using EasySep™ Mouse CD11b Positive Selection Kit II. For brain CD11b+ cells isolation, the injured or surgery hemisphere was minced in a petri dish containing 1 ml digestion media (DMEM/F-12, 15 mM HEPES, 2% FBS, 20 units/ml Papain (Catalog #07465) and 1 mg/ml DNase I). Minced brain pieces were then transferred to 50 ml conical tubes and incubated at 37 °C for 20 min on a gently shaking platform. A single cell suspension was obtained by filtering the digested tissue through a 70 µm nylon mesh strainer and the strainer rinsed with sample preparation media (DMEM/F-12, 15 mM HEPES and 2% FBS). The cell suspension was subjected to centrifugation at 300×g for 10 min at 4 °C. After removing the supernatant, 6 ml of 30% isotonic percoll solution (Catalog #17-0891-01, GE Healthcare) was added to the suspended pellet followed by centrifugation at 700×g for 10 min at 4 °C with the brake off. The upper myelin layer was carefully removed and the pellet subjected to RBC depletion and CD11b+ cell isolation as above.

RNA Isolation and Analysis
Isolation and analysis of total RNAs from CD11b+ cells followed the procedures described previously (Wang et al. 2015(Wang et al. , 2020. MiRNA/gene expression were analyzed using TaqMan assays (ThermoFisher Scientific). Small nuclear U6 RNA and Hprt (hypoxanthine phosphoribosyltransferase 1) were served as endogenous controls for normalizing miRNA and gene expression in the same sample, respectively.

Statistical Analysis
GraphPad Prism (Version 8.0.2; San Diego, CA, United States; RRID:SCR_002798) was used for statistical analyses. All data in this report presented as means ± standard error of the mean (SEM). Repeated-measures two-way analysis of variance (ANOVA) was used for comparison among groups followed by Tukey's post hoc test to identify specific differences between groups. A value of p < 0.05 with 95% confidence interval is considered statistically significant. Number of animals used in each study was represented in figure legends section of each result.

Preferential Enrichment of Inflammatory miRNAs in CD11b+ Cells from Brain and Bone Marrow
TaqMan RT-qPCR was used to measure selected inflammatory miRNAs including miR-146a-5p, miR-150-5p, miR-223-3p, and miR-155-5p levels in brain and bone marrow CD11b+ cells from female and male naïve mice. While miR-146a and miR-150 levels were detected in bone marrow CD11b+ cells in both females and males, they were significantly higher in brain CD11b+ cells compared to bone marrow in both sexes (Fig. 1). Both miRNAs showed higher levels in brain CD11b+ cells of male mice compared to females, but only miR-150 reached statistical significance. Conversely, miR-223-3p, and miR-155-5p were more than 25-fold higher in CD11b+ cells isolated from bone marrow relative to brain in both sexes, with only miR-155-5p showing significant sex-specific enrichment (males > females) in brain (Fig. 1). This pattern demonstrates that inflammatory miRNAs are differentially enriched in brain and bone marrow CD11b+ cells. Fig. 1 Enrichment of miR-NAs in CD11b+ cells of naïve mice. Levels of inflammatory miRNAs miR-146a-5p, miR-150-5p, miR-223-3p, and miR-155-5p in brain and bone marrow CD11b+ cells of male and female mice."#" indicates significant differences between bone marrow and brain and "*" indicates significant differences between female and male (n = 5 mice/per group) 1 3

Sex-Specific Alterations of Inflammatory miRNAs in CD11b+ Cells Following Severe TBI
To assess sex-specific miRNA alterations following TBI, we measured levels of these miRNA in CD11b+ cells from injured cortical tissue and bone marrow at 3 h, 24 h, 7 days, and 14 days after severe TBI. The levels of antiinflammatory miR-146a-5p were altered marginally over time (Fig. 2a). MiR-150-5p was significantly increased in females but not males at 7 days and 14 days post-injury (Fig. 2b). MiR-223-3p is another anti-inflammatory miRNA and its levels significantly increased by 24 h after TBI in both females and males, with females having significantly higher levels compared to males (Fig. 2c). Levels of proinflammatory miR-155-5p in brain CD11b+ cells were significantly increased in both female and male mice after TBI, although females showed a significantly greater induction at 3 h compared to males (Fig. 2d). Interestingly, we did not observe any TBI-related or sex differences in inflammatory miRNAs in bone marrow CD11b+ cells (Suppl Fig. 1).

Sex-Specific Alterations of Brain miR-223-3p Targets and Inflammatory Gene Expressions in CD11b+ Cells Following Severe TBI
We then examined the expression profile of miR-223-3p specific targets and related inflammatory genes in brain CD11b+ cells of female and male mice following TBI. The sham animals did not exhibit any significant sex-specific patterns in the expression levels of any of the inflammatory genes examined. Reverse-correlated with the increase in miR-223-3p reported above (Fig. 2c), we found that the levels of two miR-223-3p targets, FBXW7 and TRAF6, significantly decreased at 24 h post-injury only in brain CD11b+ cells from female mice (Fig. 3a). Interestingly, we observed an increased expression of the anti-inflammatory gene, ARG1, in both female and male injured mice compare to sham animals at 24 h post-injury, with females showing significantly higher expression than males (Fig. 3b). At this same 24 h time point, brain CD11b+ cells from female, but not male, mice had significantly higher expression of another anti-inflammatory gene, IL4 (Fig. 3b).
Several pro-inflammatory gene markers also exhibited a sex-related differential expression in brain CD11b+ cells following TBI. For example, CCL2 was significantly increased in both female and male mice after TBI, although significantly greater increase was observed in males relative to females at 24 h (Fig. 3c). In addition, there was a significant increase in COX2 expression only in male mice at 3 h and 24 h, which gradually decreased with time (Fig. 3c). The levels of TNFα and IL1b showed different sex-related expression patterns over time following injury (Fig. 3c). Specifically, both pro-inflammatory markers peaked at 3 h after TBI in males and at 24 h in females, with significantly higher levels in females compared to males at 24 h (Fig. 3c).

Fig. 2
Response of inflammatory miRNAs to TBI over time. Levels of inflammatory miRNAs a miR-146a-5p, b miR-150-5p, c miR-223-3p, and d miR-155-5p in brain CD11b+ cells of male and female mice at different time points following TBI. Mean levels of miRNAs in CD11b+ cells isolated from sham-surgery female or male mice were used to normalize respective miRNA levels in CD11b+ cells isolated from TBI female or male mice at each time point. Levels of sham group were set as 1. "#" indicates significant differences between sham and TBI group and "*" indicates significant differences between females and males (n = 5 female mice/time point except n = 4 for TBI group at 7 days; n = 6 male mice in sham and 8 male mice in TBI / time point)

Discussion
Our present study demonstrates the existence of a significant and previously unreported preferential enrichment and sex-specific response of inflammatory miRNAs in brain and bone marrow CD11b+ cells. The inflammatory miR-NAs examined in this study play key roles in inflammatory signaling in myeloid cells and are all implicated in TBI (Liu et al. 2014;Redell et al. 2009;Wang et al. 2015): miR-146a is a negative regulator of NF-κB pathway (Taganov et al. 2006); miR-223-3p plays a key role in limiting myeloid cell pro-inflammatory signaling (Jiao et al. 2021); miR-155-5p is a highly inducible, pro-inflammatory miRNA (Jablonski et al. 2016;Sun et al. 2012); and miR-150-5p is an important modulator for differentiation and activation of immune cells (Zhou et al. 2007). Our results revealed that miR-146a-5p and miR-150-5p are relatively enriched in brain CD11b+ resident microglia and that miR-223-3p and miR-155-5p are relatively enriched in bone marrow CD11b+ monocytes and macrophages. Although we did not detect significant changes of these miRNAs in bone marrow CD11b+ cells following TBI, intriguingly, the response of brain CD11b+ miRNAs to TBI are vastly different with regards to sex and time following injury. For example, an early and significant elevation of pro-inflammatory miR-155-5p in brain CD11b+ cells was observed in both female and male mice compared to shams, which returned to sham levels by 7 days post-injury but again increased significantly at 14 days. It is interesting to point out that the levels of miR-155-5p were higher in brain CD11b+ cells from naïve males, while the levels of miR155-5p in CD11b+ cells from injured female brain were significantly higher than males at the earliest time point examined (3 h). Given its pro-inflammatory role (O'Connell et al. 2007), it could be argued that the rapid increase in miR-155-5p levels following TBI reflects the early response of the innate immune system. Moreover, the sex-related increase in miR-155-5p may provide mechanistic insights into the observation that infiltrating myeloid cells are more prominent in males versus females following TBI (Doran et al. 2019). We did not observe any sex or TBI-related differences in miR-146a-5p levels from brain or bone marrow CD11b+ cells between 3 h and 7 days post-injury. However, there was a slight, albeit significant, increase in miR-146a-5p in female brain CD11b+ cells compared to males, but only at 14 days following TBI. While it was not possible to predict the pattern in miR-146a-5p levels following TBI, it was surprising not to see more robust changes, especially given the role of miR-146a-5p as a fine regulator of anti-inflammatory signaling. MiR-146a-5p is regulated by many rapidly responding inflammatory modulators such as NF-κB (Taganov et al. 2006). Therefore, it is possible that the post-injury time points chosen may miss detecting any significant alterations.
Anti-inflammatory miR-223-3p was significantly increased in brain CD11b+ cells from males at 3 h, and in both sexes at 24 h, with significant higher levels in females compared to males. However, this increase diminished at 7 and 14 days following TBI in both females and males. The Fig. 3 Changes of inflammatory marker genes and targets of miR-223-3p following TBI. Targets of miR-223-3p (a), anti-inflammatory marker genes (b), and pro-inflammatory marker genes (c) in brain CD11b+ cells of female and male mice at different time points following TBI. Mean levels of gene expression in CD11b+ cells isolated from sham-surgery female or male mice were used to normal-ize respective gene expression levels in CD11b+ cells isolated from TBI female or male mice at each time point. Levels of sham group were set as 1. "#" indicates significant differences between sham and TBI group and "*" indicates significant differences between females and males (n = 5 female mice/time point except n = 4 for TBI group at 7 days; n = 6 male mice in sham and 8 male mice in TBI /time point) dynamic changes of pro-and anti-inflammatory miRNAs is consistent with the experimental and clinical observations that while reparative anti-inflammatory signaling peak shortly following brain injury, pro-inflammatory effectors persist for weeks to months to years (Loane et al. 2014;Pierce et al. 1998).
We further examined the expression of targets of miR-223-3p and several inflammatory marker genes. MiR-223-3p is directly associated with reparative inflammatory response via regulation of key inflammatory signaling pathways such as NF-κB and the inflammasome (Bauernfeind et al. 2012;Kumar et al. 2014;Zhou et al. 2018). Our data showed a significant down regulation of the miR-223-3p validated targets FBXW7 and TRAF6, which are key modulators of the NF-κB pathway, in females compared to males at 24 h following TBI. Interestingly, the expression of the anti-inflammatory genes, ARG1 and IL4 were significantly increased in female relative to male mice and sham at the same 24 h time point. Whether the increased expression of anti-inflammatory genes is a direct result of miR-223-3p regulation will need to be further investigated. The pro-inflammatory marker gene profiles were more diverse. There is a clear increase in the levels of the pro-inflammatory genes CCL2, COX2, TNFα, and IL1b in both male and female injured brains compared to sham. However, COX2 and CCL2 were higher in males at 3 and 24 h post-injury, whereas greater levels of TNFα and IL1b were observed in females TBI at 24 h, although an initial higher level were seen in males at 3 h. These complicated sex-bias inflammatory responses could be a consequence of multiple signaling events that involve the dynamic changes of inflammatory miRNAs. Regardless, these results are in line with the findings that inflammatory factors display sexual dimorphic responses following TBI (Spani et al. 2018;Villapol et al. 2017).
We realize there are limitations and caveats of the current study. For example, there are many other inflammatory responsive miRNAs that may participate in cell/tissue-and sexual-specific modulation of inflammatory response, but our studies were limited to only four well-studied inflammatory responsive miRNAs. Moreover, we only examined two acute time points (3 and 24 h) and two chronic time points (7 and 14 days). We believe there could be additional significant sex-related changes at other time intervals. In addition, although CD11b+ cells contain major monocytes, such as macrophages and microglia, the isolation procedure used cannot distinguish subtypes of these monocytes. Finally, the use of homogenates to examine changes in miRNA species in sorted CD11b+ myeloid cells does not allow for any cell specific resolution. Future investigations will be required to fully address the tissue/cell specific (e.g., in situ hybridization or flow cytometry study) and sexual dimorphic miRNA inflammatory responses following TBI.