Myelin

For generation of microglial primary cultures, P0 WT or Spp1^KO/KO mice were decapitated, the brain was dissected from the skull and meninges were removed in ice-cold HBSS with 5% FBS. Eight to ten mice were pooled per culture preparation. Tissue was homogenized first with 2 ml pipettes (15 strokes) in 15 ml falcon tube and subsequently transferred to a prewet 50 ml tube with 70 μM strainer. The 15 ml tube was washed with HBSS and then put through filter to ensure all tissues were collected. The supernatant was removed and the cell pellet was resuspended in ice-cold 35% isotonic percoll. The interface was carefully created with HBSS. The samples were centrifuged for 40 min at 4 °C at 2,800g with no break and with slow acceleration and deceleration. The myelin layer and supernatant layers were aspirated, and the cell pellet was washed in HBSS. Cells were centrifuged and resuspended in 1 ml microglial media (DMEM F12 (Gibco), 5% FBS (Gibco), 1% pen-strep (Gibco), 50 ng ml⁻¹ CSF1 416-ML-010/CF (RnD Systems), 50 ng ml⁻¹ TGFb1 7666-MB-005/CF (RnD Systems) and 100 ng ml⁻¹ CX3CL1 472-FF-025/CF (RnD Systems)) for cell counting. Cells were used within 7–10 d of plating.

Associations between MAG and PLP with demographic (e.g. age, race, education level) and neurodegenerative disease diagnoses and non-neurodegenerative disease (e.g. modified ischaemic injury scale) pathologies were tested using Pearson’s bivariate correlation, chi-square or independent samples t-tests. Multivariable linear regressions tested the associations between the exposure to RHI proxies (i.e. years of football play and age of first exposure to football) and MAG and PLP. A separate model was done for each RHI proxy and each ELISA marker. Binary logistic regression models were used to test the association between the exposure proxies and odds for having low MAG and PLP, defined by the sample median split. This was done to facilitate within-sample interpretation of results and effect sizes as well as to account for potential non-linear associations. Previous research showed 11 years of football maximizes sensitivity and specificity for predicting CTE risk.^{14 (link)} As a sensitivity analysis, analysis of covariance compared brain donors who played 11 or more years with those who played <11 years on MAG and PLP levels. Estimated marginal mean difference (‘mean diff.’) from the analysis of covariance is reported and refers to the estimated marginal mean difference for the given outcome between groups being compared, and adjusted for the relevant covariates. Note that because years of football and level of play are highly correlated (r = 0.74, P < 0.001) we did not examine the level of play separately to limit the number of analyses performed.
In separate models, multivariable linear regression analyses tested the association between MAG and PLP and each clinical scale (i.e. FAQ, BRIEF-A BRI and BIS-11). Analysis of covariance compared brain donors who had impaired FAQ (i.e. 9 or higher) and BRIEF-A BRI (i.e. T = 65 or higher) scores versus those with unimpaired scores on PLP concentrations. Because there is no established cut-off to define clinically meaningful impairment for the BIS-11, this analysis was not performed for this scale. The estimated marginal mean diff. is reported. There is a wide age distribution of the sample and only 22 of the brain donors (of the 166 with complete clinical data) were younger than age 50 and the median age of the clinical sample was 72 (71 for the full sample). Given the sensitivity of the clinical outcomes to age, analyses were repeated and stratified by age 50.
As previously described, the MAG:PLP ratio is an indicator of subacute ischaemic injury from hypoperfusion.^{65 (link),67 (link)} Because the goal of our study was to examine the long-term and not subacute effects of exposure to RHI on myelin integrity, MAG and PLP were examined separately as opposed to examining their ratio. However, given there is a precedent in the literature for MAG:PLP, the primary analytic models described above were repeated for the MAG:PLP ratio as post hoc analyses.
A P-value <0.05 defined statistical significance. Two analyses were performed for each RHI exposure metric and three for each clinical scale. Due to the limited number of outcomes, adjustment for multiple comparisons was not performed. All primary analytic models investigating RHI and clinical associations with MAG and PLP controlled for age at death, level of education (less than high school, high school/GED degree, college degree, graduate degree) and arteriolosclerosis (moderate/severe versus none/mild). Arteriolosclerosis was included as a covariate to account for small vessel disease associated with vascular risk factors. The modified ischaemic injury scale was not used due to missingness on this variable. CTE stage (none, low and high) was included as an additional covariate in models examining years of play and clinical outcomes due to its known associations with these variables. Years of football play was included as a covariate for models that examine age of first exposure to reduce the possibility that any observed effects from age of first exposure are related to it being a proxy for duration of play.