CCL2 is associated with microglia and macrophage recruitment in chronic traumatic encephalopathy

Abstract

Background: Neuroinflammation has been implicated in the pathogenesis of chronic traumatic encephalopathy (CTE), a progressive neurodegenerative disease association with exposure to repetitive head impacts (RHI) received though playing contact sports such as American football. Past work has implicated early and sustained activation of microglia as a potential driver of tau pathology within the frontal cortex in CTE. However, the RHI induced signals required to recruit microglia to areas of damage and pathology are unknown.

Methods: Postmortem brain tissue was obtained from 261 individuals across multiple brain banks. Comparisons were made using cases with CTE, cases with Alzheimer's disease (AD), and cases with no neurodegenerative disease and lacked exposure to RHI (controls). Recruitment of Iba1+ cells around the CTE perivascular lesion was compared to non-lesion vessels. TMEM119 staining was used to characterize microglia or macrophage involvement. The potent chemoattractant CCL2 was analyzed using frozen tissue from the dorsolateral frontal cortex (DLFC) and the calcarine cortex. Finally, the amounts of hyperphosphorylated tau (pTau) and Aβ₄₂ were compared to CCL2 levels to examine possible mechanistic pathways.

Results: An increase in Iba1+ cells was found around blood vessels with perivascular tau pathology compared to non-affected vessels in individuals with RHI. TMEM119 staining revealed the majority of the Iba1+ cells were microglia. CCL2 protein levels in the DLFC were found to correlate with greater years of playing American football, the density of Iba1+ cells, the density of CD68+ cells, and increased CTE severity. When comparing across multiple brain regions, CCL2 increases were more pronounced in the DLFC than the calcarine cortex in cases with RHI but not in AD. When examining the individual contribution of pathogenic proteins to CCL2 changes, pTau correlated with CCL2, independent of age at death and Aβ₄₂ in AD and CTE. Although levels of Aβ₄₂ were not correlated with CCL2 in cases with CTE, in males in the AD group, Aβ₄₂ trended toward an inverse relationship with CCL2 suggesting possible gender associations.

Conclusion: Overall, CCL2 is implicated in the pathways recruiting microglia and the development of pTau pathology after exposure to RHI, and may represent a future therapeutic target in CTE.

Keywords: American football; CTE; Chemokine; Head impacts; Microglia; Neuroinflammation; TBI; Tau.

Fig. 1

Greater microglia and macrophages are recruited to the CTE lesion blood vessels. The Iba1+ cell density specific to the CTE pathognomonic lesion was investigated to determine if tau specific glial recruitment occurs. a Representative image of Iba1+ cells found around control and CTE lesion blood vessels at the depth of the cortical sulcus in the DLFC. Left panel is a low power image of the depth of the cortical sulcus. Right panels are high power images of control and CTE lesion blood vessels. White arrows denote Iba1+ cells with processes contacting blood vessel. Asterisk denotes a blood vessel. Scale bar = 50 μm. b Quantitation of the average number of Iba1+ cells found around lesion and control vessels in low and high stage CTE. Each dot represents a single person. c Quantitation of the percentage of TMEM119+/Iba1+ and TMEM119−/Iba1+ cells found around the lesion and control vessels in control, mild, and severe CTE. Each dot represents the Iba1+/TMEM119+ (black circles) or Iba1+/TMEM119− (white squares) percentage from a single person. d Representative image of Iba1+/TMEM119+ and Iba1+/TMEM119− cells around lesion and control vessels. Increased macrophage recruitment was observed around lesion vessels. Asterisk denotes a blood vessel. Scale bar = 100 μm. Error bars are expressed as mean ± SEM. Statistics between mild and severe CTE generated with a two-way ANOVA. *p < 0.05, **p < 0.01

Fig. 2

CCL2 levels correlate with the years spent playing American football, the number of Iba1+ and CD68+ cells in cases with RHI. Levels of DLFC CCL2 were compared against a the number of years spent playing American football, b the density of Iba1+ microglia/macrophages, c and the density of CD68+ inflammatory cells found in the DLFC at the depth of the cortical sulcus. All cases had a history of playing American football. Each dot represents a single person. Significance and slope of the line was calculated using linear regression analysis. As CCL2 was found to have a non-normal distribution, a rank bank transformation technique was used to achieve the required normal distribution needed for linear regression analysis. The transformation resulted in normally distributed Z scores which are plotted on the y axis

Fig. 3

CCL2 is elevated in CTE. Quantitative measurement of CCL2 protein levels in CTE a-c and AD d-f from the a, d dorsolateral frontal cortex (DLFC) (control n = 18, RHI without CTE n = 20, low CTE n = 27, high CTE n = 47, low AD n = 60, intermediate AD n = 28, high AD n = 24) and b, e calcarine cortex (control n = 13, RHI without CTE n = 16, low CTE n = 25, high CTE n = 42, low AD n = 34, intermediate AD n = 12, high AD n = 19). c, f To determine if CCL2 was specifically elevated in the DLFC, CCL2 in the DLFC was divided by CCL2 values in the calcarine cortex to obtain a ratio (control n = 13, RHI without CTE n = 16, low CTE n = 25, high CTE n = 42, low AD n = 37, intermediate AD n = 16, high AD n = 12). Values over 1 represent more CCL2 in the DLFC compared to the CC. Statistics were generated via a one-way ANOVA with a Kruskal-Wallis post-test comparing differences to the control cases. Each dot represents a single case. Error bars show median and interquartile range. *p < 0.05, **p < 0.01, ***p < 0.001 relative to control cases