Significance of ubiquitin carboxy-terminal hydrolase L1 elevations in athletes after sub-concussive head hits

Abstract

The impact of sub-concussive head hits (sub-CHIs) has been recently investigated in American football players, a population at risk for varying degrees of post-traumatic sequelae. Results show how sub-CHIs in athletes translate in serum as the appearance of reporters of blood-brain barrier disruption (BBBD), how the number and severity of sub-CHIs correlate with elevations of putative markers of brain injury is unknown. Serum brain injury markers such as UCH-L1 depend on BBBD. We investigated the effects of sub-CHIs in collegiate football players on markers of BBBD, markers of cerebrospinal fluid leakage (serum beta 2-transferrin) and markers of brain damage. Emergency room patients admitted for a clinically-diagnosed mild traumatic brain injury (mTBI) were used as positive controls. Healthy volunteers were used as negative controls. Specifically this study was designed to determine the use of UCH-L1 as an aid in the diagnosis of sub-concussive head injury in athletes. The extent and intensity of head impacts and serum values of S100B, UCH-L1, and beta-2 transferrin were measured pre- and post-game from 15 college football players who did not experience a concussion after a game. S100B was elevated in players experiencing the most sub-CHIs; UCH-L1 levels were also elevated but did not correlate with S100B or sub-CHIs. Beta-2 transferrin levels remained unchanged. No correlation between UCH-L1 levels and mTBI were measured in patients. Low levels of S100B were able to rule out mTBI and high S100B levels correlated with TBI severity. UCH-L1 did not display any interpretable change in football players or in individuals with mild TBI. The significance of UCH-L1 changes in sub-concussions or mTBI needs to be further elucidated.

Figure 1. Pattern of S100B, UCH-L1 and β-2 transferrin changes in players after football games.

Serum samples were drawn as described in the Methods. A total of 15 players were enrolled and analysis of samples reported here refers to two games played during the regular season. A), B) and C) refer to absolute S100B, UCH-L1 and β-2 transferrin serum levels respectively. Serum levels were measured pre- (day before the game) and post-games (within one hour from the end of a game). In D) the normalized S100B, UCH-L1 and β-2 transferrin serum levels are shown side by side to allow a direct comparison. Normalized values for a given serum markers were obtained by the following equation: . In the figures, each symbol represents a player and any given player is represented by the same symbol throughout this manuscript. Note that on average S100B and UCH-L1 were increased after a game, while the values for beta-2 transferrin remained unchanged. Also note that the beta-2 transferrin values pre-game were highly variable (C) while most of baseline values for S100B and UCH-L1 fell within a defined range. Statistical differences by student's t-test are shown as (*) for p<0.05; and n.s. for not significant.

Figure 2. S100B serum surges correlate with the extent and number of head hits, while UCH-L1 does not appear to correlate with sub-concussive head hits.

Mean percent change values of S100B and UCH-L1 levels (see equation (1)) were plotted against Head Hit Index (HHI) scores (see Methods). A statistically significant difference (by Wilcoxon Mann Whitney) was found between S100B surges at HHI of 0 and HHI of 1 and 2. UCH-L1 did not correlate with any of the HHI used for this study, and the levels of UCH-L1 measured after the game and normalized for their pre-game value as in (1) did not discriminate between a HHI of 0 (special team or played without any head hit) and any HHI (see also for details on HHI). Statistical differences were analyzed by the Wilcoxon Mann Whitney test, both in the comparison between S100B and UCH-L1 at a particular HHI and in the comparison of the individual marker at a HHI 0 vs. HHI 1, 2, 3, and 4,6. To correct for type 1 error in multiple comparisons the Dunnett’s correction was used. Significance is shown as (*) for p<0.05; and n.s. for not significant.

Figure 3. Serum levels of UCH-L1 do not correlate with S100B or levels of β-2 transferrin.

A) S100B serum levels measured at pre- and post- game did not correlate with UCH-L1 serum levels (p = 0.16; R² = 0.19). B) Lack of correlation between UCH-L1 or S100B and β-2 transferrin levels. β-2 transferrin serum levels measured at pre- and post- game (two games) plotted against UCH-L1 failed to show a statistically significant correlation (p = 0.28; R² = 0.20). C) S100B serum levels measured at pre- and post- game (two games) did not correlate with β-2 transferrin serum levels (p = 0.22; R² = 0.23). Significance is determined by ANOVA and denoted as (*) for p<0.05.

Figure 4. Markers’ change following a game confirms lack of correlation between S100B and UCH-L1.

A–B) refers to normalized serum surges of S100B and UCH-L1 after game 1 and game 2. Note lack of correlation between these markers (p = 0.19; R² = 0.37; game 1, p = 0.11; R² =  −0.45; game 2). The red line is the linear regression fit while the outer lines show confidence intervals of 95%. We measured the normalized percent change in each game in an attempt to correct for the possibility of different measuring sensitivities for subconcussive head hits for the two markers measured by ELISA. Significance is determined by ANOVA and denoted as (*) for p<0.05.

Figure 5. Lack of significant correlation between serum UCH-L1 levels and a diagnosis of mild TBI.

A) S100B levels are significantly elevated in patients with a diagnosis of mild TBI compared to healthy controls (p<0.01) and, B) S100B levels correlate with post-traumatic findings on head CT (p<0.01). C) and D) UCH-L1 levels remain unchanged in patients with a diagnosis of mTBI compared to controls as well as in those patients with positive findings on head CT. Significance is determined by ANOVA and denoted as (*) for p<0.05.