Increased severity of closed head injury or repetitive subconcussive head impacts enhances post-traumatic headache-like behaviors in a rat model

Abstract

Introduction:

Posttraumatic headache is one of the most common, debilitating, and difficult symptoms to manage after a traumatic head injury. The development of novel therapeutic approaches is nevertheless hampered by the paucity of preclinical models and poor understanding of the mechanisms underlying posttraumatic headache. To address these shortcomings, we previously characterized the development of posttraumatic headache-like pain behaviors in rats subjected to a single mild closed head injury using a 250 g weight drop. Here, we conducted a follow-up study to further extend the preclinical research toolbox for studying posttraumatic headache by exploring the development of headache-like pain behaviors in male rats subjected to a single, but more severe head trauma (450 g) as well as following repetitive, subconcussive head impacts (150 g). In addition, we tested whether these behaviors involve peripheral calcitonin gene-related peptide signaling by testing the effect of systemic treatment with an anti-calcitonin gene-related peptide monoclonal antibody (anti-calcitonin gene-related peptide mAb).

Methods:

Adult male Sprague Dawley rats (total n = 138) were subjected to diffuse closed head injury using a weight-drop device, or a sham procedure. Three injury paradigms were employed: A single hit, using 450 g or 150 g weight drop, and three successive 150 g weight drop events conducted 72 hours apart. Changes in open field activity and development of cephalic and extracephalic tactile pain hypersensitivity were assessed up to 42 days post head trauma. Systemic administration of the anti-calcitonin gene-related peptide mAb or its control IgG (30 mg/kg) began immediately after the 450 g injury or the third 150 g weight drop with additional doses given every 6 days subsequently.

Results:

Rats subjected to 450 g closed head injury displayed an acute decrease in rearing and increased thigmotaxis, together with cephalic tactile pain hypersensitivity that resolved by 6 weeks post-injury. Injured animals also displayed delayed and prolonged extracephalic tactile pain hypersensitivity that remained present at 6 weeks post-injury. Repetitive subconcussive head impacts using the 150 g weight drop, but not a single event, led to decreased vertical rearing as well as cephalic and extracephalic tactile pain hypersensitivity that resolved by 6 weeks post-injury. Early and prolonged anti-calcitonin gene-related peptide mAb treatment inhibited the development of the cephalic tactile pain hypersensitivity in both the severe and repetitive subconcussive head impact models.

Conclusions:

Severe head injury gives rise to a prolonged state of cephalic and extracephalic tactile pain hypersensitivity. These pain behaviors also develop following repetitive, subconcussive head impacts. Extended cephalic tactile pain hypersensitivity following severe and repetitive mild closed head injury are ameliorated by early and prolonged anti-calcitonin gene-related peptide mAb treatment, suggesting a mechanism linked to calcitonin gene-related peptide signaling, potentially of trigeminal origin.

Introduction

Post-traumatic headache (PTH) is one of the most commonly occurring and disabling sequalae of traumatic head injury, in particular following motor vehicle accidents, falls, and sport-related activities (1–3). PTH is defined by the International Classification of Headache Disorders (ICHD-3) as a secondary headache that develops within seven days after the head trauma, regaining of consciousness following the injury to the head, or discontinuation of medication (s) impairing ability to sense or report headache following the injury to the head (4). PTH shares similar clinical characteristics with primary headaches, in particular migraine and tension-type headache (TTH) (5–8). However, the extent to which these conditions share common pathophysiological mechanisms that can be targeted using current or novel therapeutic approaches remains unclear, in part due to the small number of studies that employ clinically relevant animal models.

A key factor thought to influence the propensity to develop PTH is pre-existing headache or migraine conditions (3). Yet, many cases of PTH arise de novo following a concussive head trauma (8). We have recently characterized a rat model of de novo PTH that displays numerous headache/migraine-like pain behaviors following a single mild closed head injury (CHI) (9). In particular, we observed the development of cephalic tactile pain hypersensitivity that could be ameliorated by systemic treatment with a blocking monoclonal antibody that targets calcitonin gene-related peptide (Anti-CGRP mAb) (9), suggesting the involvement of peripheral CGRP signaling.

Although our rat model, as well as other recently developed mouse models (10,11), are likely to provide important insights into the mechanisms underlying PTH, they do not fully capture the factors and conditions that lead to PTH as they are limited to the effects of a single mild concussive head trauma. An important factor that may influence the propensity to develop of PTH is the severity of the head trauma. PTH has been suggested to be more common after mild trauma when compared to more severe events (3). The prevalence of persistent PTH was also reported to be inversely correlated with the severity of the head trauma and the resultant brain injury (12). However, moderate and severe traumas are also associated with a high prevalence of PTH (13). Nonetheless, some studies suggest either no correlation between the severity of the trauma and incidence of PTH (14,15), increased frequency of persistent and more severe PTH following moderate or severe head trauma (16), and similar sensory abnormalities regardless of the injury severity (7). Although acute head trauma is considered the leading cause of PTH, repetitive subconcussive head impacts, such as those occurring during professional and recreational sporting events, and which are difficult to detect or quantify, can also lead to post-traumatic neurological sequelae and PTH (17). Here, we aimed to address these additional factors, and further extend the preclinical research toolbox for studying PTH. We first characterized time-course changes in cephalic and extracephalic pain sensitivity in a model that mimics a more severe head trauma. We then examined the cumulative effect of repetitive subconcussive head impacts on the development of these pain behaviors. Finally, we addressed the relative contribution of peripheral CGRP signaling in these two new PTH modeling paradigms by studying the analgesic effect of systemically administered anti-CGRP mAb.

Materials and methods

Animals

All experiments were approved and conducted in compliance with the institutional Animal Care and Use Committee of the Beth Israel Deaconess Medical Centre, and the ARRIVE (Animal Research: Reporting of in vivo Experiments) guideline. Subjects were adult male Sprague-Dawley rats (total n = 138). Animals were obtained from Taconic, USA, and were 8–9 weeks at arrival. They were housed in pairs with food and water available ad libitum under a constant 12-hour light/dark cycle (lights on at 7:00 am) at room temperature. Studies were initialized after a week of acclimatization in the vivarium. All procedures and testing were conducted during the light phase of the cycle (8:00 am to 3:00 pm). Experimental animals were randomly assigned to either sham or CHI as well as to the different pharmacological treatment groups.

Experimental closed head injury (CHI) models

We have previously used a mild CHI model involving a 250 g weight drop. Here, we tested three different CHI paradigms, employed in separate cohorts of rats. The first paradigm involved a single, more severe hit, using a 450 g weight drop. The second paradigm employed a single 150 g weight drop. The third paradigm involved three successive 150 g weight drop events, conducted 72 hours apart. All head injury paradigms were conducted while the animals were anesthetized with 3% isoflurane and placed chest down directly under the weight-drop concussive head trauma device. The device consisted of a hollow cylindrical tube (80 cm) placed vertically over the center of the rat’s head. To ensure consistency of the hit location, animals were placed under the weight drop apparatus so that the weight struck the scalp slightly anterior to the center point between the ears. A foam sponge (thickness 3.81 cm, density 1.1 g/cm³) was placed under the animals to support the head while allowing some linear anterior-posterior motion without any angular rotational movement at the moment of impact. A repeated strike was prevented by capturing the weight after the first strike. Surviving animals regained their righting reflex within 2 minutes (which likely reflects the recovery from anesthesia), and were returned immediately to their home cages for recovery. Animals were assessed during the first 7 days post-injury for any behavioral abnormalities suggestive of a major neurological (i.e. motor) impairment. For all paradigms, sham animals were anesthetized as CHI animals, but not subjected to the weight drop.

Open field behavior

Changes in locomotor activity, exploratory behavior, and anxiety-like behavior in an open field arena (43 × 43 × 30 cm) were monitored as previously described (9,18,19) using Activity Monitor (Med Associates, Vermont, USA). The monitoring system records the movement of animals in the horizontal (X-Y axis) and vertical (Z axis) planes using 16 infrared beams and detectors, spaced 2.54 cm apart. Data analyzed included total distance moved, vertical rearing events, and relative time spent in the center of the arena (30 cm × 30 cm) during a 20-minutes session. The arena was lit with a single white LED bulb on a dimmer switch to maintain homogenous lighting across the arenas (80 lux). The arena was cleaned with a mild detergent and dried to remove odor cues between successive rats. Given that activity in open field testing is based on novelty exploration, and repeated exposure decreases novelty, testing was limited to four trials: One baseline trial, and three post-injury trials, conducted on days 3, 7, and 14 post-CHI.

Assessment of tactile pain hypersensitivity

To minimize the effects of repeated handling and different testing environments, animals used for evaluating changes in pain behaviors were not previously subjected to open field testing. To assess changes in tactile pain sensitivity, we employed a method that was previously used by us and others to study PTH-and migraine-related pain behaviors (9,20–22). Briefly, animals were placed in a transparent acrylic tube (20.4 cm × 8.5 cm) closed on both ends. The apparatus was large enough to enable the animals to escape the stimulus. Animals were habituated to the arena for 15 minutes before the initial testing. To determine if the animals developed pericranial (cephalic) tactile hypersensitivity, the skin region, including the midline area above the eyes and 2 cm posterior, was stimulated with different von Frey (VF) filaments, using an ascending stimulus paradigm (0.6–15 g/force, 18011 Semmes-Weinstein Anesthesiometer kit). The development of hind paw tactile hypersensitivity was tested in the same animals by stimulating the mid-dorsal part of the hind paw using the VF filaments. We evaluated changes in withdrawal thresholds, as well as nonreflexive pain responses to the stimulation using a method previously described in this and other headache models (18,23,24) by recording four behavioral responses adapted from Vos et al. (25) as follows: 0) No response: Rat did not display any response to stimulation; 1) detection: Rat turned its head towards stimulating object and latter was explored, usually by sniffing; 2) withdrawal: Rat turned its head away or pulled it briskly away from stimulating object (usually followed by scratching or grooming of stimulated region); 3) escape/attack: Rat turned its body briskly in the holding apparatus to escape stimulation or attacked (biting and grabbing movements) the stimulating object. Starting with the lowest force, each filament was applied three times with an intra-application interval of 5 sec and the behavior that was observed at least twice was recorded. For statistical analysis, the score recorded was based on the most aversive behavior noted. The force that elicited two consecutive withdrawal responses was considered as the threshold. To evaluate pain behavior in addition to changes in the threshold for each rat, at each time point, a cumulative response score was determined by combining the individual scores (0–3) for each one of the VF filaments tested as in Vos et al. (25). All tests were conducted and evaluated in a blinded manner.

Pharmacological treatments

Anti-CGRP mAb and its corresponding isotype IgG were provided by Teva Pharmaceuticals and formulated in phosphate-buffered saline (PBS). Initial doses of the mAb and the control IgG (30 mg/kg ip) were administered immediately after the 450 g head injury or the third 150 g weight drop. Additional doses were administered every 6 days subsequently in both injury paradigms. Similar dosing regimens have been shown to alleviate cephalic tactile pain hypersensitivity following mild CHI in male rats (9,18), and male mice (11), as well as pain behaviors in chronic migraine models (26).

Data analyses

Group size was determined using a priori power analysis (G*Power 3.1 software). Calculations were based on our previous and pilot data showing that they were sufficient to detect large (d > 0.8) effect sizes in at least 80% power, and α = 0.05. Statistical analyses of experimental data were conducted using GraphPad Prism (version 8.0). All data are presented as the means ± standard error of the mean. Mechanical pain threshold data were log-transformed, which is more consistent with a normal distribution (27). Changes in open field behavior and behavioral responses to mechanical stimuli were analyzed using data from sham and CHI animals using a mixed-design ANOVA to determine the effects of time and treatment. The same analysis was used to determine the effect of anti-CGRP mAb versus the corresponding IgG control. All data included passed the Brown-Forsythe test, indicating equal variance. We used Fisher’s LSD post-hoc tests, and correction for multiple comparisons (Type I error) was conducted using the Benjamini and Hochberg false discovery rate (FDR) controlling procedures. Significant levels of the ANOVA (p-values) and post-hoc tests (q values) were set as 0.05.

Results

The effects of 450 g weight drop injury

Open field behavior.

In our earlier studies, all animals subjected to a mild CHI involving a 250 g weight drop recovered (9,18). Here, a single 450 g CHI resulted in immediate mortality (<30 sec) in ~20% of cases, suggesting a severe head (and likely brain) trauma. We further addressed the severity of the head injury by examining consequent changes in exploratory, and anxiety-like behaviors using open field testing. When compared to sham-treated animals (n = 8), rats surviving the CHI (n = 8) did not exhibit changes in total distance traveled (Time F_(3,42) = 2.9, p > 0.05; Treatment F_(1,14) = 1.1, p > 0.05, Figure 1(b)) but displayed decreased exploratory rearing activity at 3, 7, and 14 days post-CHI (Time F_(3,42) = 4.4, p < 0.01; Treatment F_(1,14) = 24.5, p < 0.001; q < 0.001 for day 3, q < 0.05 for day 7, q < 0.01 for day 14; Figure 1(c)), suggesting a brain injury-related response (28). In addition, head-injured animals displayed an acute decrease in center zone exploration at 3 days post-CHI (Time F_(3,42) = 4.5, p < 0.01: Treatment F_(1,14) = 4.7, p < 0.05; q < 0.001; Figure 1(d)), suggesting elevated anxiety levels.

Figure 1.

Changes in open field behaviors in animals subjected to a CHI involving a 450 g weight drop or a sham procedure. (a) Schematic of the experimental design. Rats were subjected to a baseline open field testing, followed by CHI, and then additional testing 3–14 days later. Changes in distance moved (b), exploratory vertical rearing (c), and time spent in the center zone (d). Mixed-design ANOVA, followed by post-hoc test between CHI and sham animals indicates decreased vertical rearing at 3, 7, and 14 days post-CHI as well as an acute decrease in time spent in the center zone (indicating increased thigmotaxis) at 3 days following CHI. Data are means ± SEM (sham: n = 8; CHI: n = 8).

*q < 0.05; **q < 0.01, ***q < 0.001 (FDR-corrected values after Fisher’s LSD post-hoc test at selected time points vs. similar time points in sham control animals).

FDR: false discovery rate; CHI: closed-head injury.

Tactile pain sensitivity.

To assess PTH-related pain behaviors, we examined changes in behavioral responses to tactile stimulation of cephalic and extracephalic sites. When compared to sham animals (n = 12), rats subjected to the 450 g CHI paradigm (n = 14) displayed prolonged cephalic tactile pain hypersensitivity manifested as decreases in pain thresholds, and increases in response scores that recovered by day 42 (Threshold: Time F_(5,102) = 7.2, p < 0.001; Treatment F_(1,24) = 50.3, p < 0.0001; q < 0.001 vs. sham for days 3–30; response score: Time F_(5,102) = 4.2, p < 0.05; Treatment F_(1,26) = 13.0, p < 0.01; q < 0.001 vs. sham for days 3 and 30, q < 0.05 vs. sham for days 7 and 14; Figures 2(b) and (c)). In addition to the pronounced cephalic tactile pain hypersensitivity, head-injured rats also displayed a delayed extracephalic tactile pain hypersensitivity that was manifested as decreases in hind paw pain thresholds on days 14–42 post-CHI (Time; F_(5,102) 3.5, p < 0.01: Treatment F_(1,24) = 15.2, p < 0.001; q < 0.001 vs. sham for days 14, 30 and 42; Figure 2(d)), and increases in response scores on days 30 and 42 post-CHI (Time F_(5,102) = 5.9, p < 0.01: Treatment F_(1,24) = 4.1, p < 0.05; q < 0.001 vs. sham for days 30 and 42; Figure 2(e)).

Figure 2.

Development of prolonged cephalic and extracephalic cutaneous tactile pain hypersensitivity following a 450 g weight drop CHI. (a) Experimental design schematic. Rats were subjected to testing of basal tactile pain sensitivity at the cephalic and extracephalic (hind paw) regions, followed by CHI, and further testing at these locations 3–42 days later. Time course changes in cephalic (b) and extracephalic (d) mechanical pain withdrawal thresholds and corresponding cumulative response scores at the cephalic (c) and extracephalic (e) regions. Mixed-design ANOVA, followed by post-hoc test between CHI and sham animals indicates prolonged decreases in withdrawal thresholds and increases in response scores following CHI at the cephalic and extracephalic regions. Data are means ± SEM (sham: n = 15; CHI: n = 11).

*q < 0.05, ***q < 0.001 (FDR-corrected values after Fisher’s LSD post-hoc test at selected time points vs. similar time points in sham control animals).

The effects of repetitive subconcussive 150 g weight drop head impacts

Open field testing.

To address the severity of single and repetitive subconcussive head impacts, we studied related changes in open field behaviors. When compared to sham-treated animals (n = 8), animals subjected to a single 150 g mild CHI (n = 8) did not display significant time-dependent changes in any of the three parameters tested (Distance moved: Time F_(3,42) = 5.0, p < 0.05; Treatment F_(1,14) = 4.3; p > 0.05; rearing: Time F_(3,42) = 0.8, p > 0.05; Treatment F_(1,14) = 3.1, p > 0.05; center zone time: Time F_(3,42) = 0.6, p > 0.05: Treatment F_(1,14) = 2.2; p > 0.05; Figures 3(b), (c), and (d)). We next asked whether the repetitive mild CHI paradigm exerts a cumulative deleterious effect on open field behavior. When compared to sham treatment (n = 10), three consecutive 150 g weight drops, conducted 72 hours apart (n = 10) had no effect on distance moved (Time F_(3,42) = 3.0, p < 0.05; Treatment F_(1,18) = 1.7; p > 0.05; Figure 3(f)), but led to a time-dependent decrease in rearing (Time F_(3,42) = 5.5, p < 0.001; Treatment F_(1,18) = 20.5, p < 0.01; q < 0.001 vs. sham for days 2, 7 and 14 post-injury; Figure 3(g)). Repetitive mild CHI did not impact center zone activity (Time F_(3,42) = 1.3, p > 0.05; Treatment F_(1,18) = 3.9; p > 0.05; Figure 3(h)), indicating no increase in anxiety level.

Figure 3.

Changes in open field behaviors in animals subjected to a single subconcussive mild closed head impact (mCHI), or repetitive subconcussive impacts (rmCHI) using 150 g weight drop. (a) Schematic of the experimental design for the 1 × 150 g mCHI model. Rats were subjected to baseline open field testing, followed by mCHI, and additional testing at 3, 7, and 14 days later. (b) Total distance moved, (c) exploratory vertical rearing, and (d) time spent in center zone (means ± SEM; sham: n = 8; mCHI: n = 8). (e) Schematic of the experimental design for the 3 × 150 g rmCHI model and (f–h) related open field test data (means ± SEM; sham: n = 8; rmCHI: n = 8). Mixed-design ANOVA, followed by post-hoc test indicates decreases in vertical rearing at 3, 7, and 14 days following a rmCHI.

*q < 0.05; **q < 0.01 (FDR-corrected values after Fisher’s LSD post-hoc test at selected time points vs. similar time points in sham control animals).

Tactile pain sensitivity.

Given the open field data suggesting that a single 150 g weight drop mild CHI is subconcussive, we next examined whether this head trauma paradigm leads to increased nociceptive behaviors, and then assessed changes in pain responses following the repetitive mild CHI paradigm. When compared to sham treatment (n = 12), animals subjected to a single 150 g mild CHI (n = 12) did not show any changes in cephalic mechanical pain thresholds (Time F_(3,54) = 3.3, p < 0.05: Treatment F_(1,22) = 0.04; p > 0.05; Figure 4(b)), or response scores (Time F_(3,54) = 3.1, p < 0.05: Treatment F_(1,22) = 0.09; p > 0.05; Figure 4 (c)). Animals subjected to repetitive mild CHI (n = 14), however, developed a prolonged, and time-dependent tactile pain hypersensitivity when compared to sham treatment (n = 12). This response was manifested as a decrease in pain threshold that was already maximal at 3 days following the last head injury and resolved by day 42 (Time F_(5,110) = 6.7, p < 0.001; Treatment F_(1,24) = 16.4, p < 0.001; q < 0.01 vs. sham for days 3–30; Figure 4(e)) and a similar temporal profile of increased response scores (Time F_(5,104) = 3.3, p < 0.01; Treatment F_(1,24) = 11.0, p < 0.01, q < 0.001 vs. sham for days 3–14, q < 0.01 for day 30; Figure 4(f)).

Figure 4.

Development of prolonged cephalic cutaneous tactile pain hypersensitivity following rmCHI but not a single mCHI. (a) Schematics of the experimental design for the 1 × 150 g mCHI model. Rats underwent baseline von Frey testing of tactile pain sensitivity at the cephalic region, followed by either a single 150 g mCHI and further nociceptive testing 3–42 days later. (b,c) Time course changes in cephalic mechanical pain withdrawal thresholds and corresponding cumulative response scores (means ± SEM; sham: n = 8; CHI: n = 8). (d) Schematic of the experimental design for the 3 × 150 g rmCHI model. (e,f) Related changes in thresholds and response scores (means ± SEM; sham: n = 8; rmCHI: n = 8). Mixed-design ANOVA, followed by post-hoc tests between CHI and sham animals indicates prolonged decreases in withdrawal thresholds and increases in response scores at 3–30 days following rmCHI.

**q < 0.01, ***q < 0.001 (FDR-corrected values after Fisher’s LSD post-hoc test at selected time points vs. similar time points in sham control animals).

Assessment of changes in extracephalic pain behavior revealed similar effects. A single 150 g mild CHI did not affect extracephalic tactile pain sensitivity (Threshold: Time F_(3,46) = 4.0, p < 0.05; Treatment F_(1,18) = 0.1; p > 0.05; response score: Time F_(3,46) = 1.5, p < 0.05; Treatment F_(1,22) = 0.3, p > 0.05; Figure 5(b) and (c)). However, animals subjected to repetitive subconcussive mild CHI exhibited extracephalic tactile pain hypersensitivity manifested as a decrease in pain threshold that developed with a delay and was significant on days 14 and 30 (Time F_(5,110) = 3.6, p < 0.01; Treatment F_{(1, 24)} = 5.5, p < 0.05; q < 0.001 vs. sham for both time points; Figure 5(e)).

Figure 5.

Development of delayed extracephalic cutaneous tactile pain hypersensitivity following rmCHI but not a single mCHI. (a) Schematic of the experimental design for the 1 × 150 g mCHI model. Rats underwent baseline von Frey testing of tactile pain sensitivity at the hind paw, followed by a single 150 g mCHI and further nociceptive testing 3–42 days later. (b,c) Time course changes in hind paw pain withdrawal thresholds and corresponding cumulative response scores (means ± SEM; sham: n = 8; CHI: n = 8). (d) Schematic of the experimental design for the 3 × 150 g rmCHI model. (e,f) Related changes in thresholds and response scores (means ± SEM; Sham: n = 8; rmCHI: n = 8). Mixed-design ANOVA followed by post-hoc test between CHI and sham animals indicate a delayed decrease in withdrawal thresholds 14–30 days following rmCHI.

***q < 0.001 (FDR-corrected values after Fisher’s LSD post-hoc test at selected time points vs. similar time points in sham control animals).

The effect of systemic anti-CGRP mAb treatment

In rats subjected to acute 450 g weight drop CHI (n = 7), treatment with the anti-CGRP mAb ameliorated the prolonged cephalic pain hypersensitivity when compared to a similar treatment approach using the control IgG (n = 7). As Figures 6(b) and (c) depict, there were statistically significant differences between the anti-CGRP mAb (n = 8) and IgG (n = 8) treatment groups with regard to changes in thresholds (Time F_(5,70) = 7.4, p < 0.0001; Treatment F_(1,12) = 9.6, p < 0.01) and response scores (Time F_(5,70) = 4.4, p < 0.01; Treatment F_(1,12) = 4.9, p < 0.05). Post-hoc analyses revealed prolonged decreases in cephalic pain thresholds in the IgG treatment group (q < 0.01 vs. baseline for days 3–14, q < 0.05 vs. baseline for day 30), but not in the anti-CGRP mAb treatment group (q > 0.05 vs. baseline for all time points). Prolonged increases in cephalic response scores were also observed in the IgG-treated animals (q < 0.01 vs. baseline for days 3–30), but not in animals treated with the anti-CGRP mAb (q > 0.05 vs. baseline for all time points).

Figure 6.

Early and prolonged treatment with anti-CGRP mAb ameliorates cephalic tactile pain hypersensitivity in rats subjected to 450 g CHI. (a) Schematic of the experimental design. Rats underwent baseline von Frey testing followed by CHI. Anti-CGRP mAb or a control IgG were administered immediately after the CHI with subsequent doses given at 6-day intervals. Mixed-design ANOVA revealed the effects of treatment on the decreases in cephalic withdrawal thresholds (b), and the increases in response scores (c), but not on the extracephalic thresholds (d) or response scores (e). Means ± SEM (anti-CGRP: n = 8; control IgG: n = 8).

*q < 0.05; **q < 0.01 (FDR-corrected values after Fisher’s LSD post-hoc test for post CHI values vs. baseline).

There was no difference between the IgG and anti-CGRP mAb treatments with regard to the extracephalic pain hypersensitivity, including for changes in thresholds (Time F_(5,70) = 2.5, p < 0.05; Treatment F_(1,12) = 0.11, p > 0.05; Figure 6(d)) and response scores (Time F_(5,70) = 0.4, p > 0.05; Treatment F_(1,14) = 0.1, p > 0.05; Figure 6(e)). Post-hoc analyses revealed no changes in thresholds or response scores in animals treated with the anti-CGRP mAb or the control IgG group (q > 0.05 vs. baseline, for both treatments, at all time points).

Treatment with anti-CGRP mAb also exerted an anti-nociceptive effect in rats subjected to repetitive subconcussive 150 g weight drop head impacts. As Figures 7(b, c) depict, there were statistically significant differences between the anti-CGRP mAb (n = 8) and IgG (n = 8) treatment groups with regard to changes in thresholds (Time F_(5,70) = 8.5, p < 0.001; Treatment F_(1,14) = 4.7, p < 0.05) and response scores (Time F_(5,70) = 4.0, p < 0.01; Treatment F_(1,14) = 5.1, p < 0.05). Post-hoc analyses revealed decreased cephalic pain thresholds in IgG-treated animals (q < 0.01 vs. baseline for days 7–30), but no change in animals treated with the anti-CGRP mAb (q > 0.05 vs. baseline for all time points). Similarly, increases in cephalic response scores were observed in the IgG treatment group (q < 0.01 vs. baseline for days 7–14; q < 0.05 vs. baseline for day 30), but not in the anti-CGRP mAb treatment group (q > 0.05 vs. baseline, for all time points). There were no differences between the IgG and anti-CGRP mAb treatments with regard to the extracephalic pain hypersensitivity including for changes in threshold (Time F_(5,70) = 3.0, p < 0.05; Treatment F_(1,14) = 1.4, p > 0.05; Figure 7(d)), and for response score (Time F_(5,70) = 2.4, p < 0.05; Treatment F_(1,14) = 0.7, p > 0.05; Figure 7(e)). Post-hoc analyses revealed decreased thresholds and increased response scores in animals treated with the anti-CGRP mAb (threshold q < 0.05 vs. baseline for days 3–30; Response score q < 0.05 vs. baseline for days 3 and 30; q < 0.01 for days 7 and 14), but no change in the control IgG-treated group (q > 0.05 vs. baseline, for both behaviors, at all time points).

Figure 7.

Early and prolonged treatment with anti-CGRP mAb ameliorates cephalic tactile pain hypersensitivity in rats subjected to rmCHI. (a) Schematic of experimental design. Rats underwent baseline von Frey testing of cephalic and extracephalic tactile pain sensitivity, followed by rmCHI. Anti-CGRP mAb or a control IgG were administered immediately after the last weight drop, with subsequent doses given at 6-day intervals. Mixed-design ANOVA revealed the effects of treatment on the decreases in cephalic withdrawal thresholds (b), and the increases in response scores (c) but not on the extracephalic thresholds (d) or response scores (e). Means ± SEM (anti-CGRP: n = 8; control IgG: n = 8).

**q < 0.01 (FDR-corrected values after Fisher’s LSD post rmCHI values vs. baseline).

Discussion

In the present study, we have extended the preclinical toolbox for studying PTH in a rat model (9), using additional closed head trauma paradigms to investigate the consequences of a more severe head injury, and repetitive subconcussive events. Our data suggest that both of these injury paradigms are associated with increased behavioral symptoms suggestive of traumatic brain injury and prolonged pain hypersensitivity when compared to a milder head injury paradigm (9).

In previous studies, we have used a rat model of diffuse closed head injury by employing a 250 g weight drop from 80 cm height. Here, to produce a more severe CHI we increased the gravity force by employing a heavier mass (450 g weight), and maintained the same height as in our previous CHI paradigm. The finding that the 450 g CHI paradigm was associated with a ~20% mortality rate, while our earlier studies indicated no mortality using the milder 250 g CHI paradigm (9,18) points to increased injury severity, and is in agreement with previous studies that investigated the effects of graded mechanical impact levels (29). Our open field data indicated no change in total distance travelled, suggesting no impairment in gross motor activity, similar to the effect of a 250 g weight drop injury. However, animals subjected to the 450 g weight drop injury exhibited decreased exploratory rearing activity that lasted much longer than that observed in animals subjected to the milder (250 g) CHI paradigm. In the context of this injury model, decreased rearing is likely to indicate traumatic brain injury (28) and possibly also increased anxiety level related to brain injury (30). Furthermore, rats subjected to the 450 g CHI also displayed increased thigmotaxis, another measurement of increased anxiety levels in open field testing (30), a behavior that was not observed previously in the milder CHI paradigm. The prolonged reduction in rearing activity following the more severe head injury could also reflect migraine-like head pain as it was also observed following noxious stimulation of the cranial meninges (22), as well as in response to peripheral and systemic administration of the headache trigger CGRP (31).

In our previous studies, male rats subjected to the 250 g weight drop head injury protocol displayed a gradual increase in cephalic tactile pain sensitivity that reached peak values at 7 days and resolved a week later. Here, increasing the severity of the head injury, using the 450 g weight drop, gave rise to an earlier onset of cephalic pain hypersensitivity that was already maximal on day 3 post-injury, and lasted for more than 4 weeks. The rapid development of the cephalic pain hypersensitivity in this CHI paradigm could be mediated by an acute trauma endured by deep cranial tissues, in particular the calvarial periosteum and intracranial meninges, and the ensuing inflammatory-related sensitization of trigeminal nociceptors that innervate these tissues (24,32). However, the earlier finding that milder CHI paradigms are also associated with rapid and robust injury-related inflammatory responses (e.g. mast cell degranulation) (18,33,34) point to other, or additional contributing factors. One plausible mechanism could involve acute injury to cortical neurons and glia and the ensuing activation and sensitization of meningeal nociceptive afferents via a similar process as occurring following cortical spreading depression (32). Mechanisms related to the acute increase in anxiety levels noted on day 3 post-injury could have also contributed to the rapid onset of the nociceptive behavior. The mechanisms underlying the prolongation of the enhanced trigeminal hyperalgesic response may involve sustained peripheral tissue damage and/or cortical injury that maintain the sensitized state of the deep cranial nociceptors, and the second-order nociceptive neuron in the medullary dorsal horn that receive their input.

Prospective studies suggest that the prevalence of persistent PTH is inversely correlated with the severity of the head trauma and the resultant brain injury (TBI), as is the case with the frequency of other TBI-related pain syndromes (12). It is not clear however whether these differences are due to biological factors, or the result of reporting bias. While persistent PTH is a common sequalae of moderate and severe TBI (16), it is also unclear whether its clinical manifestations are less severe than in PTH cases that develop following milder traumas. Our study, which assessed pain hypersensitivity as a surrogate measure of headache, points to an enhanced pain phenotype following more severe head trauma when compared to our previous data using the milder weight drop head injury model. Further studies will be required to determine whether the 450 g weight drop injury paradigm we employed herein inflicts brain damage and/or peripheral cephalic changes that are within the parameters of those considered to occur following a milder concussive head/brain trauma, or more severe ones. Nonetheless, if the 450 g weight drop injury model can be equated to a moderate/severe TBI, our data may explain the finding that subjects with PTH following these types of injuries are more likely to experience a higher frequency of headaches (several per week/daily) than those with PTH following a milder head trauma (1,16).

Repetitive concussive head injuries, as well as subconcussive events, can lead to numerous chronic neuro-behavioral problems including persistent headache (35). Although there is increased interest in the pathophysiological outcomes of repetitive concussions, including their role in PTH (36), the contribution of subconcussive head impacts to PTH remains poorly studied. Here, we first established a subconcussive weight drop paradigm by showing that a single 150 g weight drop does not exhibit any changes in the open field test. We further observed that this acute subconcussive impact does not lead to headache-like pain behaviors, suggesting a lack of acute cephalic injury. A key finding, however, was that repeating these subconcussive events gave rise to persistent deficits in open field activity, which likely implicates brain injury. In addition, we observed persistent cephalic pain hypersensitivity resembling that encountered in animals subjected to the 450 g weight drop injury. We modeled the repetitive head injury paradigm to include 72 hours of recovery time, which mimics regular events in professional as well as in many amateur contact sports, such as American football, boxing, and hockey. It will be of interest to examine in future studies the minimal duration of recovery time, as well as the number of repetitive events that may be considered safe and do not affect open field activity or produce pain hypersensitivity. Examining these factors could help to address issues such as return to play following sports-related head injuries. The mechanisms by which repetitive, subconcussive head impacts give rise to persistent headache-like pain behavior could potentially implicate the cumulative effects of the acute subconcussive impacts. The similarity to the persistent headache-like behavioral changes observed following repeated administration of inflammatory mediators to the cranial meninges (20) points to the possibility that repeated subconcussive head impacts engage pathophysiological mechanisms similar to those that perpetuate persistent migraine pain. Whether the enhanced pain behaviors are due to cumulative effects of meningeal irritation and related inflammation, or the result of a cortical injury and neuroinflammation will need to be addressed in future studies. We would also like to entertain the possibility that local or remote changes evoked by the initial head impacts promote peripheral and/or central hyperalgesic priming mechanisms (37–39) that give rise to a prolonged cephalic nociceptive response following an additional trauma event.

In addition to the rapid development of cephalic pain hypersensitivity, we also observed extracephalic pain hypersensitivity starting 2 weeks after the severe CHI as well as repetitive subconcussive CHI. This delayed onset suggests that different mechanisms are likely to be responsible for the cephalic and extracephalic pain hypersensitivity in these CHI models. Plausible mechanisms that contribute to the extracephalic hypersensitivity include damage to cortical or subcortical descending pain suppression pathways (40) and the development of brain injury-related systemic inflammatory response (41). Despite having a similarly delayed onset in the two CHI paradigm, the extracephalic pain hypersensitivity persisted at 42 days following the 450 g CHI, while ending at this time point in the repetitive subconcussive CHI paradigm. The relatively prolonged duration of the extracephalic pain hypersensitivity in the 450 CHI paradigm may be attributed to the increased severity of the trauma or to a different mechanism than that underlying this pain symptom in the repetitive subconcussive CHI paradigm.

Repeated systemic administration of the anti-CGRP mAb, starting immediately after the acute 450 weight drop injury, or the last subconcussive 150 g head impact, was able to inhibit the cephalic pain hypersensitivity when compared to treatment with the control IgG. The anti-nociceptive response of the anti-CGRP mAb was overall similar to that previously observed in the 250 g weight drop injury model in male rats with one exception: Treatment was already effective on day 3 in the 450 g impact model after one dose, while it required a second dose in the repetitive CHI paradigm as in the 250 g CHI (9). Due to their large molecular size, anti-CGRP mAb poorly cross the blood-brain barrier and are thought to exert their effect by blocking CGRP-related actions primarily outside the central nervous system. Upon head trauma or repetitive subconcussive events, local release of CGRP from trigeminal cell bodies of deep cranial nociceptors or their peripheral nerve endings might play a nociceptive role. Meningeal release of CGRP and the resultant local inflammatory response are thought to mediate migraine-like pain and the associated trigeminal hypersensitivity. However, previous work suggests that if such a mechanism plays a role following CHI, the development of the cephalic hypersensitivity may not involve mast cells (18). The possibility that the CHI paradigms used in this study lead to disruption of the blood-brain barrier and resultant action of the mAb also on CGRP-mediated processes within the central nervous system may also be considered.

The contribution of CGRP to the development of the extracephalic pain hypersensitivity in the two novel CHI paradigms we tested is less clear. In the 450 g CHI paradigm, extracephalic pain hypersensitivity did not develop in animals treated with either the anti-CGRP mAb or the control IgG. In the repetitive subconcussive CHI paradigm, however, extracephalic hypersensitivity developed in animals treated with the anti-CGRP mAb, but not following the control IgG treatment. Because animals treated with the control IgG did not exhibit extracephalic pain hypersensitivity in both CHI paradigms, we cannot conclude that CGRP mediates this pain behavior. The possibility that the extracephalic pain hypersensitivity in these CHI models is less consistent than the observed cephalic changes should also be considered.

An important limitation of the current study is the lack of female subjects. Females may be more susceptible to developing persistent PTH (8,42), and our recent study points to enhanced PTH-like behavior in female rats in the 250 g CHI paradigm (19). Whether females develop PTH-like symptoms in the new CHI paradigms we tested herein beyond those observed in males will have to be examined in future studies. Because anti-CGRP mAb treatment is less effective in female rats in the 250 g CHI model (when compared to males) (19), it will also be of interest to examine in future studies whether these CGRP-related sex differences also present in the more severe closed-head injury or repetitive subconcussive closed head impact models.

Conclusion

We characterized two additional preclinical traumatic closed head injury paradigms to extend the toolbox for studying the pathophysiology of PTH and its treatment. Our injury paradigms employing a single 450 g weight drop, or repetitive 150 g subconcussive head impacts provide a platform to study more persistent PTH-like behaviors. The repetitive head injury paradigm may also be used to investigate the origin of PTH in sport-related injuries. The anti-nociceptive effect of anti-CGRP mAb in these two novel CHI paradigms further supports this treatment approach in PTH.

Key findings.

• Increased severity of closed head injury in adult male rats is associated with enhanced headache-like behaviors when compared to data obtained in a previous study of a milder injury.

• Repetitive, non-concussive mild closed head injury gives rise to enhanced headache-like behaviors similar to those precipitated by a more severe injury.

• Extended headache-like behaviors following severe and repetitive mild closed head injury are ameliorated by early and prolonged anti-CGRP mAb treatment, suggesting a mechanism linked to peripheral CGRP signaling.