Male C57BL/6 (n = 40) mice (Jackson Laboratories) 8–12 weeks of age were studied. Mice had ad libitum access to food, water, and were group-housed. Animals were randomly assigned to Naïve, Sham, or CCI groups. All studies were approved by the Veterans Affairs Puget Sound Health Care System’s Institutional Animal Care and Use Committee. Experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and reported in compliance with the ARRIVE guidelines (McGrath et al., 2010).

Mice acclimatize to the animal facility for at least 1 week prior to experimentation. Male C57BL/6 mice were placed under 2–4% isoflurane (0.5 L/m) to shave the head and maintained under anesthesia using a nose cone once placed in a Kopf stereotactic frame. Using sterile technique of 70% alcohol and 10% betadyne were applied to the shaved head and the skull was exposed by a midline incision and skin retracted. Local lidocaine (2.5%) and prilocaine (2.5%) cream was administered to the incision site and a 3 mm diameter opening was made in the skull (+2 mm ML and −2 mm AP from Bregma). The bone was removed and the right cortex was exposed. The cortex was then compressed by using an electromagnetic Impact One device (Leica Biosystems, Buffalo Grove, IL) to deliver a 2 mm diameter rounded tip weight at a velocity of 3.5 m/s to a depth of 1 mm for a duration of 50 ms. The skin was placed back into position and closed with sterile wound clips and the mouse was placed on a heating pad until regaining consciousness before being placed back in the home cage. Sham mice went through the procedure except for delivery of impact.

Mice were anesthetized with urethane (4 g/kg; 0.2 ml; IP) and cardiac perfused with 0.1 M phosphate buffered saline (PBS) followed by 4% paraformaldehyde (PFA) in 0.1 M PBS. Brains were extracted, post fixed for 24 h in 4% PFA at 4°C, cryopreserved in 30% sucrose PBS solution, and frozen in optimal cutting temperature (OCT) compound on dry ice. Brains were stored at −80°C until cryosectioning. Twenty-four hours prior to cryosectioning, brains were acclimated overnight at −20°C and then cut with a Leica CM1950 cryostat at 30 μm-thick serial sections and stored in 0.1 M PBS + 0.02% sodium azide at 4°C as free-floating sections before processing for immunofluorescence and CS/DS-GAG extraction.

Mouse 30 μm-thick serial coronal sections underwent antigen retrieval in 10 mM trisodium citrate (pH 8.5) and heated at 90°C for 20 m. Immunostaining was performed according to our previously published methods (Alonge et al., 2019, 2020). Briefly, free-floating tissue sections were permeabilized for 30 min at room temperature (RT) in 0.1 M PBS + 0.2% Triton X-100 and blocked for 2 h at 37°C in 0.1 M PBS + 0.05% Triton X-100 (PBST) + 10% normal donkey serum (Jackson ImmunoResearch). Sections were then incubated overnight at 4°C using 1:1,000 dilution of biotin-labeled Wisteria floribunda agglutinin (WFA; Sigma: L1516) in PBST + 1% donkey serum. Other primary antibodies used include: Iba1 (WAKO: 019-19741, RRID: AB_839504, 1:1,000), GFAP (Millipore: AB5541, RRID: AB_177521, 1:2,000), and UCHL1 (Cell Signaling: 13179, RRID: AB_2798141, 1:100). The following day, sections were washed and incubated for 2 h at RT in a 1:1,000 dilution of Alexa-fluor-conjugated secondary antibodies in PBST + 1% donkey serum. Sections were then counterstained for DAPI, mounted, and coverslipped using Fluoromount-G (ThermoFisher, 4958-02). Imaging was conducted with a Zeiss Axio Observer 7 inverted microscope with a 10× objective. Confocal microscopy was performed with a Nikon A1R HD confocal 20X objective. Images were processed with Fiji open-source imaging software. To quantify the mean fluorescent intensity (MFI) for each target we defined a set region of interest within the thalamus (see outlines in Figure 1C), converted the images to 8-bit, subtracted a constant background, and quantified the MFI normalized to the outlined area (per mm; Carulli et al., 2010) as previously described (Alonge et al., 2020). To calculate colocalization between two targets, we used the Coloc 2 plugin (Fiji) to measure the Pearson’s correlation coefficient (r value) from images that were processed for MFI quantification.

Mouse 15 μm-thick coronal sections were prepared as above, mounted on microscope slides, and stored at −80°C for analysis. In situ hybridization for mouse, CHST3 mRNA was performed by RNAScope using protocols and reagents from the manufacturer (ACD, RNAScope Multiplex Fluorescent Reagent Kit v2). Briefly, sections were dehydrated through graded ethanols, treated with hydrogen peroxide to block endogenous peroxidases, followed by antigen retrieval using a preheated kit buffer for 15 m in a 95°C waterbath. After a brief rinse with distilled water and dehydration in ethanol, slides were treated with Protease III for 30 m at 40°C in a humid chamber, rinsed, and hybridized with MmCHST3probe (ACD, 572521) for 2 h at 40°C. Slides were stored overnight at RT in 5× SSC. The following day, slides were washed in RNAScope Wash Buffer and underwent three cycles of amplification at 40°C. The CHST3 signal was developed by incubation with HRP-C1 probe for 15 m at 40°C, followed by Opal570 substrate (Akoya Biosciences FP1488001KT, 1:1,500 in ACD Amplification Buffer) for 30 m at 40°C. Slides were washed, and residual peroxidase activity blocked with HRP Block (ACD kit). Slides were then blocked for 10 m at RT with Antibody Diluent/Block (Akoya ARD1001EA), incubated with mouse anti-GFAP (Covance smi-22, RRID: AB_509980, 1:1,000) for 1 h at RT, washed, and incubated with Opal polymer HRP secondary antibody (Akoya ARH1001EA) for 10 m at RT. Slides were then washed, developed with Opal520 substrate (Akoya FP1487001KT, 1:100 in 1× Plus Amplification Diluent, Akoya FP1498) for 10 m at RT. Slides were counterstained with DAPI, coverslips mounted, and imaged as above.

CS/DS disaccharides were isolated and quantified according to our previously published methods (Alonge et al., 2019, 2020; Logsdon et al., 2021). Five PFA-fixed coronal brain sections (30 μm) from each mouse were chosen for CS/DS disaccharide extraction and quantification. Sections were washed 3× in Optima LC/MS-grade water and 1× in 50 mM ammonium bicarbonate (pH 7.6) at RT. Chondroitinase ABC enzyme (ChABC, Sigma, C3667), a combinatorial endo- and exo-lyase that selectively degrades all CS/DS-GAG chains into their individual disaccharide units (Yamagata et al., 1968; Hamai et al., 1997; Prabhakar et al., 2009; Spliid et al., 2021), was reconstituted (500 mU/ml) in 50 mM ammonium bicarbonate (pH 7.6) to digest CS/DS-GAGs at 37°C in a Thermo Scientific MaxQ4000 orbital shaker for 24 h. Supernatants were collected in sterile 1.7 ml microcentrifuge tubes and spun for 10 m at 14k× g to pellet any debris. The supernatant was then dehydrated using a SpeedVac Concentrator, and the lyophilized product was reconstituted in 30 μl of LC/MS-grade water.

Isolated CS/DS samples were analyzed using a triple quadrupole mass spectrometer equipped with an electrospray ion source (Waters Xevo TQ-S) operated in negative mode ionization. LC-MS/MS was performed using a Waters Acquity I-class ultra-performance liquid chromatographic system (UPLC) coupled to the same Waters Xevo TQ-S system. Disaccharides were resolved by porous graphitic chromatography (Hypercarb column; 2.1 × 50 mm, 3 μm; ThermoFisher) as described previously (Alonge et al., 2019, 2020; Logsdon et al., 2021). Assigned MRM channels were: 4S-CS, m/z 458 > 300; 2S4S-DS and 4S6S-CS, m/z 538 > 300; 6S-CS, m/z 458 > 282; 2S6S-CS, m/z 268 > 282; 0S-CS, m/z 378 > 175. MassLynx software version 4.1 (Waters) was used to acquire and quantify all data. Under the conditions described above, the ratios between peak areas produced from equimolar CS standard runs were normalized to the highest peak intensity and relative quantification of each CS isomer within a sample was achieved using a modified peak area normalization function (Alonge et al., 2019). Each CS isomer was expressed as a relative abundance percentage of the total CS/DS isomer composition within a brain sample. Industry-grade CS disaccharide standards 0S-CS (OC28897), 4S-CS (OC28898), 6S-CS (OC01702), 4S6S-CS (OC28899) were purchased from Carbosynth (Berkshire, UK). The 2S4S-DS (CD005) and 2S6S-CS (CD006) were purchased from Galen Laboratory Supplies (North Haven, CT). Acetonitrile (optima LC/MS-grade) and all other reagents were obtained from Fisher Scientific.

To determine whether changes in CS/DS-GAG composition correlated with TBI-induced motor dysfunction through disrupted thalamocortical circuitry (Wiley et al., 2016), a rotarod apparatus (San Diego Instruments, San Diego, CA) was used to measure motor coordination. For each of three trials, the rod gradually accelerated from 8, 16, 24, 32, and 40 rpm over a 120 s duration and remained at the final speed for an additional 30 s, if necessary. 5 m inter-trial intervals were used. The latency to fall off the rotarod was recorded for each trial. On the day of Rotarod training, each mouse was exposed to an Open Field for 5 m and distance traveled was measured to determine general physical activity.

An Open Field Apparatus (40 cm, Stoelting, Wood Dale, IL) was used to measure locomotor activity in each experimental group. Each mouse was allowed to freely explore the open field for 5 m and video recordings of the total distance traveled were analyzed using Anymaze software (Stoelting, Wood Dale, IL).

Sample sizes of N = 4/group were chosen, predicated on detecting ~80% power and alpha set at P < 0.05, with two tails. A two-sample unpaired Student’s t-test was used for two-group comparisons used GraphPad Prism^® 8.0 (Graph Pad Software, Inc., La Jolla, CA). Three or more samples were compared with a two-way analysis of variance (ANOVA). A Dunnett’s post hoc test was used to compare mice with surgery to a naïve control group. Error bars represent the standard deviation (SD) of the mean. Data met the normality assumptions of the statistical tests. Investigators were not blinded to injury conditions during euthanasia or histological preparation. Investigators were blinded to study conditions during quantitative LC-MS/MS and histological analyses.

PNN matrix detection is accomplished using immunofluorescence labeling of the PNN-associated CS/DS-GAGs using Wisteria floribunda agglutinin (WFA) in mice (Alonge et al., 2020). We examined regional changes in PNN matrix integrity in association with CCI-induced gliosis at 1 and 7 days post-CCI. We observed an association between induced astrogliosis (GFAP, Figure 1A) and loss of cortical PNNs (WFA, Figure 1B) at 7 days post-CCI in coronal sections at Bregma −1.0 mm between the ipsilateral and contralateral hemispheres from mice. These histological changes radiated from the injury site along the dorsolateral layers of the cortex and into the ipsilateral thalamus (Figure 1C). At 1 day post-CCI we observed low levels of GFAP and WFA labeling in the ipsilateral thalamus (Figures 1D,E). At 7 days post-CCI, the contralateral CCI thalamus (Figures 1F–H) and ipsilateral sham control thalamus (Figures 1I,J) also exhibited low levels of GFAP and WFA labeling (Figure 1J). However, a close examination of the ipsilateral thalamus of 7 days post-CCI mice revealed the appearance of robust astrogliosis (GFAP, Figure 1K) and WFA⁺ puncta (Figures 1L,M). These spatiotemporal changes in thalamic WFA⁺ puncta are in contrast with changes observed in cortical PNNs, which showed a strong reduction in WFA⁺ PNNs after 1 day CCI (Figures 1N–P, P = 0.0026), 7 day Sham (Figures 1Q–S, P = 0.0056), and 7 day CCI (Figures 1T–V, P = 0.0094). Unlike the thalamic WFA⁺ structures, cortical PNN morphology is comprised of dense, extracellular honeycomb-like structures, with similar net morphology seen in both uninjured contralateral (Figure 1W) and injured ipsilateral (Figure 1X) cortices at 7 days post-CCI. Although we observe normal WFA⁺ PNN morphology in both the injured and uninjured cortices at 7 days post-CCI, thalamic WFA⁺ puncta appeared only after 7 days post-CCI (Figure 1M).

Changes in WFA labeling may indicate a shift in the composition of CS/DS-GAG chains that often associate with PNN disintegrity, reminiscent of a plastic environment within the developing brain. During neurodevelopment, reduced WFA⁺ PNN abundance is associated with an increase in the relative abundance of the 6S-CS isomer (Carulli et al., 2010). Moreover, elevated 6S-CS isomer overexpression delays PNN maturation (Miyata et al., 2012) and also associates with proteolytic degradation of PNN matrices in the mouse brain (Miyata and Kitagawa, 2016). Although increased 6S-CS (Properzi et al., 2005) and PNN loss (Wiley et al., 2016) have been reported in mice after CCI, the biochemical mechanisms associated with PNN integrity are poorly understood. Therefore, we sought to explore whether CCI induces changes in CS/DS-GAG sulfation patterns in association with PNN loss, gliosis, and impaired motor coordination in mice with TBI.

Using LC-MS/MS to quantitatively analyze the relative abundance of CS/DS isomers from a series of fixed-brain coronal sections expanding the rostro caudal extent of the injury site (−1.0 to −3.0 mm AP from the Bregma; Alonge et al., 2019), we found robust changes in brain CS/DS-GAG sulfation patterns 7 days post-CCI, including hypersulfation (P < 0.0001) of CS/DS-GAGs by increased 6S-CS (P < 0.0001) and 2S6S-CS (P = 0.0001) isomers and decreased 0S-CS (P < 0.0001) and 2S4S-DS (P = 0.02) isomers, compared to naïve controls (Table 1). We also found modest changes in brain CS/DS-GAG sulfation at 7 days in sham mice compared to naïve control mice, implying the act of skull removal alone to be sufficient to induce lesser changes in brain CS/DS-GAG sulfation. As such, the degree of changes observed in brain CS/DS-GAG sulfation patterns can be highlighted as a potential indicator of injury severity.

We then predicted that the robust appearance of WFA⁺ puncta in the gliotic thalamus 7 days post-CCI (Figure 1) would also associate with changes in CS/DS-GAG sulfation patterns and contribute to the whole brain CS/DS-GAG changes as described in Table 1. In the second cohort of mice, we conducted interregional LC-MS/MS to measure CS/DS-GAG sulfation changes specifically in the ipsilateral thalamic region in naïve, sham, and CCI mice 7 days post-CCI. We found a significant increase in thalamic 6S-CS isomer (P = 0.002) and a trend for increased 2S6S-CS isomer (P = 0.07) 7 days post-CCI, and to a lesser extent at 7 days after sham, when compared to naïve controls (Table 2). Although changes in CS/DS-GAG total abundance have been reported in rodents exposed to TBI (Properzi et al., 2005), our work builds upon these initial discoveries by describing the regional and temporal changes in CS/DS-GAG sulfation patterns present in the progression of CCI/TBI injury.

Previous reports describe a well-established peak in gliosis occurring between 1 and 7 days after CCI/TBI, which can persist depending on injury severity (Burda and Sofroniew, 2014). Based on these previous reports, we next predicted that the spatiotemporal appearance of ipsilateral thalamic WFA⁺ puncta (Figure 1M) may associate with temporal expression of glial scarring (Figure 1K) within the ipsilateral thalamus. As expected, based on our initial findings described in Figure 1, we did not observe a robust appearance of WFA⁺ puncta (Figure 2A), microgliosis (Iba1, Figure 2B), or astrogliosis (GFAP, Figure 2C) in the ipsilateral thalamus at 1-day post-CCI (Figure 2D), which were also absent in the ipsilateral thalamus 7 days post-sham (Figures 2E–H). However, we observed a robust appearance of WFA⁺ puncta (Figure 2I; quantified in Figure 2M, P = 0.0012), astrogliosis (GFAP, Figure 2J; quantified in Figure 2N, P < 0.0001), and microgliosis (Iba1, Figure 2P; quantified in Figure 2S, P < 0.0001) in the ipsilateral thalamus 7 days post-CCI. Surprisingly, thalamic WFA⁺ puncta showed weak colocalization with astrogliosis (Figures 2I–L; Pearson’s r = 0.11) and microgliosis (Figures 2O–R, Pearson’s r = 0.13) within this region, suggesting the production of WFA⁺ puncta may be present in another CNS cell type.

In addition to GFAP, patients with TBI also exhibit increased biofluid expression of the clinical biomarker of neuronal repair, ubiquitin C-terminal hydrolase L1 (UCHL1; Brophy et al., 2011; Wang et al., 2021). Since thalamic WFA⁺ puncta is only weakly associated with astro- and micro-gliosis, we next examined the level of colocalization between thalamic WFA⁺ puncta and UCHL1⁺ neurons 7 days post-CCI injury. Interestingly, although protein expression of UCHL1 remained unchanged in 7-days post-sham, 1-days post-CCI, and 7-days post-CCI mice (Figure 2U; quantified in Figure 2X, P = 0.26), we discovered a moderate colocalization of UCHL1 with WFA⁺ puncta within this region (Figures 2T–W; Pearson’s r = 0.47), implying that the appearance of WFA⁺ puncta associates with neuronal repair, rather than gliosis, after CCI injury.

We next sought to determine the cellular source for increased levels of the 6S-CS isomer by measuring the expression of the 6S-CS-specific sulfotransferase (CHST3) in mice 7 days post-CCI exposure. Since the induction of astrogliosis appeared in association with increased levels of the 6S-CS isomer, we combined immunohistochemistry with RNA scope to determine whether CHST3 mRNA colocalized with GFAP immunofluorescence. As expected, we observed an increase in ipsilateral cortical CHST3 (P < 0.0001) and GFAP (P < 0.0001; Figures 3A,B; quantified in Figures 3C,D) 7 days post-CCI, complementing an increase in the 6S-CS isomer (P = 0.0007; Figure 3E) within this region compared to the uninjured contralateral cortex. This finding was recapitulated in the ipsilateral thalamus 7 days post-CCI, which also showed increased CHST3 (P < 0.0001) and GFAP (P < 0.0001; Figures 3F,G; quantified in Figures 3H,I) in association with increased thalamic 6S-CS isomer (P = 0.0243; Figure 3J) compared to the uninjured contralateral thalamus. Despite previous reports of an association between CHST3 and injury-induced GFAP (Properzi et al., 2005), CHST3 expression weakly colocalized with GFAP immunoreactivity in both the cortex (Pearson’s r = 0.22) and thalamus (Pearson’s r = 0.24).

The lack of a strong association between CHST3 and GFAP suggests additional cell type(s) may be responsible for the increase in 6S-CS sulfation after CCI. We predicted that the increase in thalamic 6S-CS may be linked to the specific appearance of thalamic WFA⁺/UCHL1⁺ cells (Figure 2W). Surprisingly, RNAscope analysis found that CCI-induced thalamic ipsilateral CHST3 expression (P = 0.0004) only weakly colocalized with UCHL1 immunoreactivity (Figures 3K,L; quantified in Figures 3M,N; Pearson’s r = 0.10) in the thalamus. Again, no difference in the thalamic expression of UCHL1 was observed at 7 days post-CCI (Figure 3N), agreeing with our previous finding (Figure 2X). The stronger association between UCHL1 and WFA⁺ puncta (Figure 2W) compared to UCHL1 and CHST3 (Figure 3L) implies the appearance of CCI-induced ipsilateral thalamic WFA⁺ puncta may not be strongly related to the increased levels of 6S-CS sulfation detected after CCI.

Mice with TBI are known to develop motor coordination deficits (Meabon et al., 2016; Bhowmick et al., 2018) that can be assessed using a rotarod task (Logsdon et al., 2020). In order to specifically address deficits in motor coordination with the rotarod task, any confounds of general locomotion impairment were addressed using the Open Field Task. Mice exposed to CCI and sham procedures did not exhibit locomotor deficits 7 days post-procedures compared to naïve mice, as evidenced by similar distances traveled in the Open field task (Figure 4A). In contrast, a significant decrease in rotarod latency was observed in mice at 7 days post-CCI compared to naïve control mice and, to a lesser extent, sham mice (Figure 4B, P < 0.0001), which demonstrates severe motor impairment selectively in these CCI-exposed mice.

The thalamus is an important brain region involved in relaying information to the cortex for motor function (Biane et al., 2016; Tanaka et al., 2018). In our model system, the cortex is mostly removed from this circuit, yet the thalamus remains intact (Figure 1A). Since 6S-CS sulfation is required for neuroplasticity and associated behavioral deficits in mice (Yang et al., 2021), we sought to determine whether the significant differences in thalamic 6S-CS abundance (Table 2) were associated with motor coordination defects in mice with CCI. As such, we correlated the latency to fall with the thalamic 6S-CS levels for naïve, 7 days post-sham, and 7 days post-CCI mice. Here, we found a strong correlation between the abundance of thalamic 6S-CS isomer and rotarod performance (Figure 4C; R² = 0.73, P < 0.001). These findings provide a functional association between thalamic CS/DS-GAG sulfation and motor coordination deficits, warranting future mechanistic studies targeting the 6S-CS isomer in preclinical models for human TBI.