Complement Component 5 (C5) Deficiency Improves Cognitive Outcome After Traumatic Brain Injury and Enhances Treatment Effects of Complement Inhibitors C1-Inh and CR2-Crry in a Mouse Model

A potent effector of innate immunity, the complement system contributes significantly to the pathophysiology of traumatic brain injury (TBI). This study investigated the role of the complement cascade in neurobehavioral outcomes and neuropathology after TBI. Agents acting at different levels of the complement system, including 1) C1 esterase inhibitor (C1-Inh), 2) CR2-Crry, an inhibitor of all pathways acting at C3, and 3) the selective C5aR1 antagonist, PMX205, were administered at 1 h post-TBI. Their effects were evaluated on motor function using the rotarod apparatus, cognitive function using the active place avoidance (APA) task, and brain lesion size at a chronic stage after controlled cortical impact injury in C5-sufficient (C5+/+) and C5-deficient (C5–/–) CD1 mice. In post-TBI C5+/+ mice, rotarod performance was improved by CR2-Crry, APA performance was improved by CR2-Crry and PMX205, and brain lesion size was reduced by PMX205. After TBI, C5–/– mice performed better in the APA task compared with C5+/+ mice. C5 deficiency enhanced the effect of C1-Inh on motor function and brain damage and the effect of CR2-Crry on brain damage after TBI. Our findings support critical roles for C3 in motor deficits, the C3/C5/C5aR1 axis in cognitive deficits, and C5aR1 signaling in brain damage after TBI. Findings suggest the combination of C5 inhibition with C1-Inh and CR2-Crry as potential therapeutic strategies in TBI.


Introduction
Severe traumatic brain injury (TBI) causes significant chronic neurological, cognitive, and behavioral disability, resulting in an immense socioeconomic burden.Effective treatments are required to combat the chronic effects of TBI.3][4][5] Targeting the complement system is a potential therapeutic target for developing future treatments for TBI.
The complement system comprises >40 proteins involved in a series of enzymatic cleavages and membrane binding events.It plays a critical role in clearing pathogens, dying cells, and misfolded proteins and can be activated by the classical, lectin, or alternative pathways, resulting in activation of common major effectors.Complement components 3 and 5 (C3 and C5) lie at the center of the complement system. 6C3 is cleaved to form C3a, which promotes chemotaxis and activation of microglia through the receptor C3aR, and C3b which mediate opsonization.Cleavage of C5 mediates proinflammatory effects by: 1) generation of the C5a anaphylatoxin that, primarily through its G-protein-coupled receptor C5aR1, initiates inflammatory responses including recruitment and activation of inflammatory cells, blood-brain barrier disruption, and cytokine release; 2) the terminal pathway, which culminates in the formation of the cytolytic membrane attack complex (MAC). 7ysregulation at different levels of the complement cascade may have different effects on neuropathological and -behavioral outcomes post-TBI.Hence, we hypothesized that the administration, after TBI, of agents targeting different levels of the complement cascade have different effects on motor and cognitive function and brain damage.
To test this hypothesis, we used three inhibitors that act at different levels of the complement system (depicted in Fig. 2B): 1. C1 esterase inhibitor (C1-Inh).C1-Inh is a member of the serpin family of protease inhibitors, which inhibits the complement system, contact (kinin) system, and fibrinolytic/coagulation system. 8Human-derived C1-Inh, a U.S. Food and Drug Administration (FDA)-approved drug, is active in mice; hence, it can be tested in murine models. 92. Complement receptor 2-complement receptor 1-related gene/protein (CR2-Crry).Crry is a membrane complement regulator that inhibits C3 convertase.A CR2-targeting moiety targets Crry to the site of complement activation and C3d deposition, 10,11 allowing recombinant CR2-Crry to act as a C3 inhibitor at the injury site.3. PMX205.3][14] The inclusion of a C5aR1 antagonist allowed us to determine whether any observed benefit of C5aR1 blockade in TBI is mediated through the attenuation of C5a-C5aR1 signaling.
Up to 39% of murine strains, including C57BL/10Sn, DBA/2J, A/HeJ, AKR/J, NZB/BINJ, SWR/J, and B10.D2/oSnJ, have been reported to be C5 deficient and to lack detectable blood levels of C5. 15 In C5deficient mouse strains, the 2-base-pair ''TA'' deletion at positions 661 and 662 of the C5 messenger RNA (mRNA) coding frame results in failure to secrete C5 protein, although it is biosynthesized. 15C5-deficient mouse strains represent valuable tools for investigating the role of C5 in TBI. 16We hypothesized that the absence of C5 affects functional outcomes after TBI and influences the therapeutic effects of complement inhibitors by comparing them in C5 +/+ and C5 -/- CD1 mice, an outbred strain that is commonly used for testing the efficacy and safety of new drugs in TBI and post-traumatic epilepsy models.

Experimental design
First, we used enzyme-linked immunosorbent assay (ELISA) to compare the C5a level in brains of shaminjured and TBI mice (n = 8 in each group) at 4 h post-injury.Because the variation among animals was very high in the brain samples, we genotyped mice for the TA deletion reported in some outbred mouse strains (Fig. 1A).Then, we examined the treatment effects of C1-Inh, CR2-Crry, and PMX205 in C5 +/+ and C5 -/-mice from three cohorts of animals (Figs.2A and 3A).Motor function was assessed using the rotarod task before controlled cortical impact (CCI) injury and at 1 and 2 weeks after TBI.Spatial learning ability was examined using the active place avoidance (APA) task over a 5-day testing period at 5 weeks post-TBI.Size of the brain lesion was measured on sections stained with cresyl violet, obtained 16 weeks after TBI.Figures 2B and 3B are summary diagrams of the complement pathways and the site of action of the inhibitors used in C5 +/+ and C5 -/-mice.Sample sizes of treatment groups and different assays are summarized in Supplementary Table S1.

Animals
Outbred CD1 male mice (Envigo, Indianapolis, IN) were used in this study.Mice were housed individually under controlled laboratory conditions (12-h light/ dark cycle, with lights on at 7:00

Genotyping
DNA was extracted from ear-notch tissue samples using standard HotSHOT DNA extraction procedures.DNA sequencing and mutagenically separated polymerase chain reaction (MS-PCR) methods were used to identify the TA deletion.For sequencing, a 543-basepair (bp) fragment from the C5 gene that encompassed the site of the potential TA deletion was amplified using PCR.Primer sequences were C5-F1: TAGG AATTAGTTAAATTGTCTAGGG and C5-R1: GATT CAGCTACTCGTAGTTAC.The PCR annealing temperature was 52°C.PCR fragments were sequenced for detection of the TA deletion as previously described. 17MS-PCR-amplified PCR fragments were detected by capillary electrophoresis.

Controlled cortical impact injury model
Adult mice, 9-10 weeks of age, were subjected to a severe unilateral cortical contusion by computercontrolled impact delivered by a beveled steel tip 3 mm in diameter (TBI-0310; Precision Systems and Instrumentation, Fairfax, VA) as previously described. 18Briefly, mice were deeply anesthetized with a mixture of tiletamine/zolezepam (Zoletil Ò 100, 50 mg/kg; Virbac, Carros, France) and xylazine (20 mg/kg; Troy Laboratories, Glendenning, NSW, Australia) administered, intraperitoneally (i.p.), before being placed in a stereotaxic frame (World Precision Instruments, LLC, Sarasota, FL).The skull was exposed by a midline incision, before a 4-mm craniotomy was made just lateral to the sagittal suture and centered between the bregma and lambda, to allow removal of the skullcap without damage to the dura.A CCI injury was subsequently delivered by compressing the cortex of the left hemisphere to a depth of 2.0 mm at a velocity of 5.0 m/s for a duration of 100 ms.The incision was sutured without cranioplasty.Sham controls underwent the craniotomy procedure, but did not receive the CCI injury.

Drug treatments
A single dose of C1-Inh (15.0 IU; Berinert Ò ; CSL Behring GmbH, Marburg, Germany) or 0.9% saline solution was administered, intravenously (i.v.), by the tail vein 1 h after the CCI injury; CR2-Crry (10 mg/kg) or phosphate-buffered saline (PBS) were administered, i.v., by the tail vein at 1 and 7 h post-CCI; and PMX205 (1 mg/kg, synthesized in-house 19 ) or 5% glucose vehicle solution were administered by i.p. injection at twice-daily intervals, commencing 1 h post-TBI over a 10-day period.The i.v.injections were performed with mice under isoflurane anaesthesia, ‰ FIG. 1. C5 deficiency in outbred CD1 mice.(A) Schematic representation of the timeline for the experimental procedures.Screen shots of representative data for sequencing (B) and MS-PCR (C) from C5 +/+ (top), C5 +/- (middle), and C5 -/-(bottom) outbred CD1 mice.In the sequencing screenshots, the TA base-pair that followed TGCA after sequence number 100 was deleted in the mutational C5 allele.The MS-PCR screenshots showed that PCR fragments with two different sizes were present in C5 +/+ , C5 +/-, and C5 -/-mice.(D) Brain C5a levels in sham and CCI-injured C5 +/+ , C5 +/-, and C5 -/-mice 4 h after injury.C1-Inh, C1 esterase inhibitor; CCI, controlled cortical impact; ELISA, enzyme-linked immunosorbent assay; MS-PCR, mutagenically separated polymerase chain reaction.and the i.p. injections were administered with mice restrained.Doses and administration times were based on the available literature documenting the effectiveness of each complement inhibitor, or the known half-life of the agent. 10,14,20Control mice received the equivalent amount of vehicle at each dosing interval.

Motor function
The rotarod apparatus (Ugo Basile, Comerio, Italy) was used to assess motor function.Mice underwent initial training on an accelerating rotarod (4-16 rpm over 60 sec) in three trials separated by 5-min intervals.
Motor function was subsequently assessed commencing the day before CCI injury and at 1 and 2 weeks after TBI, in three trials on an accelerating rotarod (4-40 rpm over 180 sec) with a minimum 15-min intertrial interval.Average time to fall from the rotating cylinder in each trial was recorded as the fall latency.

Active place avoidance task
The APA apparatus (Bio-Signal Group, Acton, MA) consisted of an elevated arena (diameter 77 cm) with a metal grid floor (bar spacing 5 mm and diameter 3 mm), surrounded by a transparent cylindrical boundary (height, 32 cm).A visual cue consisting of a large black and white symbol/shape (A3 in size) was located on each of the four walls that housed the APA apparatus.The arena rotated counterclockwise (1 rpm), and a 60-degree region of the grid, which remained constant in relation to the room coordinates, served as the shock zone.A mild electric foot shock (500 ms, 60 Hz, 0.6 mA at 1.5-sec intervals) was delivered when the mouse entered the shock zone.
Mice were handled for *30-60 sec daily in the week before testing.On the day before commencing testing, mice were placed in the arena without visual cues and shocks for 5 min, to allow habituation to the testing environment and APA apparatus.Mice subsequently underwent 10-min trials over 5 consecutive days.An overhead tracking camera recorded each trial.Trials were subsequently analyzed using Track Analysis software (Bio-Signal Group), to determine the number of shocks received and shock-zone entries.
C5a enzyme-linked immunosorbent assay Blood samples (*0.5 mL) were collected by cardiac puncture before perfusion, transferred into ethylenediaminetetraacetic acid tubes, mixed immediately with 10 lL of complement serine protease inhibitor FUT175 (5 mg/mL; BD Biosciences, San Jose, CA) to prevent ex vivo activation, 21 and stored on ice.Plasma was collected after centrifugation at 13,000g for 10 min at 4°C.Injured ipsilateral hemispheres were snapfrozen in liquid nitrogen and stored at À80°C, before being ground to fine powder using a mortar and pestle on dry ice.Powdered brain tissue was subsequently dissolved in 1 mL of NP-40 lysis buffer (Invitrogen, Carlsbad, CA) containing 1 mM of phenylmethylsulfonyl fluoride (reconstituted in in dimethyl sulfoxide; Sigma-Aldrich, St. Louis, MO), 92.6 lM of FUT175 (BD Biosciences), and 10 lL of protease inhibitor mixture (Sigma-Aldrich).Dissolved brain tissue samples were then vortexed for 1 min, left to stand on ice for 1 h, centrifuged at 13,000g for 30 min at 4°C, and the supernatant subsequently collected.Both plasma samples and brain tissue supernatants were stored at À80°C.Protein concentrations in brain tissue supernatants were determined using the bicinchoninic acid protein assay procedure (ThermoFisherScientific, Waltham, MA).C5a expression levels in plasma and brain tissue supernatants were determined by ELISA (R&D Systems mouse C5a DuoSet ELISA; R&D Systems, Minneapolis, MN), according to the manufacturer's instructions.C5a concentrations in brain tissue samples were adjusted to the protein concentration present in the sample and expressed as pg/lg protein.

Tissue preparation for histological procedures
At 16 weeks post-TBI, mice were administered a lethal dose of sodium pentobarbital (600 mg/kg, i.p.) and perfused transcardially with 20 mL of PBS (pH, 7.4), followed by 30 mL of 4% paraformaldehyde dissolved in PBS.The brain was subsequently removed and postfixed overnight in the same fixative solution, before being transferred to 15% and 30% sucrose in PBS for cryoprotection, frozen in FSC 22 Ò (Leica Microsystems, Wetzlar, Germany) frozen embedding medium on dry ice, and stored at À20°C until sectioning.Coronal sections (20 lm, between bregma À1.23 and À2.03 mm) were cut on a Leica cryostat, mounted directly to Super-FrostPlus Ò slides, and subsequently stored at À20°C.

Cresyl violet staining and assessment of lesion size
Series of every one in ten sections (20 lm thickness, 200 lm apart) were collected on one slide at the site of the injury.Slides were allowed to return to room temperature and briefly washed in deionized water (10-15 sec).They were then passed through 70% ethanol (3 min), 100% ethanol (2 • 3 min), and xylene (2 • 3 min) to defat the sections; before being rehydrated by passing back through 100% ethanol (2 • 1 min), 70% ethanol (1 min), and briefly rinsed in deionized water (10-15 sec).Sections were then transferred to 0.1% cresyl violet acetate in acetate buffer (5 mL of 1 M of sodium acetate in a final volume of 200 mL adjusted to pH: 3.85-3.90with glacial acetic acid) for 5-10 min, before differentiation in 70% ethanol (30 sec), dehydration in 100% ethanol (2 • 1 min), and cleared in xylene (2 • 3 min).Sections were subsequently mounted with DPX Ò neutral mounting medium.Lesion size was measured and averaged from three brain sections at anteroposterior levels À1.50, À1.70, and À1.90 mm from bregma using ImageJ (National Institutes of Health, Bethesda, MD). 22Percent area of tissue loss in the ipsilateral hemisphere was calculated as shown below.
area of contralateral hemisphere À area of ipsilateral hemisphere area of contralateral hemisphere • 100 where ipsi-and contralateral respectively refer to the cerebral hemisphere ipsi-and contralateral to the site of injury.

Statistical analysis
Statistical comparisons were performed using Graph-Pad Prism Ò software (version 8.3; GraphPad Software Inc., La Jolla, CA).Normality of continuous data was assessed using the Shapiro-Wilk test, and homoscedasticity was assessed using Bartlett's test.Where the assumptions of normality and homoscedasticity were not violated, data were analyzed using unpaired Student's t-tests, one-way analysis of variance (ANOVA), or repeated-measures two-way ANOVA as appropriate.Non-parametric analyses were performed if these assumptions were violated as specified in the Results section.The criterion for statistical significance was p < 0.05.Data are presented as meanstandard error of the mean values.
CR2-Crry, but not C1-Inh or PMX205, improved rotarod performance after controlled cortical impact injury In this study, we examined the therapeutic effect of complement inhibitors on motor and cognitive functions in C5 +/+ mice.The contribution of C5 to brain damage and the influences of C5 deficiency on the therapeutic impacts of complement inhibitors were studied using C5 -/-mice.
We investigated the effects of complement inhibitors on rotarod performance by comparing the effects of each inhibitor against the effects of vehicle treatment in C5 +/+ mice.Because fall latency at 1 and 2 weeks after CCI was not significantly different between vehicle-treated groups (one-way ANOVA, F (3, 48) = 0.74, p = 0.53 and F (3, 47) = 2.00, p = 0.13), we combined data from all vehicle-treated mice.We examined rotarod performance in C5 +/+ mice after CCI injury and found that fall latencies at 1 and 2 weeks after CCI were significantly lower than the pre-CCI baseline (Kruskal-Wallis' test, ****p < 0.0001; Dunn's multiple comparisons test, *p < 0.05, ****p < 0.0001; Fig. 2D).In the injured C5 +/+ mice, CR2-Crry treatment improved rotarod performance to the pre-CCI baseline level at 1 week post-CCI, and fall latency of CR2-Crry-treated mice was significantly higher compared with vehicle-treated mice (Mann-Whitney U test, *p < 0.05; Fig. 2G,H).In the injured C5 +/+ mice, fall latency over the 2-week testing period was not significantly affected by C1-Inh treatment (Mann-Whitney U test, p = 0.85 at 1 week post-CCI and p = 0.25 at 2 weeks post-CCI; Fig. 2E,F) and PMX205 treatment (Mann-Whitney U test, p = 0.74 at 1 week post-CCI and p = 0.24 at 2 weeks post-CCI; Fig. 2I,J).
Taken together, these findings indicate that inhibition of all complement pathways at the C3 level with CR2-Crry can accelerate the recovery of motor function.However, inhibition of the classic and lectin pathways with C1-Inh, and inhibition of C5aR1 signaling with PMX205, did not alleviate deficits in motor function post-TBI.

C5 deficiency promoted the effects of C1-Inh and CR2-Crry on the recovery of motor function
We compared rotarod performance between C5 +/+ and C5 -/-mice before CCI injury and observed no significant difference (Mann-Whitney U test, p = 0.41; Fig. 3C), indicating that these genotypes had no effect on motor function in CD1 mice, as is the case for other inbred C5-deficient mouse strains. 24Because fall latency at 1 and 2 weeks after CCI was not significantly different between vehicle-treated groups (one-way ANOVA, F (3, 20) = 0.6073, p = 0.62 and F (3, 21) = 2.55, p = 0.083), we combined data from all vehicle-treated mice.In the injured C5 -/-mice, fall latencies at 1 week, but not at 2 weeks, after CCI were significantly lower than the pre-CCI baseline (Kruskal-Wallis' test, *p < 0.05; Dunn's multiple comparisons test, *p < 0.05; Fig. 3D).We examined rotarod performance in C5 +/+ and C5 -/-mice after CCI injury by comparing fall latency at 1 and 2 weeks post-CCI and observed no significant main effect on fall latency for genotype (Mann-Whitney U test, p = 0.67 at 1 week post-CCI and p = 0.90 at 2 weeks post-CCI; Fig. 3E,F), suggesting that C5 deficiency does not alter the recovery of motor function after CCI.
In the injured C5 -/-mice, we also observed a significant increase in fall latency at 1 week after CCI for CR2-Crry-treated mice compared with vehicle-treated mice (Mann-Whitney U test, *p = 0.02 at 1 week post-CCI and p = 0.11 at 2 weeks post-CCI; Fig. 3I,J).Though C1-Inh failed to improve the recovery of motor function after TBI in C5 +/+ mice, in the injured C5 -/-mice treated with C1-Inh, fall latency of C1-Inhtreated mice was significantly higher compared with vehicle-treated mice at 1 week after CCI (Mann-Whitney U test, *p < 0.05 at 1 week post-CCI and p = 0.20 at 2 weeks post-CCI; Fig. 3G,H), suggesting that C5 deficiency enhanced the treatment effect of C1-Inh.Similar to injured C5 +/+ mice, there was no significant effect on fall latency in the rotarod test after treatment with PMX205 in CCI-injured C5 -/-mice (Mann-Whitney U test, p = 0.39 at 1 week post-CCI and p = 0.14 at 2 weeks post-CCI; Fig. 3K,L).
CR2-Crry and PMX205, but not C1-Inh, improved performance in the active place avoidance task after controlled cortical impact injury APA is a challenging spatial learning task that relies on hippocampal function.In a 5-day APA testing para-digm, TBI mice made significantly more mistakes than sham-operated mice and failed to show an improvement in performance over time. 25We assessed spatial learning ability using the APA task over a 5-day testing period at 5 weeks after CCI injury (Fig. 4A).Improvement during the APA task was analyzed by calculating the reduction in shock number compared to the number on the first trial in each mouse and expressing this value as a percentage.Average velocity was used to examine the movement during APA testing.Among vehicle-treated groups, there were no significant main effects on shock number (repeatedmeasures two-way ANOVA, F (3, 73) = 2.50, p = 0.066), percent improvement (repeated-measures two-way ANOVA, F (3, 73) = 2.36, p = 0.23), and velocity during the 5-day testing period (repeated-measures two-way ANOVA, F (3, 69) = 0.67, p = 0.57).Hence, we combined data from all vehicle-treated mice and compared them with the data sets for each complement inhibitor treatment.
Taken together, these findings support a role for complement activation at the level of C3 or C5aR1 in the loss of cognitive function after CCI injury.
C5 deficiency improved performance in the active place avoidance task, but did not further enhance the treatment effects of C1-Inh, CR2-Crry, and PMX205 To test for the effects of genotype, we compared shock numbers, percent improvement, and average velocity of the fifth trial with the first trial in vehicle-treated C5 +/+ and C5 -/-mice.Shock number was significantly lower and percent improvement was higher on day 5 compared with day 1 in injured C5 -/-mice treated with vehicle (Friedman's test, **p < 0.01; Dunn's multiple comparisons test, ***p < 0.001; Fig. 5A and Friedman's test, **p < 0.01; Dunn's multiple comparisons test, ***p < 0.01; Fig. 5B).In the injured C5 -/-mice treated with vehicle, average velocity was also slightly, but significantly, lower on day 5 than in day 1 (Friedman's test, ***p < 0.001; Dunn's multiple comparisons test, ****p < 0.0001; Fig. 5C).These findings indicate that C5 deficiency attenuates loss of cognitive function after CCI injury.
Like vehicle treatment, average velocity was also reduced on day 5 compared to day 1 in injured C5 -/- mice treated with these treatments (Friedman's test, *p < 0.05; Dunn's multiple comparisons test, *p < 0.05; PMX205, but not C1-Inh and CR2-Crry, reduced lesion size after controlled cortical impact injury We measured lesion size using percentage area of the hemisphere ipsilateral to the site of impact that was injured on brain sections stained with cresyl violet (Fig. 6A-E) at 16 weeks after CCI injury.There was no significant difference in lesion size between vehicletreated groups (one-way ANOVA, F (3,58) = 2.57, p = 0.06); hence, the effect of each complement inhibitor on lesion size was compared against pooled data for their vehicle treatments.In the injured C5 +/+ mice, lesion size was significantly reduced after treatment with PMX205 (t (55) = 2.28, p = 0.027, unpaired Student's t-test; Fig. 6H), but not after treatment with C1-Inh (t (49) = 1.88, p = 0.07, unpaired Student's t-test; Fig. 6F) and CR2-Crry (t (51) = 0.87, p = 0.39, unpaired Student's t-test; Fig. 6G).

Discussion
In the present study, we examined the effects of C1-Inh, CR2-Crry, and PMX205 on motor and cognitive function, and brain lesion size after TBI in C5 +/+ and C5 -/-mice.As summarized in Figure 8A, CR2-Crry improved rotarod and APA performances, and PMX205 improved APA performance and reduced brain lesion size in C5 +/+ mice.C5 deficiency improved performance in the APA task after CCI injury.C5 deficiency enhanced the effect of C1-Inh on rotarod performance.C1-Inh and CR2-Crry also reduced brain lesion size in C5 -/-mice.
C3 degradation products C3a/C3b and membrane attack complex signaling are the main drivers of motor deficits In C5 +/+ mice, we showed that CR2-Crry treatment improved motor function after TBI, whereas PMX-205 and C5 deficiency had no significant effects, supporting the critical role of C3a/C3b in motor deficits post-TBI (Fig. 8B).7][28] The therapeutic efficacy of CR2-Crry has been assessed using a battery of motor function tasks post-TBI in the mouse CCI model. 10reatment with CR2-Crry substantially improved motor function and suppressed microgliosis and astrogliosis post-TBI. 10,11In this study, CR2-Crry treatment improved recovery in the rotarod test of motor function, further supporting the critical role of C3 in motor deficits post-TBI.In previous research, administration of C1-Inh 1 h after CCI injury in mice attenuated motor deficits, assessed using a composite neuroscore that included lateral pulsion, forelimb function, hindlimb function, and the angle board test. 29In a rat weight-drop model, treatment with C1-Inh before TBI induction reduced brain edema 2 days after injury. 30These studies have attributed the beneficial effects of C1-Inh to inhibition of the complement system, even though brain C3a levels were reduced by only 15% after treatment with C1-Inh. 30However, as assessed in the rotarod test, we found that C1-Inh administered 1 h after CCI injury did not improve grip strength, balance, and motor coordination.This discrepancy may reflect the different methods used to assess motor function.
Although C5 deficiency and PMX205 treatment by themselves did not affect recovery in the rotarod test, C5 deficiency further enhanced the effect of C1-Inh on motor function.Our findings support a downstream role for MAC signaling.Our ELISA data demonstrated the presence of C5a in brain tissue in C5 -/-mice, perhaps reflecting release from damaged brain cells.It appears that even reduced amounts of C5/C5a released in the brain after TBI contribute to motor deficits in C5 -/-mice.Whereas blockade of the C5a receptor with PMX205 did not have a significant effect, we postulate that further reductions in MAC signaling from targeting upstream of the complement system with C1-Inh was sufficient to improve motor function in C5 -/-mice.
The C3/C5/C5a receptor axis plays a detrimental role in cognitive function after traumatic brain injury Consistent with previous studies in mice where treatment with C1-Inh 1 h after CCI injury did not improve cognitive function assessed using the Morris water maze, 29 we found that treatment with C1-Inh did not improve performance in the APA task.We showed that CR2-Crry improved performance in the APA task after CCI injury in mice, supporting the key role of C3 in cognitive deficits after TBI.Treatment with CR2-Crry has also been reported to substantially improve cognitive function assessed using the Barnes maze. 10 Complement activation upstream of C3 promotes chronic TBI neuropathology, but the contribution of downstream molecules is still unclear.The use of C5deficient mouse strains has led to significant progress in understanding normal and pathological functions of the C5 gene. 24,31Previous work in a TBI cryoinjury model has shown that neutrophil extravasation was significantly reduced in C5 -/-mice, but the study did not address whether C5 deficiency affected functional outcomes. 16By comparing C5 +/+ and C5 -/-CD1 mice after CCI injury, we found that C5 deficiency significantly improved performance in the APA task.
Inhibition of the terminal MAC using CR2-CD59 post-CCI injury in mice was found to reduce acute deficits, but not prevent chronic deficits, indicating that opsonins and anaphylatoxins contributed to chronic neuroinflammation in a MAC-independent manner. 10n a previous study, C5aR1 mRNA expression was upregulated in cerebellar Purkinje cells in rats after diffuse axonal injury, 32 whereas a C5aR1 antagonist peptide reduced neutrophil extravasation in a cryoinjury mouse model, suggesting that inhibition of the C5a receptors may ameliorate TBI. 16We examined the specific role of the C5aR1 in cognitive function and found that inhibition of C5aR1 using PMX205 improved performance in the APA task.Treatment with CR2-Crry and PMX205 and C5 deficiency improved performance in the APA task, indicating that activation of the C3/C5/C5aR1 axis all play detrimental roles in cognitive function post-TBI (Fig. 8B).
C5/C5a/C5aR1 signaling contributes to brain damage We found that brain lesion size did not differ significantly between C5 +/+ and C5 -/-mice post-CCI injury whereas treatment with PMX205 reduced brain lesion in both C5 -/-and C5 +/+ mice.These findings suggest a key role for C5aR1 in brain damage.That PMX205 also reduced brain lesion in C5 -/-mice is in keeping with the ELISA data demonstrating the presence of low levels of C5a in brain tissue in these animals.Even these small amounts of brain C5/C5a and a low level of C5aR1 activation could be sufficient for inducing brain damage, potentially also explaining the protective effects of C1-Inh and CR2-Crry in C5 -/-mice.In C5 +/+ mice, it is likely that targeting upstream parts of the complement cascade, using C1-Inh and CR2-Crry, do not inhibit downstream C5aR1 signaling fully, whereas in C5 -/-mice, the combination of C5 deficiency with C1-Inh or CR2-Crry provides more effective inhibition of C5aR1 signaling resulting in brain-protective effects.In previous studies, MAC inhibition significantly reduced lesion volume at 30 days after TBI, but the effect is less robust compared with the inhibition of C3. 10 These findings support a critical role for downstream elements of the complement system in brain damage post-TBI.
Although treatment with CR2-Crry5 and PMX205 in C5 +/+ mice, as well as C5 deficiency, improved performance in the APA task, it appears from the histological images (Figs. 6 and 7) that the hippocampus is still severely damaged after these treatments.Spatial cognition involves a network of structures, such as the medial pre-frontal cortex, in addition to the ventral hippocampus. 33We focused on the hippocampus in this study and did not observe the protective effect on the hippocampus, suggesting that improved performance after CR2-Crry and PMX205 treatment are related to protective effects elsewhere in the network.

Limitations
A limitation of our study is that we did not validate the effects of C5 deficiency by using C5 inhibitors, such as C5 targeting monoclonal antibodies.It is possible that deep anesthesia during surgery added a component of hypoxic brain injury, and during drug administration, depending on the treatment, some animals had an additional inhalational anesthetic agent or were restrained, which may alter inflammation/immune function.However, we consider the potential degree of added variability to be mild and unlikely to negate our findings.

Future direction
Our study demonstrated that C3 inhibition alleviates both motor and cognitive deficits after TBI, suggesting CR2-Crry as a potent therapeutic strategy in TBI.Our findings also suggest that even a low level of C5 released in damaged brain tissue is sufficient to contribute to motor deficits and brain damage.Upstream targeting of the complement cascade by CR2-Crry may not achieve adequate inhibition of terminal components; C5 inhibition/deficiency enhanced the treatment effects of CR2-Crry on brain damage.The findings suggest that combination of C3 and C5 inhibitors as an attractive therapeutic strategy for TBI.A previous study demonstrated that alternative completement pathway inhibitor CR2-fH provided similar beneficial effects to CR2-Crry in the CCI model. 10A few humanized C5 antibodies have been approved for clinical use or are advanced in their clinical development.For instance, eculizumab and ravulizumab (ALXN210) are FDA approved for treating paroxysmal nocturnal hemoglobinuria. 34,35BB5.1 is a mouse analogue to eculizumab and ravulizumab, and future studies should examine the treatment effects of the combination therapy of BB5.1 and C3 inhibitors on TBI outcomes.

Conclusion
We evaluated the effectiveness of treatment with three complement inhibitors targeting different levels of the complement system in both C5 +/+ and C5 -/-CD1 mice post-CCI injury.Our findings suggest a combination treatment using a C5 inhibitor together with CR2-Crry as a potential novel therapeutic strategy for TBI.In addition, this study demonstrates that genetic variability in CD1 mice is a fact that can affect their response to injury.

FIG. 8 .
FIG. 8. Effects of complement inhibitors in CCI injury.(A) Summary of the treatment effects of complement inhibitors after CCI injury in the present study.(B) Role of the complement cascade in functional deficits and brain damage after CCI injury.APA, active place avoidance; C1-Inh, C1 esterase inhibitor; CCI, controlled cortical impact; RR, rotarod.
AM, temperature 22°C -1°C, air humidity 50-60%) with ad libitum access to food and water.Experimental procedures were approved by the University of Queensland Animal Ethics Committee (Approval No.: CAI/300/17) and the Animal Care and Use Review Office of the U.S. Army Medical Research and Development Command.All experiments were conducted in accordance with the guidelines of the Australian National Health and Medical Research Council.Behavioral tests were performed by operators blinded to genotypes and treatments.