Dose–response,therapeutic time-window and tPA-combinatorial efficacy of compound 21: A randomized,blinded preclinical trial in a rat model of thromboembolic stroke

Abstract

The aim of this translational, randomized, controlled, blinded preclinical trial was to determine the effect of compound 21 (C21) in embolic stroke. Rats were subjected to embolic-middle cerebral artery occlusion (eMCAO). They received C21 (0.01, 0.03 and 0.06 mg/kg/d) or saline (orally) for five days, with the first-dose given IV at 3 h post-eMCAO. For the time-window study, the optimal-dose of C21 was initiated at 3, 6 or 24 h post-eMCAO and continued for five days. For the combinatorial study, animals received IV-tissue plasminogen activator (tPA) at either 2 or 4 h, with IV-C21 (0.01 mg/kg) or saline at 3 h post-eMCAO and daily thereafter for five days. After performing the behavior tests, brains were collected for analyses. The dose–response study showed significant motor improvements with the lowest-dose (0.01 mg/kg) of C21. In the time-window study, this same dose resulted in improvements when given 6 h and 24 h post-eMCAO. Moreover, C21-treated animals performed better on the novel object recognition test. Neither the single treatment with C21 or tPA (4 h) nor the combination therapy was effective in reducing the hemorrhage or infarct size, although C21 alone lowered sensorimotor deficit scores post-eMCAO. Future studies should focus on the long-term cognitive benefits of C21, rather than acute neuroprotection.

Introduction

Stroke has dropped from the fourth to the fifth leading cause of death in the United States. ¹ This decline in stroke mortality is likely due to better acute management and more effective prevention. Tissue plasminogen activator (tPA) is still the only agent approved by the US Food and Drug Administration (FDA) for stroke treatment and is currently delivered to less than 5.0% of patients, due to its narrow time window and high risk-to-benefit ratio.^2,3 Finding a safe and effective therapeutic strategy for stroke remains a high priority.

Many promising drugs have failed clinical trials. These failures may have been attributable, to some extent, to a failure to conform with the Stroke Therapy Academic Industry Roundtable (STAIR) guidelines.^4,5 More recently, the RIGOR guidelines^6,7 for effective translational stroke research concluded that pre-clinical studies should consider the method of blinding, randomization, power/ statistical analysis and combination with tPA, in different and clinically relevant animal models, aged animals, animals with comorbidities, long-term outcome, dose–response/treatment time-window evaluation and without exclusion of any data.

A number of studies suggested that angiotensin type 2 receptor (AT2R) stimulation has beneficial effects in several experimental neuronal injury models, including ischemic stroke.^8,9 For instance, AT2R knockout mice exhibited larger infarcts compared to wild-type/controls. ¹⁰ Moreover, it has been demonstrated that intracerebroventricular administration of CGP, a selective peptide AT2R agonist, improves stroke outcome in a conscious rat model of stroke. ¹⁰ Despite its neuroprotective properties in animal studies, the clinical use of CGP is limited by its poor pharmacokinetic properties. Compound C21 (C21) is a novel selective non-peptide AT2R agonist (4000-fold more selective to AT2R versus AT1R).^11,12 It has a molecular mass of 497.6 g/mol, oral bioavailability of 20–30% and a half-life of 4 h in rats. ¹¹ Being hydrophilic, it is water soluble and has poor CNS penetration through the intact blood–brain barrier (BBB). ¹³ However, C21 is thought to be able to adequately reach the brain after BBB disruption, showing therapeutic efficacy in a variety of experimental stroke models.^8,14,15 Recent studies from our laboratory have shown promising neurovascular and neurobehavioral benefits of C21 in a mechanical vessel occlusion model of stroke.^14,15 C21 was shown to improve stroke outcome through a number of different mechanisms, including upregulation of endothelial nitric oxide synthase and growth factor expression, in addition to amelioration of oxidative stress and inflammation. ⁸

Most studies of C21 as a neuroprotective agent in stroke have been done with mechanical or intraluminal middle cerebral artery occlusion (MCAO) models.^9,14,15 Thromboembolic occlusion of cerebral arteries is the most common type of focal ischemic stroke in humans.^16,17 In this study, we used the more clinically relevant embolic stroke model (eMCAO), the only technical and physiologically suitable choice to test fibrinolytic therapy with tPA. The use of the eMCAO model allows us to evaluate the impact of C21 alone and in combination with tPA. The present study was performed in accordance with both the STAIR and RIGOR guidelines (translational model, randomization, preclinical trial, blinding, and power analysis) to determine C21's dose–response, therapeutic time-window and tPA-combinatorial effects on outcome in an embolic stroke model.^18,19

Materials and methods

Animals and study design

A total of 257 Male Wistar rats (320–340 g; 8–10 weeks Charles River Laboratories, Wilmington, MA) were used in the present study. We used young males exclusively in these experiments, knowing that there could be important sex differences in the outcome of embolic stroke that are dependent upon the hormonal cycling of the females in response to C21 treatment. The animals were housed in individual cages in a room maintained at 21–25℃, 45–50% humidity and a 12-h light/dark cycle with free access to food and water. The study is reported here in accordance with the ARRIVE (Animal Research: Reporting in Vivo Experiments) guidelines. ²⁰ All experiments were conducted according to the procedures approved by the Institutional Animal Care and Use Committee (Protocol #14-02-064) of the Charlie Norwood VA Medical Center, and followed the highest standards of experimental rigor (STAIR criteria NIH guide for care and use of laboratory animals, anesthetized and temperature controlled, housing and husbandry, randomization, blinding, interpretation and statistical analysis). The rats were allocated to treatment groups using SAS® PROC PLAN this was provided by the biostatistician. Treatments were prepared and coded based on the randomization schemes by experimenters who knew the codes but were otherwise not involved in any of the experiments or data analysis. For IV injections, C21 or saline-filled syringes were coded and provided to the surgeon. For oral gavage, C21 and saline were prepared in coded solutions to be administered by the researcher performing the behavioral testing. Coded results after each experiment were sent to the biostatistician for analysis. Therefore, Biostatistician, Surgeon and Behavioral assessor/infarct analysis investigator were blinded to the experimental conditions of the animals. Unblinding was performed only after all statistical analyses were completed. The complete set of data is reported, and no exclusions of any data points were made. Results were examined only at the end of each experiment and there was no interim analysis performed within experiments. Four separate experiments were pre-planned to examine the optimal dose, and time-window for C21 as well as the effect of its interaction with tPA on the short and long-term stroke outcomes (Figure 1).

Figure 1.

Schematic description of experimental design with animal group numbers.

Experiment I (dose finding study): IA (48-h study)

This experiment aimed to identify the optimal dose of C21 (0.01, 0.03 and 0.06 mg/kg) on stroke outcome at 48 h post-eMCAO. Rats (N=12/group) were subjected to eMCAO and treated at 3 h with three different doses of C21 or saline-IV. The four groups were: eMCAO + vehicle (saline); eMCAO + C21 (0.01 mg/kg IV); eMCAO + C21 (0.03 mg/kg IV); eMCAO + C21 (0.06 mg/kg IV). Behavioral outcomes were assessed at 24 and 48 h after eMCAO, followed by animal sacrifice and infarct size analysis. IB (seven days study):

This experiment was conducted to identify the optimal dose of C21 (0.01, 0.03 and 0.06 mg/kg) on stroke outcome at seven days post-eMCAO. Rats (N=12/group) were subjected to eMCAO and treated at 3 h with three different doses of C21 or saline. The first dose was administered IV (Day 1) with successive daily doses given by oral gavage (Days 2–5), calculated based on the oral bioavailability of C21 being 20–30% in rats. ²¹ The four groups were: eMCAO + saline; eMCAO + C21 (0.01 mg/kg IV followed by 0.04 mg/kg oral); eMCAO + C21 (0.03 mg/kg IV followed by 0.12 mg/kg oral); eMCAO + C21 (0.06 mg/kg IV followed by 0.24 mg/kg oral). Behavioral outcomes were assessed at days 1, 3, 5 and 7 after eMCAO and animals sacrificed for tissue collection.

Experiment II (time-window study)

This experiment was carried out to evaluate the time window (3, 6 and 24 h) for C21 on stroke outcome at seven days post-eMCAO. The optimal dose of C21 (0.01 mg/kg), chosen according to our results from experiment I, was initiated at 3, 6 or 24 h post-eMCAO and continued daily for a total of five days. The four groups (N=6–12) were: eMCAO +vehicles (saline) treatment at 3 h; eMCAO + C21 (0.01 mg/kg) treatment at 3 h; eMCAO + C21 (0.01 mg/kg) treatment at 6 h; eMCAO + C21 (0.01 mg/kg) treatment at 24 h. Note: six rats were included in the 3 h group since this group was already shown to be effective in experiment IB. For the time-window study, behavioral outcomes were assessed at days 1, 3, 5 and 7 after eMCAO and animals sacrificed for tissue collection.

Experiment III (delayed tPA-combinatorial study)

This experiment aimed to evaluate the potential interaction of C21, in combination with delayed IV-tPA (10 mg/kg at 4 h), on behavioral outcome, hemorrhage, and infarct size at 48 h post-eMCAO. C21 was given at 3 h (0.01 mg/kg IV) and again (0.04 mg/kg oral gavage) at 24 h. The four groups (N=12) were: eMCAO + vehicle (saline); eMCAO + C21 (0.01 mg/kg, at 3 h); eMCAO + vehicle followed by tPA (10 mg/kg at 4 h); eMCAO + C21 (0.01 mg/kg, at 3 h) followed by tPA (10 mg/kg at 4 h). Behavioral outcomes were assessed at days 1 and 2 after eMCAO and animals sacrificed for hemorrhage and infarct analysis.

Experiment IV (early tPA-combinatorial study)

This experiment was carried out to evaluate the effect of C21 in combination with early IV-tPA (10 mg/kg at 2 h) on long-term functional outcome at 28 days. C21 was given IV (0.01 mg/kg) at 3 h followed by oral gavage (0.04 mg/kg) daily thereafter for five days. The four groups (N=15) were: eMCAO + vehicle (saline); eMCAO + C21 (0.01 mg/kg, at 3 h); eMCAO + vehicle followed by tPA (10 mg/kg at 2 h); eMCAO + C21 (0.01 mg/kg at 3 h) followed by tPA (10 mg/kg at 2 h). Behavioral outcomes were assessed at days 7, 14, 21 and 28 after eMCAO.

Rat model of embolic stroke

Preparation of the clot, induction of embolic stroke and tPA administration were performed as described in our previous report with slight modifications (For details, please see the supplementary file). ¹⁸ Human fibrinogen supplemented clot was used to induce embolic stroke. Briefly, a modified catheter containing a 3.5 ± 0.5 cm long clot was inserted into the right external carotid artery, advanced into the internal and the clot was gently delivered into the brain. Moreover, either tPA (10 mg/kg body weight), or PBS was infused intravenously at 2 or 4 h post eMCAO as 10% bolus, and the balance over a period of 30 min.

Surgical exclusion criteria

A Bederson score (see below) of 3 was consistent at 3 h with eMCAO, and only animals with a score of ≥ 2 at 24 h after embolization (eMCAO) were included in further analysis. In the tPA study, only animals with a score of 3 prior to tPA/saline infusion were included in the analysis of infarct size, hemoglobin, and neurological function. Rats that did not develop sufficient deficits after eMCAO were excluded. Rats that died during the observation period were excluded from all analyses, but differences in the mortality rate between the treatment groups during the observation period were determined. In addition, rats (N = 6) that developed a subarachnoid hemorrhage were excluded from the analysis. Excluded animals (N = 27) in the study were replaced so that treatment groups contained the same number of animals.

Assessment of functional outcome

The experiment was performed between 9:00 a.m. and 4:00 p.m. Animals underwent neurobehavioral testing before eMCAO and at days 1, 3, 7, 14, 21 and 28 after eMCAO.

Motor function tests

The motor function tests (Bederson score, paw grasp, beam walk, grip strength and rotarod) were assessed as previously described by us (for details, please see the supplementary file). ¹⁵

Cognitive function test

The novel object recognition (NOR) test was performed to evaluate non-spatial memory at the end of the experiment (7 or 28 days). ²² This test, based on the spontaneous tendency of rodents to explore and interact with a novel object more than a familiar one, consisted of two trials separated by a retention period and preceded by habituation phase. The habituation phase was conducted on two separate days, before the start of the test, to allow animals to acclimate to their arena, which consisted of an empty standard size box for 15 min. On the designated test day, animals were first subjected to an acquisition/sample trial, where the animal is presented with two identical (sample) objects and allowed to explore for 10 min.^22,23 Following sample object exposure, the animal was returned to its home cage for a 1 h retention period. The second preference trial/test session (5 min), which follows the retention period, was conducted in the same manner as the first trial, except that a novel object replaces one of the familiar objects. The arena and objects were cleaned after each session with 70% ethanol. The time spent in exploring each object during the preference trial /test session was recorded and the discrimination index (DI), which is the difference in exploration time for the objects divided by total time of exploration, was taken as an indication of working memory. ²²

DI = (T_N − T_F)/ (T_N + T_F); T_F is the time spent interacting with the familiar object and T_N is the time spent interacting with the novel object.

Infarct size analysis and hemorrhage

Rats were anesthetized with ketamine/xylazine (IM), transcardially perfused with ice cold PBS and the brains were removed. Each brain was sliced into seven 2 mm-thick coronal sections (A to G) and stained with 2% 2, 3, 5-triphenyltetrazolium chloride (TTC) (Sigma Chemical Co., Missouri, USA). Areas of the infarct, ipsilateral and contralateral hemispheres were measured using ImageJ software. Infarct size was calculated with edema correction using the following formula: 100 × (contralateral – (ipsilateral – infarct))/contralateral. Hemorrhage was quantified using a colorimetric hemoglobin detection assay kit (Quamticrome, BioAssay Systems, Haywood, CA) as described previously. ¹⁸

Monitoring blood pressure

To determine the effect of 0.06 mg/kg C21 dose on blood pressure, a separate set of experimental animals (N = 6) were used. Blood pressure was monitored before eMCAO, after occlusion, and after C21 administration for seven days, using continuous BP telemetry with transmitters (Data Sciences International, St Paul, Minnesota, USA). These were implanted in rats, under 3–5% isoflurane inhalation anesthesia, as reported previously, according to the manufacturer's specifications. ²⁴

Mortality

Mortality was monitored and recorded for the full duration of the study. First, we compared mortality to determine the effect of each main effect and a possible interaction for mortality. Second, we modified the Bederson scoring, and a score of 5 was assigned to those animals that died. We used this data separately to compare the deficit score for all animals, not just those that survived, as “Bederson incorporating mortality” rate.

Data analysis

Sample size estimation

All sample size determinations were made at alpha=0.05 and data are reported as mean ±S.E. In experiment I: Based on an expected infarct size of 40 ± 10% for saline, 0.01 mg/kg and 0.06 mg/kg C21-treated rats and a reduction to 25 ± 10% for rats treated with 0.03 mg/kg C21, sample size of 12/group was determined to provide at least 90% power to detect this difference. For experiment II: C21 given at any time point was expected to provide the same improvement in infarct size over saline (25 ± 10% vs. 40 ± 10%). Sample size of 12/per group was determined to provide at least 90% to detect this difference. Experiment III: Based on an expected excess hemoglobin data of 45 ± 10 ug/mg for untreated animals, 80 ± 10 ug/mg for tPA, and a decrease to 25 ± 10 ug/mg for C21 treatment and 45 ± 10 ug/mg for tPA+C21 treatment, sample size of 13/group was determined to provide 80% power to detect a significant interaction. Experiment IV: Based on an expected 28-day paw grasp data of 2.5 ± 0.8 for saline-treated animals, an effect of C21 and tPA of 1.5 ± 0.8, and the combination treatment providing additive improvement (0.9 ± 0.8), sample size of 12/group was determined to provided 90% power to detect the main effects of C21 and tPA. Since 30% mortality was expected due to long-term follow up, there were 15 animals randomized to each group.

Statistical analysis

All statistical analyses were performed using SAS 9.4 and statistical significance was assessed at an alpha level of 0.05, unless otherwise noted. All tests were two sided. Descriptive statistics for all variables were determined within the blinded group and measurement time. Figures for grip strength, weight, hemoglobin content, and infarct size are presented as mean ± SD. Figures for Bederson, Bederson with mortality, beam walk, and paw grasp are presented as median ± 25th and 75th percentiles. Baseline comparisons between the four dose groups on weight and on grip strength were examined using one-way ANOVA. A Kaplan–Meier survival curve with a log-rank test was used to compare mortality between the groups with time to death being defined as the day of death following surgery for those who died and those who did not die were censored at the time of sacrifice (Supplementary Figures 4 to 7). For continuous outcome measures, repeated measures mixed models were used to examine differences between the four dose groups over time for the weight, grip strength, rotarod time, rotarod RPM, and NOR at 3 min and 5 min assuming an unstructured covariance structure with denominator degrees of freedom calculated using the Kenward–Roger method. For ordinal outcomes (Bederson, Bederson with mortality incorporated, paw grasp, and beam walk), repeated measures mixed models were used on the ranks of the data using the minimum variance quadratic unbiased estimation of covariance parameters method of estimation and assuming an unstructured covariance structure. For ordinal outcome models, the denominator degrees of freedom were calculated using the Kenward–Roger method for factors in the model that did not involve time, and for factors that did involve time the denominator degrees of freedom were infinity. Each model included fixed effects of dose group and time as well as the two-factor interaction between dose group and time/day. For the tPA-combinatorial study, each model included fixed effects of tPA, C21 and day as well as the two-and three-factor interactions between tPA, C21 and day. The F-test for the highest order interaction term with measurement time for each outcome was the statistical test of interest and for ordinal outcomes, the significance of this F-test was determined using the Brunner et al. ²⁵ method. Post hoc pairwise comparisons between groups within time points and between time points within group were examined using a Bonferroni adjustment to the overall alpha level. One-way ANOVA was used to examine differences in infarct size and percent tissue loss between the four dose groups. Post hoc pairwise comparisons between groups for the one-way ANOVA models were performed using a Tukey–Kramer multiple comparison procedure.

Results

Dose response effects of C21 on stroke outcome

Our previously published data indicated that C21 (0.03 mg/kg), administered IV at 3 h after transient suture MCAO, reduces infarct size and improves functional outcome at 24 h and seven days. ¹⁵ In this preclinical study, we aimed to test C21 in a clinically relevant embolic stroke model. First, we conducted a dose–response study using three different doses of C21 (experiments IA and IB). In experiment IA, C21 was given at 3 h after eMCAO as a single treatment. In contrast to what we reported before in a temporary occlusion model, ¹⁵ there was no difference in infarct size or behavioral outcomes at 24 and 48 h after eMCAO (Supplemental Figure 1). Therefore, another experiment (IB) was conducted to examine the dose–response of repeated C21 treatment on stroke outcome at seven days. In this experiment, treatment was started at 3 h, and repeated for five days after eMCAO. There were no statistically significant differences in time to death between saline-treated and C21-treated groups (Kaplan–Meier Supplementary Figure 4 for Exp 1B). Repeated-measures mixed models on the ranks of the data showed significant (p < 0.05) improvements with the low dose (0.01 mg/kg) of C21 on Bederson on days 3 and 5 compared to the 0.03 mg/kg dose and on day 5 compared to saline. For Bederson incorporating mortality, the low dose (0.01 mg/kg) of C21 had lower scores compared to the saline-treated group. For paw grasp scores on days 5 and 7, the low dose showed improvement compared to the 0.03 mg/kg treated group, and on day 7, the low dose showed improvement compared to the 0.06 mg/kg dose. The beam walk showed improvement on day 5 for the low dose compared to the 0.03 mg/kg dose, and the 0.03 mg/ kg dose showed significant decline compared to saline at seven days. Low dose (0.01 mg/kg) C21 showed significant (p < 0.05) improvement in the grip strength on day 5 compared to the saline-treated group (Figure 2). There were no statistically significant interactions between group and time indicating that the difference in means or medians over time had a similar pattern in each group. Furthermore, we examined the dose response effect of C21 on infarct size at seven days after eMCAO. There were no statistically significant differences between treatment groups for percent infarct size (Figure 2).

Figure 2.

Dose–response effect of C21 on functional outcome after eMCAO (a–g). Motor function tests were performed at days 1, 3, 5 and 7 after stroke. Repeated-measures ANOVA showed significant improvement with the low dose (0.01 mg/kg) of C21 on functional outcomes in Bederson (*p < 0.05 vs. 0.03 dose, ^#p < 0.05 vs. saline and 0.03 dose) (a), Bederson incorporating mortality (b), paw grasp (c), beam walk (*p < 0.05 vs. 0.03 dose, ^#p < 0.05 saline vs. 0.03 dose) (d), and grip strength (*p < 0.05 vs. saline) (e) tests after stroke. This improvement was seen at days 3, 5 or 7, and shows a neurorestorative effect. Statisticians Guidance on “Superior” Group – From the plots and the statistical analysis it appears that low dose (0.01 mg/kg) of C21 is superior. There were no statistically significant differences between saline-treated and C21-treated groups for weight (f) and percent infarct size (g). Values are expressed as mean ±SD and median ± 25th and 75th percentiles.

To check the effect of C21 on cognitive function, the NOR test was performed at day 7. Saline-treated animals showed no preference for the novel object, whereas animals treated with C21 seemed to remember the familiar object and spend more time exploring the novel object. All the different groups treated with C21 showed better performance in the NOR test compared to the saline-treated group but the higher dose (0.03 mg/kg) showed the maximum protective effect (Figure 3).

Figure 3.

Dose–response effect of C21 on NOR at seven days after eMCAO. NOR was performed at seven days, prior to sacrifice. The discrimination index is the time spent with the novel object – time spent with the familiar object/ total time (done by video collection). A positive number indicates more time spent with the novel object. As can be seen, the saline-treated animals showed no preference for the novel object, whereas the animals treated with C21, seemed to remember the familiar object and spend more time with the novel object. NOR: novel object recognition. Values are expressed as mean ±SD.

Therapeutic time window for low dose C21 (0.01 mg/kg) on stroke outcome

After identifying the optimal dose of C21 from experiment I (dose response study), we then assessed the therapeutic time window for the effects of C21 after eMCAO. In our recently published study with the mechanical stroke model, we administered C21 at 3 h after tMCAO. ¹⁵ In this experiment (time-window study), we used the most effective dose from experiment I, initiated at either 3, 6 or 24 h after eMCAO, and compared it to saline, with sacrifice at seven days. Behavioral assessments were carried out at repeated intervals (days 1, 3, 5, and 7) after eMCAO. The C21 (0.01 mg/kg) group showed an improvement in Bederson scores compared to saline. This was apparent after a delay of 3, 6 or 24 h post-stroke but was only statistically significant at 6 h at day 7 and 24 h at day 5 (Figure 4) and beam walk scores were lower at 24 h in the C21 group compared to saline. There were no differences between time courses for paw grasp. The 6 h C21 and 24 h C21 showed superiority with regard to grip strength compared to 3 h C21, and weight was superior in the 6 h C21 compared to saline (Figure 4). There were no statistically significant differences between groups in time to death, baseline weight or baseline grip strength between saline-treated and C21-treated groups.

Figure 4.

Therapeutic time window of C21 on functional outcome after eMCAO (a–f). Motor function tests were performed at days 1, 3, 5 and 7 after stroke. C21 (0.01 mg/kg) group exhibited improvement in Bederson score (a) with 3, 6 and 24 h delays after stroke onset (*p < 0.05 saline vs. 6 h, ^#p < 0.05 saline vs. 24 h). Compared with placebo, compound C21 showed a significant improvement in Bederson score (a) as late as 6 h and 24 h after stroke. There were no statistically significant differences between groups in Bederson with mortality (b), paw GRASP (c), beam walk (d), grip strength (e) and weight (f) between saline-treated and C21-treated groups. Statisticians Guidance on “Superior” Group – There is no statistical significance. From the plots, it appears that 6 h and 24 h groups are superior to saline. Values are expressed as mean ± SD and median ± 25th and 75th percentiles.

C21 in combination with early (2 h) or delayed (4 h) IV-tPA

tPA remains the most effective biologic treatment for ischemic stroke. Unfortunately, tPA is associated with a significant risk of hemorrhage and this, along with its narrow therapeutic time window, remains an important impediment to widespread adoption of this therapy.^2,3 Our recent findings indicate that C21 reduces infarct size, hemorrhage and improves functional outcome at 24 h after a transient suture induced MCAO. ¹⁵ So, it was imperative to test C21 treatment for interactions with tPA on acute (48 h) vascular damage (hemorrhage) and long-term functional recovery in an embolic stroke model. In the acute outcomes study when tPA was delayed to 4 h after eMCAO, there was a great variability in infarct size and hemorrhage. This may have resulted in the lack of difference in any of the outcomes in the 48 h study; however, the tPA-treated groups had the highest hemorrhage values (Supplemental Figure 2). Moreover, there were no significant difference in the Bederson, beam walk and grip strength tests at 24 and 48 h after eMCAO (Supplemental Figure 2). There was no statistically significant difference in time to death between tPA and C21 groups.

In the long-term outcome study, when tPA was administered at 2 h after eMCAO, there was not a statistically significant interaction between tPA, C21 and day indicating that the differences in weight over time were similar in the four tPA by C21 treatment groups. There were no statistically significant differences in time to death between the four groups. The interaction between tPA, C21 and day was not statistically significant indicating that the differences in the Bederson, Bederson incorporating mortality, paw grasp, beam walk and grip strength over time were not different in the four tPA/C21 groups (Figure 5). For Bederson in the no tPA group, the no C21 group had lower scores over the C21 group at 7 and 14 days, and, among the C21 group, the tPA group had lower scores compared to the no tPA group. For beam walk in the tPA group, the C21 had lower scores over the no C21 group at 28 days and at 7 days in the no C21 group, the no tPA group had lower scores compared to the tPA group. There were no statistically significant differences between groups for Bederson incorporating mortality or paw grasp. The interaction between tPA, C21 and day was statistically significant indicating that the differences in rotarod time and RPM across days were varied in the four tPA/C21 groups (Figure 5). The groups treated with saline, C21, and tPA + C21 showed decreases in rotorod time from day 0 to day 1 (evidence of an acute stroke) and then steady improvement from day 7 through day 28. However, the group that received tPA only showed a decrease in time from day 0 to day 1 and then little to no improvement from days 7 to 28 (Figure 5). There were no differences between groups within rotarod measurement days.

Figure 5.

C21 in combination with tPA (2 h) for 28 days survival after eMCAO (a–j). Motor function tests were performed at days 7, 14, 21 and 28 after stroke. Combination therapy with tPA did not show any benefit in hemorrhage, weight (a), Bederson (b), Bederson incorporating mortality (c), paw grasp (d), beam walk (e) and grip strength (f) over time. In the rotarod (g, h), tPA group only showed decrease in time from day 0 to day 1 and then little to no improvement from days 7 to 28. There were no differences between groups within rotarod measurement days. There was not a statistically significant interaction between all groups and day for NOR DI at 3 min (i) or 5 min (j). C21 group only exhibited significant improvement in NOR DI at 5 min (j) than group C21 + tPA on day 28 after eMCAO (*p < 0.05). Values are expressed as mean ±SD and median ± 25th and 75th percentiles.

We further examined the effect of C21 in combination with tPA on cognition (NOR at 28 days after eMCAO (Figure 5). There was not a statistically significant interaction between tPA, C21 and day for NOR DI at 3 min or 5 min indicating that the NOR DI from day 0 to day 28 were not different for the four tPA/C21 groups. Only one statistically significant post hoc pair-wise difference was found at day 28, where the group C21 had significantly greater NOR DI at 5 min than C21+tPA (p = 0.002). Despite complete recovery of sensorimotor function, the tPA-treated animals showed a trend of cognitive decline at 28 days. Administration of C21 alone displayed a trend toward the significant improvement in the NOR DI on day 28 compared to tPA-treated groups.

High dose C21 had no effect on BP

In order to ensure that the doses of C21 employed did not affect blood pressure after stroke, we performed telemetry on a separate set of eMCAO animals that received the highest dose of C21 (0.06 mg/kg IV at 3 h followed by daily 0.24 mg/kg oral for five days). Stroke animals showed higher BP from baseline by over 50 mmHg after eMCAO and this was slightly lower after 6 h, but remained elevated for at least 24 h. C21 at 0.06 mg/kg treatment had no effect on BP compared to saline, up to seven days after stroke (Supplemental Figure 3). This is similar to what we previously observed in the suture occlusion model. ¹⁵

Discussion

After numerous clinical trial failures and the variability of preclinical evaluation, the STAIR and RIGOR recommended scientific guidelines for developing effective translational stroke research.^6,7,26 The current preclinical trial was designed to adhere to these guidelines in the evaluation of C21's neuroprotective effects in a rat model of embolic stroke.

We and others have shown the beneficial effects of a single dose of C21 (0.03 mg/kg), administered at 3 h post-occlusion, on stroke outcome.^{14,15,27–29} Nevertheless, from a clinical stand point, it is important to determine the most effective dose and therapeutic window of efficacy for C21 after ischemic stroke. In this study, we first determined the dose response effects of C21 on stroke outcome in eMCAO, a more clinically relevant model of ischemic stroke. After confirming the optimal dose of C21, we assessed the therapeutic time window of C21 treatment on stroke outcome.

We examined the dose response effect using three different doses of C21 (0.01, 0.03 and 0.06 mg/kg) starting at 3 h after eMCAO and continued for five days, using a panel of neurobehavioral tests over seven days. We report here that the lowest dose of C21, 0.01 mg/kg, administered 3 h after eMCAO, resulted in significant (P < 0.05) improvements in a variety of behavioral tests including Bederson, beam walk and grip strength. The lack of a dose–response effect in our data was of concern, however, and we wondered whether the higher doses were associated with an unintended drop in blood pressure. However, continuous BP monitoring both prior to and after eMCAO demonstrated a lack of effect on BP despite administration of the highest dose, 0.06 mg/kg (Supplementary Figure 3).

Others have demonstrated that C21 effectively preserved cognitive function in experimental models of vascular cognitive impairment.^30,31 We were curious as to whether C21 would have similar cognitive benefits in our model. We found that rats treated with C21, regardless of dose, showed preserved cognitive function compared to saline-treated animals. Unlike the sensorimotor data, however, the group treated with 0.03 mg/kg C21 demonstrated superior performance on the NOR test, compared to all other treatment groups.

In the time-window study, we examined the effects of C21 (0.01 mg/kg), when initiated at either 3, 6 or 24 h after eMCAO, on sensorimotor function. We report here that when treatment was delayed for 6 h or 24 h after eMCAO, C21 was still effective in reducing the behavioral impairments. After a treatment delay of 6 h and 24 h, we observed significant improvement in Bederson score. In addition, the 6-h group was superior with regard to grip strength. There were no differences between doses or time-windows for either percent infarct size or tissue loss.

Thrombolytic therapy with tPA is the only proven effective therapy for stroke, but this therapy has limitations due to a significant risk of vascular toxicity (e.g. BBB disruption, edema and hemorrhage). We hoped that combination therapy with tPA and C21 would achieve better outcomes than using either drug alone. In order to do this, we examined the effect of C21 in combination with tPA on acute vascular damage (48 h) and long-term (28 days) functional outcomes in eMCAO with delayed (4 h) and early (2 h) reperfusion (tPA), respectively.

To examine whether the combination of C21 with tPA affects the presence of brain hemorrhage, we measured the frequency of gross hemorrhage and total hemorrhage brain volume when tPA was administered at 4 h after eMCAO. Although our previous work using the mechanical model suggested a vascular protective effect of C21 with a reduction of hemorrhagic transformation at 24 h, ¹⁵ this was not the case in the embolic model. Moreover, the combination of C21 with tPA did not appear to reduce the incidence of hemorrhage. Indeed, consistent with previous studies in embolic and other experimental models,^32–36 we have now found that treatment with tPA at 4 h results in a trend toward higher incidence of gross hemorrhage, as expected, but it did not achieve statistical significance. In addition, the 4 h tPA-treated group with eMCAO showed no difference in neurological deficits and infarct size compared to the saline-treated, which is consistent with previous publications.^32–36 These studies demonstrated that exogenous tPA administration at 1 to 4 h after MCAO in mice exerted neurotoxic and hemorrhagic effects and did not affect infarct size or neurological outcomes. However, this time frame is shorter than the 4.5 or 6 h therapeutic window, for tPA, used in humans. Species differences, faster metabolism, embolic model methodology, and dose of tPA may have all contributed to the quicker onset of tPA associated complications in rodents.^35,37,38

In the long-term outcome study, we administered tPA early (2 h) in order to reduce mortality (compared to when administered at 4 h) and combined it with 0.01 mg/kg C21 for a total of five days. It is an important limitation in the translatability of our findings that we were unable to demonstrate a difference in Bederson, paw grasp, beam walk, rotarod or grip strength over time in the tPA-treated animals. In fact, this was true for all four groups. Only in the cognitive test, on day 28, did C21 treatment show a benefit, but only as monotherapy. The rodent embolic model of MCAO has mostly been employed in short-term outcome studies^18,19,39 and there is very little evidence of a lasting effect of tPA on stroke outcome in these animals. This is likely due to the robust recovery seen in these animals even with no treatment (Figure 5). As far as C21 goes, our previous investigation demonstrated an improved outcome at seven days in a mechanical model of temporary MCAO when a single dose of C21 was administered at 3 h. The embolic model, with its inherent higher variability, appeared to “wash out” any neuroprotective effect of C21 in young male Wistar rats.

In summary, these rigorously conducted, randomized, blinded, controlled preclinical trials were essentially negative in supporting the future development of C21 as a therapeutic modality for acute ischemic stroke. Although small, albeit significant, improvements in sensorimotor function were seen at three, five or seven days, there was a lack of a dose–response, significant time window effect, or impact on long-term outcome. In addition, we were unable to demonstrate any additive effects with tPA in either the short term (48 h) or the long-term studies.

We have not abandoned our interest in C21, however. We believe that the reproducible benefits of C21 on cognition in our study (at 7 and 28 days) are intriguing and we are continuing to pursue this. Also, given the known effects of C21 on inflammation,^14,40 it is possible that our five-day treatment paradigm was insufficient to favorably impact long-term outcomes. One might expect the impact to be even greater at later time points, where inflammation and apoptosis are even more important contributors to the ultimate secondary damage after stroke. Further, longer term studies with multiple days/doses will be needed to determine the promising effects of C21 on cognitive function and we are proceeding in this manner.

Supplemental Material

Supplemental material for Dose–response, therapeutic time-window and tPA-combinatorial efficacy of compound 21: A randomized, blinded preclinical trial in a rat model of thromboembolic stroke

Supplemental material for Dose–response, therapeutic time-window and tPA-combinatorial efficacy of compound 21: A randomized, blinded preclinical trial in a rat model of thromboembolic stroke by Tauheed Ishrat, Abdelrahman Y Fouda, Bindu Pillai, Wael Eldahshan, Heba Ahmed, Jennifer L Waller, Adviye Ergul and Susan C Fagan in Journal of Cerebral Blood Flow & Metabolism