Attenuation of the induction of TLRs 2 and 4 mitigates inflammation and promotes neurological recovery after focal cerebral ischemia

Abstract

The intense inflammatory response triggered in the brain after focal cerebral ischemia is detrimental. Recently we showed that the suppression of toll-like receptors (TLRs) 2 and 4 attenuates infarct size and reduces the expression of pro-inflammatory cytokines in the ischemic brain. In this study, we tested the effect of unrestricted induction of TLRs 2 and 4 on the expression of its downstream signaling molecules and pro-inflammatory cytokines one week after reperfusion. The primary purpose of this study was to investigate the effect of simultaneous knockdown of TLRs 2 and 4 on M1/M2 microglial polarization dynamics and post-stroke neurological deficits and the recovery. Transient focal cerebral ischemia was induced in young adult male Sprague-Dawley rats by the middle cerebral artery occlusion (MCAO) procedure using a monofilament suture. Appropriate cohorts of rats were treated with a nanoparticle formulation of TLR2shRNA and TLR4shRNA (T2sh+T4sh) expressing plasmids (1 mg/kg each of T2sh and T4sh) or scrambled sequence inserted vector (vehicle control) expressing plasmids (2 mg/kg) intravenously via tail vein immediately after reperfusion. Animals from various cohorts were euthanized during reperfusion, and the ischemic brain tissue was isolated and utilized for PCR followed by agarose gel electrophoresis, real time PCR, immunoblot, and immunofluorescence analysis. Appropriate groups were subjected to a battery of standard neurological tests at regular intervals until fourteen days after reperfusion. The increased expression of both TLRs 2 and 4 and their downstream signaling molecules including the pro-inflammatory cytokines was observed even at one-week after reperfusion. T2sh+T4sh treatment immediately after reperfusion attenuated the post-ischemic inflammation, preserved the motor function, and promoted recovery of the sensory and motor functions. We conclude that the post-ischemic induction of TLRs 2 and 4 persists for at least seven days after reperfusion, contributes to the severity of acute inflammation, and impedes neurological recovery. Unlike previous studies in TLRs 2 or 4 knockout models, results of this study in a pharmacologically relevant preclinical rodent stroke model have translational significance.

Introduction

The intense inflammatory response triggered in the ischemic brain contributes to post-stroke pathogenesis [1–3]. The immune system associated inflammatory responses play a pivotal role in the post-stroke brain damage [4]. Innate immunity is primarily initiated by the toll-like receptors (TLRs), and the activation of these TLRs could possibly underlie the intense inflammatory response seen after ischemic stroke. Whereas the post-ischemic inflammation in the late or delayed phase could protect the brain during the acute phase, the robust inflammatory response exacerbates brain damage [5].

Of the thirteen TLRs identified to date, TLRs 1–10 are related to (expressed in) humans [6]. Based on their sequence homology, human TLRs are classified into five subfamilies: TLR1/2/6/10, TLR3, TLR4, TLR5, TLR7/8/9 [7, 8]. Multiple lines of evidence suggest the expression of TLRs on neurons and glial cells, including microglia, astrocytes, and oligodendrocytes [9–18]. While it is evident that microglia and astrocytes initiate the inflammatory cascade by recognizing the ischemia and reperfusion injury-associated molecules through TLRs, infiltration of macrophages and neutrophils into the ischemic brain due to a disrupted blood-brain barrier (BBB) leads to a massive release of inflammatory cytokines, proteolytic enzymes, and other cytotoxic mediators [19–21]. Several preclinical and clinical studies suggest that the post-ischemic induction of TLR2 and TLR4 is detrimental and exacerbates the post-stroke brain damage [16, 22–36].

We believe that novel treatments capable of reducing the severity of inflammation during the acute phase after stroke have the potential to mitigate brain injury and facilitate neurological recovery. Although previous studies in TLR2 and TLR4 knockout models have confirmed the detrimental role of TLR2 and TLR4 in post-stroke pathogenesis, they could not offer any novel therapeutic interventions. Moreover, the translational significance of the outcome of these studies using the gene knockout models is limited and questionable. We have recently shown that attenuation of post-ischemic inflammation, by using the shRNA-mediated gene silencing of TLR2 and TLR4, mitigated the ischemic brain damage in rats [37]. Reduced infarct size and swelling of the ipsilateral hemisphere made this evident. Following treatment, the expression of TLRs 2 and 4 downstream signaling molecules (MyD88, IRAK1, and NFκB p65) and pro-inflammatory cytokines (IL-1β, IL-6, and TNFα) was also greatly reduced in the ischemic brain [37]. Post-ischemic inflammation in the ipsilateral brain is known to contribute to neuronal damage [38–40]. In this study, we examined the effect of TLRs 2 and 4 suppression on the expression of its downstream signaling molecules and pro-inflammatory cytokines in the ischemic brain one-week after reperfusion. Also, we investigated the effect of simultaneous knockdown of TLRs 2 and 4 on M1/M2 microglial polarization dynamics and post-stroke neurological deficits and recovery.

Materials and Methods

ShRNA plasmids and formulation preparation

Four plasmid clones each for TLR2shRNA and TLR4shRNA or their scrambled sequence shRNA (which served as vehicle control) (Qiagen USA) were transfected into E. coli cells (JM109 competent cells), cultured overnight and purified with plasmid mini/maxi kit (Qiagen, USA). The positive clones were confirmed by gene sequencing analysis at the University of Illinois at Urbana-Champaign. The positive plasmid clones were transfected into PC12 cells (ATCC), and PCR confirmed their efficiency to prevent TLR2 and TLR4 expression [37]. Plasmids of the selected clone were purified using phenol-chloroform extraction procedure. Plasmids expressing TLR2shRNA (T2sh), TLR4shRNA (T4sh), or scrambled sequence shRNA (vehicle control) were formulated as nanoparticles by using the in vivo-jetPEI (Polyplus transfection, Illkirch, France) according to the manufacturer’s instructions.

Animals, study design, experimental procedure and treatments

Seventy seven healthy young adult male Sprague-Dawley rats (Envigo Laboratories, USA) were used in this study. Upon arrival, the animals were 7–9 weeks old and weighed 200–220 g. Animals were randomly assigned to four groups, consisting of at least six animals/procedure/group (table 1). Rats were subjected to a suture model transient two-hour right middle cerebral artery occlusion (MCAO) followed by reperfusion, as described recently by our group [41]. At the time of the MCAO procedure, the animals were 8–10 weeks old, and their body weight was 230–260 g. Appropriate cohorts of rats were injected intravenously via tail vein within 30 min after reperfusion with the freshly prepared T2sh (1mg/kg) +T4sh (1mg/kg) or vehicle control (2 mg/kg) formulation. Various neurobehavioral tests, as detailed in the subsequent sections of this manuscript, were performed on rats of appropriate cohorts at regular intervals until fourteen days reperfusion. Baseline scores of these neurobehavioral evaluations were obtained before the MCAO procedure. Untreated and treated ischemia- and reperfusion-induced rats were euthanized at various time points during reperfusion, and their brain tissues were collected and subjected to multiple experimental methods (Fig. 1).

Table 1.

Description of experimental groups:

Group No.	Group Description	Number of animals
		IF	Immunoblot/PCR	Neurological	Total
1	Rats subjected to focal cerebral ischemia procedure without suture insertion (Sham)	6	6		12
2	Rats subjected to 2-h focal cerebral ischemia followed by reperfusion (untreated)	6 (7d AR)	6 (3d AR) 6 (7d AR)	-	18
3	Rats subjected to 2-h focal cerebral ischemia followed by reperfusion, and subsequent treatment with formulated scrambled sequence shRNA (Vehicle control)	-	-	9 (14d AR)	9
4	Rats subjected to 2-h focal cerebral ischemia followed by reperfusion, and subsequent treatment with formulated TLR2shRNA+TLR4shRNA (T2sh+T4sh)	-	6 (3d AR)	10 (14d AR)	16
		Number of animals excluded from the study			22

IF-immunofluorescence; PCR-polymerase chain reaction; 3d AR-euthanized three days after reperfusion; 7d AR-euthanized seven days after reperfusion; 14d AR-euthanized fourteen days after reperfusion.

Fig. 1.

Schematic representation of the experimental design and methods executed. Abbreviations: AR, after reperfusion; mNSS, modified neurological severity score; PCR, polymerase chain reaction. The figure was created with the support of https://biorender.com under the paid subscription.

RT-PCR analysis and agarose gel electrophoresis

RT-PCR analysis was performed using the cDNAs synthesized from various brain tissue samples and the primers as listed in table 2 by using GoTaq^® Green Master Mix (Promega, USA) according to the manufacture’s protocol. RT-PCR was performed in C1000 Touch Thermocycler (Bio-Rad Laboratories, USA) using the following PCR cycle: [95°C for 5min, (95 °C for 30 sec, 56–60 °C for 30 sec, 72 °C for 30 sec) × 35 cycles, and 72 °C for 5 min]. RT-PCR products were resolved on a 2% agarose gel, stained with ethidium bromide, and visualized under UV light. The mRNA expression of target genes as compared to the expression of housekeeping gene β-actin was quantified using Image J analysis (NIH) software.

Table 2.

Rat specific primers for PCR:

Gene	NCBI Reference Sequence	Primer Sequence
		Forward (5’ - 3’)	Reverse (5’ - 3’)
TLR2	NM_198769	ggaagcaggtgacaaccatt	cgcctaagagcaggatcaac
TLR4	NM_019178	gtgggtcaaggaccagaaaa	ggctaccacaagcacactga
TIRAP	XM_017596001	ccaagaagcctcgagacaag	atgactccgaggtgaactgc
MyD88	NM_198130	atcccactcgcagtttgtt	gatgcggtccttcagttcata
IRAK1	NM_001127555	cctccctggaagctagaggt	agaggccaggaacactctca
NFκB p65	NM_199267	catgcgtttccgttacaagtg	cccgtgtagccattgatctt
IL-1β	NM_031512	gctttcgacagtgaggagaat	cgagatgctgctgtgagatt
IL-6	NM_012589	ccaacttccaatgctctcctaa	ttgccgagtagacctcatagt
TNFα	NM_012675	cagccgatttgccatttcatac	aggtacatgggctcatacca
CD16	NM_207603	acttctgcagagggatcattg	cagcaggcagaaagtgatttg
TGFβ	NM_021578	ttcagctccacagagaagaac	gtgtccaggctccaaatgta
βActin	NM_031144	gtcgtaccactggcattgtg	ctctcagctgtggtggtgaa
18SrRNA	NR_046237	acgtctgccctatcaactttc	ttggatgtggtagccgtttc

Real-time PCR analysis

Real-time PCR analysis was performed using the SYBR Green method. Briefly, the reaction set-up for each cDNA sample was assembled using the FastStart SYBR Green Master (Roche, Indianapolis, Indiana, USA) as per the manufacturer’s instructions. The forward and reverse primer sequences of target genes are listed in table 2. Samples were subjected to the following PCR cycle: [95°C for 5min, (95 °C for 30 sec, 56–60 °C for 30 sec, 72 °C for 30 sec) × 35 cycles, and 72 °C for 5 min] in iCycler IQ (Multi-Color Real-Time PCR Detection System; Bio-Rad Laboratories, Hercules, California, USA). Data were collected and recorded using the iCycler IQ software (Bio-Rad Laboratories) and expressed as a function of the threshold cycle (Ct), which represents the number of cycles at which the fluorescent intensity of the SYBR Green dye is significantly above that of the background fluorescence. β-actin and 18SrRNA served as housekeeping genes. The average Ct values were normalized with average Ct values of βactin. After normalization of the Ct values, fold changes were calculated by using the formula 2^-(ΔCt of Test)/2^-(ΔCt of controls).

Immunofluorescence analysis

Appropriate cohorts of rats at seven days reperfusion were placed under deep anesthesia with pentobarbital and perfused through the left ventricle with 70–100 mL of PBS, followed by 100–150 mL of 10% buffered formalin (Fisher Scientific, NJ). The brains were removed, fixed in 10% buffered formalin, and embedded in paraffin. Serial coronal brain sections were cut at a thickness of 6–8 μm with a microtome. Immunofluorescence analysis was used to identify the changes in the expression of TLR2 and TLR4 proteins in sham and untreated ischemia- and reperfusion-induced animals euthanized on the seventh day of reperfusion. Paraffin-embedded brain sections were de-paraffinized, subjected to antigen retrieval, permeabilized, processed with anti-TLR2 and anti-TLR4 primary antibodies followed by fluorescent-labeled secondary antibodies, counterstained with DAPI, and cover slipped. The images were captured using a confocal microscope (Olympus Fluoview).

Immunoblot analysis

Protein samples prepared from the ipsilateral brain hemispheres of rats from appropriate cohorts were subjected to immunoblot analysis using antibodies for MyD88, IRAK1, NF-κB p65, IL-1β, IL-6, TNFα, and GAPDH (Santa Cruz Biotechnology, USA) followed by HRP-conjugated secondary antibodies. Immunoreactive bands were visualized using chemiluminescence ECL Western blotting detection reagents (Bio-Rad Laboratories, USA). Band intensities were quantitated using Image J analysis software (NIH) and normalized using GAPDH as a housekeeping protein.

Modified neurological severity score (mNSS) assessment

The mNSS is a widely accepted, standardized method of assessing the severity of post-stroke injury and recovery [42] and is a composite of motor, sensory, reflex and balance tests. A cumulative score from all the tests determines the severity of post-stroke injury in each animal. Modified NSS score of 1–6 indicate mild injury; score of 7–12 indicate moderate injury; and score of 13–18 indicate severe injury. The mNSS assessment was performed on rats of both cohorts before ischemia (baseline) and at regular intervals (1d, 3d, 5d, 7d, and 14d) post-ischemia, until fourteen days reperfusion.

Modified adhesive removal (sticky tape) test

This test provides an assessment of post-stroke somatosensory dysfunction [43] and can reliably quantify the degree of focal sensory impairment in animals without requiring any prior training. In this test, a strip of adhesive tape was wrapped around the forepaw, and the animal’s interaction time with the tape (i.e. licking/biting or shaking of the taped limb) during a 30-sec period was recorded for both the affected limb and the unaffected limb. The sticky tape test was performed on both cohorts of rats before ischemia (baseline) and at regular intervals (1d, 3d, 5d, 7d and 14d) post-ischemia, until fourteen days reperfusion. In each animal, the average of three trials for both the affected limb and the unaffected limb was calculated at each time point and used for data analysis. In addition, a sticky-tape ratio (affected limb/unaffected limb) was calculated for all the animals. The interaction time and sticky-tape ratio provide complimentary, quantitative assessment of post-stroke somatosensory function.

Beam walk test

This test, often referred to as the foot fault test, was used to assess deficits in motor coordination and integration, especially those involving the hindlimbs of rats [44]. The beam walk apparatus consisted of a square, rectangular beam (2 cm × 2 cm cross section and 152 cm long, with a 110 cm walking distance), which was supported 30 cm above the tabletop surface by two stands positioned at opposite ends of the beam. A thick layer of bubble wrap was placed under the beam to cushion the fall, in the event an animal slipped off the beam. Prior to ischemia induction, rats from both cohorts were trained for two or three days to traverse the beam, and by the end of the training period, all rats had learned the task. The beam walk test was performed on rats of both cohorts before ischemia (baseline) and at regular intervals (1d, 3d, 5d, 7d, and 14d) post-ischemia, until fourteen days reperfusion. Beam walk performance of rats was rated as follows: 0 - the rat was not able to stay on the beam; 1 - the rat was able to stay on the beam, but did not move; 2 - the rat tried to traverse the beam, but fell; 3 - the rat traversed the beam with more than 50% foot slips of the affected forelimb and/or hindlimb; 4 - the rat traversed the beam with more than one-foot slip, but less than 50% foot slips of the affected forelimb and/or hindlimb; 5 - the rat traversed the beam with only one-foot slip of the affected forelimb and/or hindlimb; 6 - the rat traversed the beam without any foot slips of either the affected forelimb or hindlimb. For each testing session, the mean of three trials was used to evaluate the beam walk performance of each rat.

Accelerating Rotarod performance test

This test was used to evaluate the motor coordination of stroke-induced rats that had received different treatments. Animals were trained to locomote on a rotating Rotarod (Rotamex, Columbus instruments) (initial speed = 10 rpm; acceleration rate = 0.3 rpm/sec; maximum speed = 80 rpm) for two or three days prior to ischemia induction. By the end of the training period, all rats had learned the task to a satisfactory level (latency to fall ≥ 50 sec). As with the other tests, the Rotarod was performed before ischemia (baseline) and at regular intervals (1d, 3d, 5d, 7d, and 14d) post-ischemia, until fourteen days reperfusion. During testing, rats were challenge to remain on the accelerating Rotarod for a maximum period of 300 sec, employing the same conditions used during training. The latency to fall from the Rotarod apparatus was recorded for each rat with a minimum of three trials, which are then averaged. Animals were allowed to rest between trials for at least 15 min, and bubble wrap was used to cushion their falls during the sessions. The Rotarod latency of each rat at the different time points was calculated and expressed as the percentage of baseline value, which was considered 100%.

Data collection and exclusion criteria

Neurobehavioral evaluation tests were performed by trained research personnel who were blind to treatments. Animals that showed signs of hemorrhage at the origin of MCA following terminal euthanasia were excluded from the study. Furthermore, animals that did not show or showed only mild neurobehavioral changes after reperfusion (mNSS < 8 within 3h post-reperfusion) were excluded from the study.

Statistical Analysis

Statistical analysis of the data were carried out using Prism v.6.04 for Windows (Graph Pad Software, San Diego, CA). Outliers in the data were identified by the Grubb’s test. Quantitative data of all the experiments were tested for normality, followed by F test for equality of variances. For the real time PCR and immunoblot data, differences between groups were analyzed by unpaired (two-tailed) t-test, with or without Welch’s correction. Body weight, bodyweight gain, and neurobehavioral assessments (mNSS, sticky tape, beam walk, and Rotarod tests) data were analyzed by two-way repeated-measures ANOVA (with treatment as the between-subject factor and time as the within-subject factor). In some cases (sticky tape test), a repeated measures ANOVA for both factors was used. Following a significant ANOVA test result, post-hoc comparisons were made using Sidak’s test (for main treatment effects) or Tukey’s multiple comparison test (for main time effects). All data are expressed as means ± SEM. Differences between groups were considered significant at p<0.05.

Results

TLR2shRNA and TLR4shRNA are efficient and attenuate infarct volume

We have recently shown that TLR2shRNA and TLR4shRNA plasmids transfected into PC12 cells reduced the expression of the target genes [37]. Furthermore, TLR2shRNA and TLR4shRNA plasmids formulated as nanoparticles and delivered intravenously via tail vein in rats reduced the expression of both the TLRs 2 and 4 in the ischemic brain [37]. These experiments clearly demonstrate the in vitro and in vivo efficacy (and by inference brain penetrability) of the plasmids used in this study. The administration of either TLR2shRNA or TLR4shRNA plasmids to stroke-induced rats significantly reduced the percent infarct volume, compared to rats that did not receive any treatment [37]. Furthermore, rats that received both TLR2shRNA and TLR4shRNA plasmids showed a greater reduction of infarct volume than those that received either treatment alone.

TLRs 2 and 4 mediated post-ischemic inflammation during reperfusion

Recently, we showed post-ischemic induction of TLRs 2 and 4 in the rat brain at five days after reperfusion [37]. In this study, we observed their continued elevation in the ischemic rat brain at seven days after reperfusion. Real time PCR analysis data revealed that TLR2 and TLR4 mRNA expression was ~13 fold and ~11 fold higher at seven days reperfusion as compared to their respective mRNA expressions in the sham cohort (Fig. 2A). The increase in TLRs 2 and 4 mRNA expression observed in this study at seven days reperfusion was much higher than their expression we had reported earlier at five days reperfusion [37]. Furthermore, the mRNA expression of toll-interleukin 1 receptor domain containing adaptor protein (TIRAP), which is specific to TLR downstream signaling, was increased to ~3 fold at seven days after reperfusion as compared to its expression in the sham cohort (Fig. 2B). Consistent with the PCR data, immunofluorescence analysis demonstrated strong protein expression of both TLRs 2 and 4 in the ischemic brain at seven days after reperfusion as compared to the sham cohort (Fig. 2C). In contrast to the strong and clear labeling of TLRs observed in the ischemic animals, we detected little or no specific labeling in the sham animals. The sparse and negligible staining observed in the sham animals precluded us from carrying out any meaningful quantitative analysis of the immunofluorescence staining. Nonetheless, the clear distinction between the barely visible staining in the sham group versus the obvious staining in the ischemic group strongly indicates that TLRs were upregulated in the stroke-induced animals. The mRNA expressions of TLR2 and TLR4 common downstream signaling molecules MyD88, IRAK1, and NFκB p65 (p65) were all significantly increased (p<0.0001) at seven days reperfusion (Fig. 3A). Similarly, the protein expressions of these signaling molecules were also significantly increased (p<0.0001) as compared to their expression in the sham cohort (Fig. 3B). Of significance, MyD88-dependent TLRs 2 and 4 downstream signaling leads to the release of pro-inflammatory mediators such as IL-1β, IL-6, and TNFα. As expected from the preceding MyD88 results, the mRNA expressions of these pro-inflammatory mediators in the ischemic brain at seven days reperfusion were all significantly (p<0.0001) increased (Fig. 3C). Likewise, the protein expressions of IL-1β, IL-6, and TNFα were also significantly increased (p<0.001 for IL-1β and p<0.0001 for IL-6 and TNFα) as compared to their protein expressions in the sham cohort (Fig. 3D).

Fig. 2.

Post-ischemic induction of TLRs 2 and 4 in the ischemic brain. (A) Quantitative real-time PCR analysis depicting the mRNA expression of TLRs 2 and 4 as fold change over sham in the ischemic brain of male rats subjected to 2-h focal cerebral ischemia followed by reperfusion for seven days. Beta actin served as the house-keeping gene. Histograms and error bars indicate the mean and the SEM, respectively; n=6. (B) Quantitative real-time PCR analysis depicting the mRNA expression of TIRAP as fold change over sham in the ischemic brain of male rats subjected to 2-h focal cerebral ischemia followed by reperfusion. 18SrRNA served as the house-keeping gene. Histograms and error bars indicate the mean and the SEM, respectively; n=6. (C) Representative immunofluorescence images depicting the protein expression of TLRs 2 and 4 (green) in the ipsilateral hemisphere of rats subjected to 2-h focal cerebral ischemia followed by reperfusion for seven days. Nuclei were stained with DAPI (blue). Scale bar = 100 μm.

Fig. 3.

Increased expression of TLRs 2 and 4 downstream signaling molecules in the ischemic brain of rats subjected to 2-h focal cerebral ischemia and reperfusion for seven days. (A) Bar graph represents the densitometry analysis of TLRs 2 and 4 downstream signaling molecules (MyD88, IRAK1, and NFκB p65) mRNA expression obtained from the agarose gel electrophoresis of RT-PCR products, normalized to beta-actin; n=6. (B) Representative immunoblots depicting the protein expression of MyD88, IRAK1, and NFκB p65. GAPDH served as a loading control. Bar graph represents the densitometry analysis of MyD88, IRAK1, and NFκB p65 protein bands, normalized to GAPDH; n=6. Statistical tests: Unpaired t test with Welch correction. ****p<0.0001 vs sham. (C) Bar graph represents the densitometry analysis of pro-inflammatory molecules (IL-1β, IL-6, and TNFα) mRNA expression obtained from the agarose gel electrophoresis of RT-PCR products, normalized to beta-actin; n=6. (D) Representative immunoblots depicting the protein expression of IL-1β, IL-6, and TNFα. GAPDH served as a loading control. Bar graph represents the densitometry analysis of IL-1β, IL-6, and TNFα protein bands, normalized to GAPDH; n=6. Statistical tests: Unpaired t test with Welch correction. ***p<0.001; ****p<0.0001 vs sham.

Suppression of both the TLRs 2 and 4 attenuates post-ischemic inflammation

We recently reported that the simultaneous knockdown of TLR2 and TLR4 by shRNA-mediated gene silencing reduced the infarct size and expression of pro-inflammatory mediators, including IL-1β, IL-6, and TNFα in the ischemic brain [37]. In this study, real time PCR analysis data revealed that the expression of M1 marker CD16 increased to approximately 9 fold at three days reperfusion (Fig. 4). Administration of T2sh+T4sh immediately after reperfusion significantly (p<0.01) decreased the induction of CD16. Previously, a gradual increase in the expression of M2 markers, including IL-10 and TGFβ, was reported in the ischemic mouse brain until five days reperfusion, followed by a decrease [45]. In our study, we observed an increase in the expression of TGFβ to approximately 3 fold at three days reperfusion in rats (Fig. 4). Importantly, T2sh+T4sh treatment significantly (p<0.01) decreased the expression of TGFβ in the ischemic rat brain.

Fig. 4.

Simultaneous knockdown of TLRs 2 and 4 reduces the post-ischemic expression of M1 and M2 polarization markers. Quantitative real-time PCR analysis depicting the mRNA expression of M1 and M2 microglia markers (CD16 and TGFβ) as fold change over sham in the ischemic brain of male rats, subjected to 2-h focal cerebral ischemia followed by reperfusion for three days. The appropriate cohorts were treated with the nanoparticle formulation containing TLR2shRNA and TLR4shRNA plasmids (T2sh+T4sh). 18SrRNA served as the house-keeping gene. Histograms and error bars indicate the mean and the SEM, respectively; n=6. Statistical tests: Unpaired t test. **p<0.01 vs untreated.

Effect of TLRs 2 and 4 suppression on body weight changes and neurological scores

We observed a significant decrease (p<0.0001) in post-stroke body weight at one-day reperfusion as compared to the body weight of animals before the MCAO procedure in both the vehicle control and T2sh+T4sh cohorts (Fig. 5A). We calculated the bodyweight gain at 3, 5, 7, and 14 days of reperfusion as compared to the body weight at one-day reperfusion. We observed that rats from both the cohorts continued to lose body weight after one-day reperfusion but reached the day 1 body weight approximately at seven days reperfusion. Although the bodyweight gain was significantly higher (p<0.0001) in both the vehicle control and T2sh+T4sh treated cohorts at fourteen-day reperfusion as compared to the respective weight gain at three days after reperfusion, treatment-related differences were not seen at any of the reperfusion time points (Fig. 5B).

Fig. 5.

Post-stroke changes in body weight and modified neurological severity score (mNSS). (A) Bar graph represents the post-stroke body weight change in rats at one-day reperfusion subsequent to a 2-h focal cerebral ischemia. Error bars indicate SEM; n=10 (T2sh+T4sh treated group) and n=9 (vehicle control). Statistical tests: Two-way repeated-measures ANOVA followed by Sidak’s multiple comparisons test. ⁺⁺⁺⁺p<0.0001 vs before ischemia. (B) Bar graph represents the percent bodyweight gain during reperfusion as compared to the body weight at one-day reperfusion. Error bars indicate SEM; n=10 (T2sh+T4sh treated group) and n=9 (vehicle control). Statistical tests: Two-way repeated-measures ANOVA followed by Sidak’s or Tukey’s multiple comparisons test. ⁺p<0.05; ⁺⁺⁺⁺p<0.0001 vs 3 day reperfusion. (C) Bar graph represents the mNSS of rats subjected to 2-h focal cerebral ischemia followed by reperfusion. Error bars indicate SEM; n=10 (T2sh+T4sh treated group) and n=9 (vehicle control). Statistical tests: Two-way repeated-measures ANOVA followed by Sidak’s or Tukey’s multiple comparisons test. ⁺⁺⁺p<0.001; ⁺⁺⁺⁺p<0.0001 vs one-day reperfusion. *p<0.05 (T2sh+T4sh vs vehicle control).

The neurobehavioral symptoms as evaluated by the mNSS assessment revealed the scores of 9.44 ± 0.59 and 8.60 ± 0.83 in vehicle control and T2sh+T4sh cohorts, respectively, at one-day after reperfusion (Fig. 5C). These scores indicate that the magnitude of post-stroke brain injury induced in this study is moderate and comparable between both cohorts of animals. The scores gradually declined over time in both groups until fourteen days after reperfusion. The decrease was significant at days 7 (p<0.001) and 14 (p<0.0001) in the vehicle control group and at days 5 (p<0.001), 7 (p<0.0001), and 14 (p<0.0001) in the T2sh+T4sh group. At the end of the study (day 14 after reperfusion), the scores were reduced to 5.56 ± 0.75 and 2.90 ± 0.84 in the vehicle control and the T2sh+T4sh cohorts, respectively. The score of the T2sh+T4sh group was significantly lower (p<0.05) than the score of the vehicle control group, indicating that they improved more.

Suppression of TLRs 2 and 4 facilitates the recovery of somatosensory function

As expected, in the vehicle control group, the differences in interaction time between the affected and unaffected forelimbs was significant (p<0.0001) at all the tested reperfusion time points (Fig. 6A). In contrast, in the T2sh+T4sh group, the differences in interaction time between the two limbs was significant at reperfusion day 1 (p<0.0001), 3 and 5 (p<0.001), but not at day 7 or 14. The lack of an appreciable significant difference between the affected limb versus the unaffected limb at the later time points indicates that the treatment had facilitated the recovery of post-stroke somatosensory function. This is further supported by a significant time-dependent increase in the interaction time of the affected limb in the T2sh+T4sh group at reperfusion days 5 (p<0.05), 7 (p<0.0001), and 14 (p<0.0001) as compared to reperfusion day 1. No such effect was observed for the unaffected limb in the T2sh+T4sh group, or for either limb in the vehicle control group.

Fig. 6.

Suppression of TLRs 2 and 4 facilitates the post-stroke recovery of somatosensory function. Bar graphs represent the quantitative data of the sticky-tape interaction time (A) and the sticky-tape ratio (B) obtained from rats subjected to 2-h focal cerebral ischemia followed by reperfusion. Error bars indicate SEM; n=10 (T2sh+T4sh treated group) and n=9 (vehicle control). Statistical tests: Two-way repeated-measures ANOVA followed by Sidak’s or Tukey’s multiple comparisons test. ⁺p<0.05; ⁺⁺⁺⁺p<0.0001 vs one-day reperfusion. *p<0.05 (T2sh+T4sh vs vehicle control). ***p<0.001; ****p<0.0001 (affected limb vs unaffected limb).

Prior to the induction of ischemia, the sticky-tape ratio of rats from the appropriate cohorts was approximately 1, as would be expected in normal animals, and indicates little or no difference between the two limbs in terms of somatosensory function. However, following the MCAO procedure, the sticky-tape ratio of rats dropped below 1, and even approached 0 (zero) in some cases, depending on the extent of stroke-induced somatosensory dysfunction. The sticky-tape ratios of 0.23 ± 0.11 and 0.31 ± 0.13 at one-day after reperfusion in the vehicle control and T2sh+T4sh cohorts, respectively, indicates that post-stroke somatosensory function was compromised in both groups at this early time point (Fig. 6B). In the vehicle control group, the sticky-tape ratio was increased from 0.23 ± 0.11 at one-day after reperfusion to a maximum value of 0.55 ± 0.11 at fourteen-day of reperfusion. However, the increase was not significant at any of the reperfusion time points tested in the study. In the T2sh+T4sh group, the sticky-tape ratio was increased from 0.31 ± 0.13 at one-day after reperfusion to a maximum value of 0.95 ± 0.05 at fourteen-day of reperfusion. The increase was significant at reperfusion days 7 (p<0.001) and 14 (p<0.0001). Furthermore, at fourteen-day of reperfusion, the sticky-tape ratio of the T2sh+T4sh group was significantly higher (p<0.05) than the ratio of the vehicle control group. Overall, these data suggest that while T2sh+T4sh treatment could not prevent the loss of post-stroke somatosensory function, it did promote the recovery of somatosensory function.

Suppression of TLRs 2 and 4 facilitates the recovery of motor function

The beam walk scores of the rats provide a quantitative measure of the animal’s coordination and integration of motor movement (with a score of 6 being normal). The beam walk scores of 1.11 ± 0.28 and 2.10 ± 0.71 at one-day after reperfusion in the vehicle control and T2sh+T4sh cohorts, respectively, indicate a severe impairment in post-stroke motor coordination and integration in both cohorts of animals (Fig. 7A). In the vehicle control group, the beam walk score was increased from 1.11 ± 0.28 at one-day after reperfusion to a maximum score of 3.11 ± 0.67 at fourteen-day of reperfusion. Compared to day 1, the increase in beam walk scores was significant at reperfusion days 7 (p<0.05) and 14 (p<0.01). In the T2sh+T4sh group, the beam walk score was increased from 2.10 ± 0.71 at one-day after reperfusion to a maximum score of 5.40 ± 0.36 at fourteen-day of reperfusion. The increase in beam walk scores was modestly significant at reperfusion day 5 (p<0.05) and highly significant (more so than the vehicle control group) at reperfusion days 7 (p<0.0001) and 14 (p<0.0001). Moreover, at fourteen-day reperfusion, the beam walk score of the T2sh+T4sh group was significantly higher (p<0.05) than the beam walk score of the vehicle control group.

Fig. 7.

Simultaneous knockdown of TLRs 2 and 4 facilitates the post-stroke recovery of motor function. Line graphs represent the quantitative data of beam walk scores (A) and Rotarod latencies, expressed as a percent baseline (B) obtained from rats subjected to 2-h focal cerebral ischemia followed by reperfusion. The baseline Rotarod latencies were 96.04 ± 8.81 for the vehicle control group and 98.13 ± 8.20 for the T2sh+T4sh treated group. Error bars indicate SEM; n=10 (T2sh+T4sh treated group) and n=9 (vehicle control). Statistical tests: Two-way repeated-measures ANOVA followed by Sidak’s or Tukey’s multiple comparisons test. ⁺p<0.05; ⁺⁺p<0.01; ⁺⁺⁺⁺p<0.0001 vs one-day reperfusion; *p<0.05 (T2sh+T4sh vs vehicle control).

The accelerating Rotarod performance of the rats provides a quantitative measure of motor function. The Rotarod latencies of 21.7% ± 6.95% and 54.8% ± 11.67% of baseline values at one-day after reperfusion in the vehicle control and T2sh+T4sh cohorts, respectively, indicate an impairment in post-stroke motor function in both groups (Fig. 7B). However, the Rotarod latency was significantly higher (p<0.05) in the T2sh+T4sh group than of the vehicle control group. These results indicate that the T2sh+T4sh treatment may have attenuated the loss of post-stroke motor function. In the vehicle control group, the Rotarod latency was increased from 21.7% ± 6.95% of baseline at one-day after reperfusion to 50.4% ± 8.46% of baseline at fourteen-day of reperfusion. Compared to day 1, the increase in Rotarod latency was significant at reperfusion days 5 (p<0.05) and 14 (p<0.05). In the T2sh+T4sh group, the Rotarod latency was increased from 54.8% ± 11.67% of baseline at one-day after reperfusion to 80.1% ± 7.44% of baseline at fourteen-day of reperfusion. This increase in Rotarod latency was significant at reperfusion day 14 (p<0.05) compared to day 1. Furthermore, at fourteen-day of reperfusion, the Rotarod latency of the T2sh+T4sh group was significantly higher (p<0.05) than the Rotarod latency of the vehicle control group. These results indicate that the T2sh+T4sh treatment may have both attenuated the loss of and promoted the recovery of post-stroke motor function.

Discussion

In this study, we demonstrated the continued elevation of TLRs 2 and 4 and their associated downstream signaling molecules, including the pro-inflammatory cytokines in the ischemic brain at seven days of reperfusion after a two-hour transient focal cerebral ischemia. Also, we showed that attenuation of the post-ischemic induction of both TLR2 and TLR4 by shRNA-mediated gene silencing mitigates inflammation, preserves somatosensory and motor functions, and promotes neurological recovery.

In ischemic stroke patients, the expression of TLR2 and TLR4 on monocytes at one-day, three-day, and seven days after admission was significantly higher in patients with poor outcome [30]. Paralleling these findings, we have recently found elevated expression of TLR2 and TLR4 up to five days of reperfusion in the ischemic brains of rats subjected to transient focal cerebral ischemia [37]. In the current study, we further show the continued and persistent elevation of both TLRs 2 and 4 in the ischemic rat brains even one-week after reperfusion. It is well established that MyD88-dependent TLR2 and TLR4 downstream signaling leads to the elevation of pro-inflammatory mediators such as IL-1β, IL-6, and TNFα [5, 46]. In ischemic stroke patients, the observed correlation of the expression of TLRs 2 and 4 with serum IL-1β, IL-6, and TNFα levels after ischemic stroke suggests that these TLRs may be involved in the activation of inflammatory responses during the acute phase after ischemic stroke [30]. Increased expression of TLR2 and TLR4 common downstream signaling molecules (MyD88, IRAK1, and NFκB p65) and pro-inflammatory mediators (IL-1β, IL-6, and TNFα) in the ischemic rat brains up to seven days after reperfusion, as demonstrated herein and in our previous work, supports the concept that TLRs 2 and 4 play a major role in mediating post-ischemic inflammatory responses [37]. The post-ischemic induction of IL-1β, IL-6, and TNFα observed in our experiments is in agreement with earlier reports [47–52] and formed part of the basis for specifically targeting TLRs 2 and 4 by shRNA-mediated gene silencing approach, with the specific aim of reducing the severity of post-ischemic inflammation during the acute phase of ischemic stroke. Suppression of both TLRs 2 and 4 immediately following reperfusion significantly reduced the expression of downstream signaling molecules, including MyD88, IRAK1, and NFκB p65 and the final pathway mediators of inflammation, including IL-1β, IL-6, and TNFα in ischemic rat brains [37] (Fig. 8).

Fig. 8.

Schematic representation of the effects of ischemic stroke in rats subjected to 2-h focal cerebral ischemia followed by reperfusion, and subsequent treatment with T2sh+T4sh nanoparticle formulation. Ischemic stroke leads to the induction of TLRs 2 and 4 and their downstream signaling molecules (MyD88, IRAK1, and NFκB p65) and the release of pro-inflammatory mediators (IL-1β, IL-6, and TNFα) in the ischemic brain. T2sh+T4sh treatment by attenuating the induction of TLRs 2 and 4 and their downstream signaling, reduces infarct volume and expression of M1 markers, preserves neurological function, and promotes recovery [37].

Microglia/macrophages that mediate multiple facets of neuroinflammation represent the first line of innate immune reaction to ischemic brain injury. The activation of microglia/macrophages helps the injured brain by removing cell debris and restoring tissue integrity [53, 54]. However, growing evidence suggests that drastic activation of microglia/macrophages exacerbates infarct volume and worsens neurological outcomes. The differentiation of microglia/macrophages into a “classically activated” sick M1 phenotype leads to the release of destructive pro-inflammatory mediators such as IL-1β, IL-6 and TNFα [55–57], whereas differentiation into an “alternatively activated” healthy M2 phenotype possesses neuroprotective and repair properties [58–60]. At early stages of ischemic stroke, local microglia and newly recruited macrophages assume the M2 phenotype but gradually transform into the M1 phenotype in peri-infarct regions [45]. The expression of several M1 markers, including CD11b, CD16, CD32, and iNOS in the ischemic mouse brain was gradually increased over time and remained elevated for at least fourteen days after reperfusion. In contrast, the expression of M2 markers, including CD206, IL-10, TGFβ, Arg1, CCL22, and Ym1/2 was induced one to three days post-ischemia and peaked at three to five days of reperfusion followed by a decrease. In agreement with the reported studies, we observed a 9 fold increase in CD16 and a 3 fold increase in TGFβ expression in the ischemic rat brain at three days of reperfusion. The M1 and M2 marker genes mentioned earlier are expressed in microglia/macrophages, as well as in other brain cells or infiltrating immune cells. Therefore, the real-time PCR results obtained in this study reflect the overall changes of M1 and M2 signature genes in the ischemic brain tissue of mixed cell types.

TLR4 is important for microglial polarization. TIRAP functions as a sorting adaptor that links MyD88 to activated TLR4 on cell membrane [61, 62]. In this study, although there is no change in the TIRAP mRNA expression at three days after reperfusion in the untreated group compared to the sham cohort, we observed a 3 fold increase in its mRNA expression at seven days after reperfusion. TIRAP promotes microglial M1 polarization via TLR4-mediated TAK1/IKK/NF-κB, MAPKs and Akt signaling pathways [63]. The increase in TIRAP expression observed at seven days after reperfusion could facilitate the M1 polarization of microglia beyond three days after ischemia. In agreement with this, we found a much larger 23-fold increase in CD16 mRNA expression over sham at seven days, compared to a 9-fold increase at three days after reperfusion (Supplementary Fig. 1). It has been shown that prevention of TLR4‐mediated signaling suppresses microglial activation and interferes with the release of destructive cytokines [64]. Therefore, in the current study, we expected that the suppression of TLRs 2 and 4 would reduce the expression of M1 signature genes while increasing the M2 markers. As predicted, the simultaneous knockdown of TLR2 and TLR4 attenuated the expression of the M1 marker CD16 in the ischemic brain. Although we anticipated an increase in the expression of M2 marker, suppression of TLRs 2 and 4 significantly reduced the expression of TGFβ, contrary to our expectations. Evidence suggests that the pro-inflammatory cytokines IL-1β, IL-6 and TNFα released by activated microglia contribute to a BBB breakdown, which plays a significant role in neuroinflammation [65]. BoxA, one of the highly conserved DNA binding domains of HMGB1 protein, competitively binds to TLRs and does not initiate pro-inflammatory responses [66, 67]. It has been shown that BoxA treatment reduces TLR2 and TLR4 activation and prevents the breakdown of the BBB [68]. Prevention of the post-ischemic BBB breakdown could reduce the infiltration of monocytes, which would become macrophages followed by their initial activation to M2 phenotype. Our results should not be interpreted so much as evidence that our treatment inhibited the expression of the M2 phenotype, but rather that our treatment, by reducing BBB disruption, infarct size, and inflammation, alleviated the need for the body’s defense mechanism. Therefore, we believe that the reduction in the expression of M2 markers in the ischemic brains of TLR2 and TLR4 suppressed animals could be attributed to a treatment-associated attenuation of BBB breakdown.

As attenuation of post-ischemic induction of TLRs 2 and 4 has been shown to reduce the severity of acute brain inflammation and infarct size, we hypothesized that their suppression would also preserve post-ischemic neurological function and promote recovery of somatosensory and motor functions. The degree of post-stroke brain damage induced in our experimental subjects could be graded as moderate to severe, based on the mNSS obtained following reperfusion. Consistent with our proposed hypothesis, we found that shRNA-mediated gene silencing of TLRs 2 and 4 preserved post-stroke motor function and/or facilitated recovery of somatosensory and motor functions. Several studies (mentioned earlier) have reported on the role of TLR2 and TLR4 in post-stroke pathogenesis. However, to our knowledge, this is the first study to demonstrate conclusively that TLRs 2 and 4 exert a detrimental influence on post-stroke neurological function and recovery. The limitations of this study include a lack of experiments in both females and in another animal species (such as the mouse) to further confirm the general findings of a deleterious influence of TLRs 2 and 4 on post-stroke neurological recovery.

Conclusions

We conclude that the expression of TLRs 2 and 4 in the ischemic brain is maintained to a high degree until at least seven days following reperfusion, contributes to the severity of acute inflammation, and impedes neurological recovery. Future studies in our laboratory will address these limitations while also focusing on other key molecules involved in TLR2 and TLR4 downstream signaling. We believe that either TLRs 2 and 4 or their downstream signaling molecules could evolve as promising therapeutic targets in the future for ischemic stroke treatment.

Supplementary Material

1663518_Supp_Fig1

Supplementary Fig. 1. Post-ischemic induction of CD16 in the ischemic brain. Quantitative real-time PCR analysis depicting the mRNA expression of CD16 as fold change over sham in the ischemic brain of male rats subjected to 2-h focal cerebral ischemia followed by reperfusion for three and seven days (CD16 mRNA expression at three days after reperfusion shown in Fig 4 was shown for comparison). 18SrRNA served as the house-keeping gene. Histograms and error bars indicate the mean and the SEM, respectively; n=6.