Impairment of neuronal tyrosine phosphatase STEP worsens post-ischemic inflammation and brain injury under hypertensive condition

Abstract

Hypertension is associated with poor outcome and higher mortality in patients with ischemic stroke. The impairment of adaptive vascular mechanisms under hypertensive condition compromises collateral blood flow after arterial occlusion in patients with acute ischemic stroke resulting in hypoperfusion. The increased oxidative stress caused by hypoperfusion is thought to be a trigger for the rapid evolution of ischemic infarct volume under hypertensive condition. However, the cellular factors and pathways that contribute to the exacerbation of ischemic brain injury under hypertensive condition is not yet understood. The current study reveals that predisposition to hypertension leads to basal loss of function of the neuron-specific tyrosine phosphatase STEP, which plays a crucial role in neuroprotection against excitotoxic insult. The findings further show that a mild ischemic insult in hypertensive rats triggers an early onset and sustained activation of the neuronal extracellular signal regulated kinase (ERK MAPK), a member of the mitogen activated protein kinase family and a substrate of STEP. This leads to rapid increase in the activation of neuronal NF-κB, expression of neuronal cyclooxygenase-2 and subsequent biosynthesis of the pro-inflammatory mediator prostaglandin E2, resulting in rapid morphological transformation of microglia to the pro-inflammatory state and subsequent exacerbation of ischemic brain injury. Restoration of STEP signaling with intravenous administration of a STEP-derived peptide mimetic reduces the pro-inflammatory response in neurons, activation of microglia, and ischemic brain injury. The findings suggest that the basal loss of STEP function under hypertensive condition contributes to the exacerbation of ischemic brain injury by enhancing post-ischemic inflammatory response. The study not only presents a novel role of STEP in regulating neuroimmune communication but also highlights the therapeutic potential of a STEP-mimetic in mitigating ischemic brain damage under hypertensive condition.

Introduction

Hypertension is the most prevalent comorbidity in human stroke patients, and it is known to worsen stroke outcome, with large infarct volumes and less salvageable tissue (penumbra) compared to normotensive patients [1]. Studies in animal models of chronic hypertension have also demonstrated larger infarct volume after stroke [1,2,3]. One of the most important consequences of hypertension is its effect on cerebral arteries. Changes in vascular morphology and tone as well as structurally smaller arterial lumens results in impairment of the adaptive vascular mechanisms required to maintain stable cerebral blood flow [1, 4, 5]. These vascular changes increase the brain’s vulnerability to ischemic injury after arterial occlusions, as they compromise the development of collateral blood flow from adjacent non-ischemic vascular territories [6]. The resulting hypoperfusion induces oxidative stress in the brain, which along with the canonical pathway of ischemic brain injury caused by overactivation of glutamate receptors evokes a robust post-ischemic inflammatory response to further amplify stroke-induced brain damage [3, 7,8,9,10]. However, the causal link between the high basal oxidative stress and enhanced post-ischemic inflammatory response observed under hypertensive condition is not yet fully understood.

Increased oxidative stress under hypertensive condition leads to aberrant alteration in intracellular redox homeostasis that can affect the function of multiple proteins in the brain [7, 11, 12]. We have identified the brain enriched and neuron specific tyrosine phosphatase STEP (STriatal Enriched Phosphatase) as one such protein, which is specifically expressed in neurons and plays a critical role in neuronal survival against excitotoxic insults [13,14,15,16,17,18]. STEP, also known as Ptpn5 includes both membrane associated (STEP₆₁) and cytosolic (STEP₄₆) isoforms formed by alternative splicing of a single gene [19]. STEP₆₁ is the most ubiquitously expressed STEP isoform in the brain [19], and its activity is regulated by several post-translational modifications. Dopamine/D1 receptor-mediated dephosphorylation of a critical serine residue (ser 221) in a conserved domain termed kinase interacting motif (KIM) of STEP₆₁ renders it inactive in terms of its ability to bind to its substrates [20]. Whereas dephosphorylation of this residue following glutamate-mediated N-methyl-D-aspartate receptor (NMDAR) stimulation allows STEP₆₁ to bind to its substrates [15], which includes mitogen activated protein kinase/extracellular regulated kinase (ERK MAPK), stress activated kinase p38, as well as the non-receptor tyrosine kinases Pyk2 and Fyn [15, 18, 21, 22]. It has also been shown that phosphorylation of two SP/TP sites in a second conserved domain of STEP₆₁ termed as kinase specificity sequence helps maintain its stability [23]. STEP₆₁ is a component of the NMDAR signaling pathway and its rapid activation following an excitotoxic insult in neuron cultures or ischemic insult in normotensive rats have been shown to provide initial neuroprotection. While degradation of active STEP over time leads to secondary activation of deleterious processes resulting in progression of neurotoxicity and ischemic brain damage [18, 24]. These findings demonstrate that an ischemic insult not only triggers a multitude of cytotoxic pathways but also triggers some endogenous protective responses to limit the extent of brain injury. Furthermore, studies in aged animals and cell cultures have showed that oxidative stress due to depletion of the cellular antioxidant glutathione leads to dimerization and subsequent loss of function of STEP₆₁ [25, 26]. Since hypertension is also associated with increased oxidative stress in the brain, it raises the possibility that impairment of basal STEP function under hypertensive condition could adversely affect the severity of neurological disorders. As such, the present study sought to investigate whether impairment of STEP function under hypertensive condition could play a role in enhancing post-ischemic inflammatory response and amplifying stroke-induced brain damage when compared to normotensive controls.

Methods

Materials and reagents

Male Sprague-Dawley (SD) rats and spontaneously hypertensive (SHR) rats weighing 290–300 g and 10–14 weeks of age were purchased from Envigo (Indianapolis, IN, USA). Rats were housed in pairs in standard cages, maintained in a 12-h light/dark vivarium (light off at 18 h) and with access to food and water ad libitum. Approval for animal experiments was given by the University of New Mexico, Health Sciences Center, Institutional Animal Care and Use Committee. The Animal Welfare Assurance number is D16-00228 (A3350-01) and USDA registration number is 85-R-0014. The study was conducted in strict accordance with the recommendations in the “Guide for the Care and Use of Laboratory Animals” of the National Institutes of Health and complies with the ARRIVE guidelines.

Antibodies used were as follows: monoclonal anti-phospho-ERK 1/2 (Cat #: 9106; Cell Signaling), polyclonal anti-phospho-ERK1/2 (Cat #: AF1018; R&D systems), polyclonal anti-ERK2 (Cat #: 9108; Cell Signaling); polyclonal anti-COX-2 (Cat #: ab15191; Abcam), monoclonal anti-β-tubulin (Cat #: T0198; Sigma), monoclonal anti-STEP (Cat # NB300-202; Novus), monoclonal anti-NeuN (Cat # NBP1-92693; Novus), polyclonal phospho-NF-κB p65 (Cat # 3031: Cell signaling), monoclonal IκBα (Cat # 4814: Cell signaling) and polyclonal anti-Iba-1 (Cat #: 019-19741; Wako). Secondary antibodies used were as follows: Horseradish peroxidase (HRP) conjugated goat anti-rabbit (Cat #: 7074; Cell Signaling), HRP-conjugated goat anti-mouse (Cat #: 7076; Cell Signaling), AlexaFluor-488 conjugated goat anti-mouse (Catalog #: A48286, Thermo Fisher Scientific, MA, USA) and Cy3 conjugated goat anti-rabbit (Cat # 111-165-144, Jackson Immuno-Research Laboratories). Additional information on the above antibodies is presented in Supplementary Table 1. Vectashield Antifade Mounting Medium with DAPI was obtained from VectorLabs (Catalog #H-1200). Prostaglandin E2 enzyme immunoassay kit was obtained from Arbor Assays (Catalog # K051-H1). Protein gel electrophoresis reagents were obtained from Bio-Rad (California, USA), BCA protein assay kit (Catalog #: 23225) and West Pico Plus Chemiluminescent Substrate (Catalog # 34580) were obtained from Thermo Fisher Scientific (Waltham, MA, USA) and all other reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA).

Induction of transient focal cerebral ischemia

To induce focal cerebral ischemia, both SD and SHR rats were subjected to middle cerebral artery occlusion (MCAO) using the intraluminal suture method as described earlier [27]. Briefly, animals were anesthetized by spontaneous inhalation of isoflurane (2%) in oxygen. Rectal temperature was maintained at 37 ± 1^o C with an electrical heating pad both during surgery and recovery. The right common carotid artery (CCA) and the external carotid artery were exposed through an incision in the ventral midline neck region. A 6 − 0 silk suture was tied around the ECA and the smaller vessels extending from it to permanently ligate the vessel. Another suture was loosely tied around the ECA stump. The CCA and the internal carotid artery (ICA) were temporarily clipped before insertion of occluding filament. Thereafter, a small incision was made in the ECA close to the permanent suture and a silicon-rubber-coated 4 − 0 monofilament (Doccol Corporation) was advanced through the ECA into the internal carotid artery to a length of 18 mm from the bifurcation to occlude the middle cerebral artery. Depending on the experiment, the occluding filament was either kept in place for 10–60 min and animals were sacrificed immediately after filament removal, or the filament was gently retracted after 60 min to allow reperfusion for varying time periods (3–24 h). For reperfusion the incision was closed, and the animals were allowed to recover in their cages. A sub-group of SHR rats received a single intravenous (through femoral vein) dose of vehicle (PBS) or the STEP-mimetic, TAT-STEP-myc (3 nmol/g of body weight) at the onset of reperfusion [24]. An observer blinded to the study design assigned the SHR rats subjected to MCAO arbitrarily to one of the two groups: vehicle or TAT-STEP-myc treatment. Severity Both SD and SHR rats were processed for biochemical, molecular, and histochemical studies at the specified time periods.

Functional assessment

Neurological severity score: At 24 h after MCAO and reperfusion severity of neurologic deficit was assessed using 5-point neurological severity score: 0 = no observable deficits, 1 = failure to extend left forepaw, 2 = circling to the left, 3 = falling to left, 4 = no spontaneous walking with a depressed level of consciousness, 5 = death [27].

Infarct volume measurement

To evaluate infarct size by 2,3,5-triphenyl tetrazolium chloride monohydrate (TTC) staining 24 h after MCAO and reperfusion, SD and SHR rats were anesthetized using sodium pentobarbitol (150 mg/kg, i.p.). The brains were removed, sliced into six 2-mm thick sections using the coronal rat brain matrix, incubated in 2% TTC solution in saline for 30 min at 37^oC and scanned on a Umax Powerlook scanner. For detection of neuronal degeneration with Fluoro-Jade C staining, at the specified time points following MCAO and reperfusion (6 h, 12 h and 24 h after reperfusion) or following sham surgery, SD and SHR rats were anesthetized, perfused intracardially with ice-cold 4% paraformaldehyde in 0.01 M phosphate buffered saline (PBS). Brains were rapidly removed and post-fixed in the same solution for 4 h, cryoprotected in 15% and 30% sucrose in PBS, and then frozen in Optimal Cutting Temperature (OCT) compound for cryosectioning. Fluoro-Jade C staining was performed on 16 μm brain sections as described earlier [28]. Briefly, brain sections were air-dried, dehydrated, incubated with 0.06% potassium permanganate (15 min), rinsed in water, and stained with 0.001% Fluoro-Jade C (Histo-Chem) in 0.1% acetic acid for 30 min with gentle agitation on ice. Sections were rinsed in water, air-dried, and mounted using DPX mounting media followed by imaging using Olympus IX-71 fluorescent microscope as described earlier [24]. For infarct size measurement, in each section the total area in the contralateral side and the non-infarcted area on the lesioned side were measured using Image J software. The areas on each side were summed over the number of sections evaluated, and the volumes were calculated as follows: [(volume of contralateral side - non-infarcted volume of the lesioned side)/volume of contralateral side] × 100 [28, 29].

Detection of STEP dimers using non-reducing SDS-PAGE

To detect dimers of STEP, brains from SD and SHR rats were processed as reported previously [25]. Briefly, cortical tissues were homogenized in a lysis buffer containing 50-mM Tris-HCl, pH 7.5, 100-mM NaCl, 1% Nonidet P-40, and cocktail of protease inhibitors, incubated on ice for 45 min and then centrifuged at 13,000 rpm for 10 min to collect the supernatant. Equal protein from each sample was diluted 1:1 with 2x Laemmli sample buffer, without β-mercaptoethanol (non-reducing condition) and then separated on 6% SDS-PAGE. In a parallel series of experiments protein sample were diluted 1:1 with 2x Laemmli sample buffer, with β-mercaptoethanol (reducing condition), boiled for 10 min and then separated on 7.5% SDS-PAGE. The proteins in the gel were transferred to polyvinylidene difluoride membrane and processed for immunoblot analysis with anti-STEP (1:2000) antibody.

Tyrosine phosphatase assay

Cortical tissues from SD and SHR rats were processed as described in the previous section. Equal amount of protein from supernatant of each sample (1.5-2 mg) was processed for immunoprecipitation with anti-STEP antibody [25]. The immune complexes bound to protein G beads were then washed 3 times in a buffer containing 50-mM Tris-HCl, pH 7.4, 150-mM NaCl, and 0.1% NP-40 and then once in a buffer containing 30-mM HEPES (pH 7.4) and 120-mM NaCl. For phosphatase assay, the beads were incubated for 30 min at 30^oC in 100 µl of reaction buffer (30-mM HEPES, pH 6.0, 150-mM NaCl, and 10-mM pNPP) and reaction was stopped by addition of 0.9 mL of 0.2-N NaOH. Phosphatase activity of STEP₆₁ was measured by colorimetric quantitation of the formation p-nitrophenolate at 410 nm using a spectrophotometer [25]. To assess the amount, STEP immunoprecipitated immune complexes bound to the beads were eluted with SDS sample buffer and processed for immunoblot analysis with anti-STEP (1:2000) antibody.

Immunoblotting

For immunoblotting of ERK MAPK and IκB (inhibitor of nuclear factor-κB) proteins, SD and SHR rats after sham surgery, MCAO or reperfusion were perfused with PBS, brains were removed and sliced into 2-mm thick sections. Cortical tissue from the third rostral section of ipsilateral hemisphere was homogenized in Laemmli sample buffer, boiled at 100^oC for 10 min, centrifuged at 14,000 x g for 10 min and equal amount of protein from the lysates were processed for SDS-PAGE and immunoblotting [28] with the following primary antibodies: monoclonal anti-phospho-ERK1/2 (1:4000), polyclonal anti-ERK2 (1:4000), monoclonal IκBα (1:1000) and monoclonal anti-β-tubulin (1:10,000). Horseradish peroxidase-conjugated goat anti-rabbit (1:2000) and goat anti-mouse (1:2000) were used as secondary antibodies. The immune complexes on the membrane were detected on X-ray film after treatment with West Pico Plus Chemiluminescent reagent. Quantitation of p-ERK1/2, ERK2 and IκBα was done by computer-assisted densitometry and ImageJ analysis.

Immunohistochemistry

SD and SHR rats after sham surgery or following MCAO and reperfusion (6 h) were processed for cryosectioning as described earlier. For immunohistochemistry, 16 μm brain sections were blocked with 10% normal goat serum and 3% BSA in PBS-T (PBS with 0.2% Triton-X-100) for 1 h at room temperature and incubated overnight with one of the following antibodies: polyclonal anti-phospho-ERK1/2 (1:100), polyclonal anti-COX-2 (1:100), anti-phospho-NF-κB p65 (1:100), monoclonal anti-NeuN (1:300) or polyclonal anti-Iba-1 (1:200) antibody. After extensive washing with PBS-T, sections were incubated with AlexaFluor-488 conjugated goat anti-mouse (1:300) or Cy3 conjugated goat anti-rabbit (1:400) for 1 h at room temperature. Sections were then washed three times with PBS-T and mounted using Vectashield mounting medium with DAPI for counterstaining of nuclei. All sections were imaged using fluorescent microscope (Olympus IX-71) and the fluorescence micrographs were processed for morphological assessment of microglia, as described below.

Assessment of microglial morphology by single cell skeletal and fractal analysis

Morphological changes in Iba-1 positive microglial cells from SD and SHR rats were assessed from fluorescence photomicrographs using ImageJ and following previously published protocol [30, 31]. Individual microglia were randomly selected from several areas of the ipsilateral cortex (n = 45–50 cells from 3 independent experiments) and processed for skeletal and fractal analysis. Briefly, fluorescence photomicrographs were converted to 8-bit grayscale image and the brightness/contrast of the converted images was adjusted to best visualize the branches. The threshold was then adjusted to obtain the binary images, which were then saved and utilized for both skeletal and fractal analysis. For skeletal analysis, despeckle, close and remove outliers’ functions were applied to the binary images to eliminate any single-pixel background noise and gaps between processes. The binarized images were then skeletonized, and the skeletonized images were analyzed using the AnalyzeSkeleton (2D/3D) plugin (http://imagej.net/AnalyzeSkeleton). The number of branches, number of junctions, maximum branch length and the longest shortest path were calculated for each isolated cell using the analyze skeleton function. For fractal analysis, the binary image was converted to an outline. The FracLac plug-in for ImageJ (https://imagej.nih.gov/ij/plugins/fraclac/fraclac.html) was used to analyze the cells with “box counting” applied and the “grid design Num G” set to 4 for evaluation of the following morphological parameters for each isolated cell: (1) fractal dimension (D), to assess overall morphological complexity of microglia, where a higher D means greater complexity; (2) lacunarity (λ), to measure heterogeneity in shape, where lower λ value imply homogeneity; (3) area (µm²), to measure total cell area; (4) perimeter (µm), defined as the total length of the cell boundary; and (5) circularity, a non-dimensional measure of the roundness of a cell, where the circularity value of a circle is 1.

Enzyme linked immunoassay for prostaglandin E2

Prostaglandin E2 (PGE₂) level was measured in cortical lysates from SD and SHR rats subjected to sham surgery or MCAO and reperfusion (6 h). Brains were rapidly removed from anesthetized rats and sliced into 2-mm thick sections. The ipsilateral cortex from the third rostral brain slice was homogenized in ice cold PBS containing protease inhibitor cocktail and centrifuged at 14,000 rpm for 10 min at 4^oC. The supernatant was collected and equal amount of protein from each sample (10 µg) was used for PGE₂ assay using the PGE₂ enzyme immunoassay kit (K051-H1) according to manufacturer’s protocol and as described earlier [32]. Briefly, a fixed amount of a sample (or standard PGE₂), HRP-conjugated PGE₂ and the antibody against PGE₂ were added to a microtiter plate well coated with IgG to capture the antibody specific for PGE₂. The assay is based on competitive binding technique, where the PGE₂ present in the sample competes with the HRP-conjugated PGE₂ for sites on the bound PGE₂ antibody. After 2 h incubation with gentle shaking at room temperature, and a brief washing to remove unbound sample and excess HRP-conjugated PGE₂, a substrate solution was added to each well to determine the enzyme activity of the bound HRP-conjugated PGE₂. The intensity of the color was measured at 450 nm wavelength using a microplate reader (BioTek Synergy H1, USA).

RNA extraction and quantitative real time PCR

SD and SHR rats subjected to sham surgery or MCAO followed by reperfusion (6 h) were processed for RNA extraction. Briefly, brains were rapidly removed from anesthetized rats and sliced into 2 mm thick coronal sections. Ipsilateral cortex obtained from the third rostral brain section was processed for total RNA extraction using RNeasy Lipid Tissue Mini Kit (Catalog no. 74804; Qiagen, Hilden, Germany), and 2 µg of total RNA from each sample was then converted to cDNA using the High-Capacity cDNA Reverse transcription kit (Catalog no. 4368814, Applied Biosystems) according to manufacturer’s protocol. Quantitative real-time PCR (RT-PCR) was performed using the cDNA product, TaqMan™ Universal PCR Master Mix (Catalog no. 4304437, Applied Biosystems) and gene-specific TaqMan probes, which were obtained from Thermo Fisher Scientific. The TaqMan probes and their accession number are as follows: NF-kB1 (p50) - Rn01399572_m1; RelA (p65) - Rn01502266_m1; COX-2 (PTGS2) - Rn01483828_m1; CD68 - Rn01495634_g1; TNFα - Rn01525859_g1; β-actin - Rn00667869_m1; and HPRT1 - Rn01527840_m1. Amplification conditions included 2 min at 50^oC and 10 min at 95^oC and then run for 40 cycles at 95^oC for 15 s and 60^oC for 1 min on the Quant studio-4 sequence detection system (Applied Biosystems). The mRNA expression levels of the target genes were normalized to the reference genes β-actin or HPRT level in each sample and relative gene expression was calculated using the 2^ΔΔCT method.

Construction and purification of recombinant STEP-derived peptide

A recombinant DNA construct for the STEP-derived peptide mimetic (TAT-STEP-Myc peptide) was generated using a bacterial expression vector, expressed in E. Coli and purified as described previously [18]. Briefly, STEP₆₁ cDNA lacking the PTP domain and encoding only 173–279 amino acids (ΔPTP) was sub-cloned into a pTrc-His-Myc-TOPO expression vector (Invitrogen). A 11 amino acid TAT peptide (trans-activator of transcription of human immunodeficiency virus) nucleotide sequence was inserted at the N-terminal of the STEP ΔPTP cDNA to render the peptide cell permeable. A point mutation was introduced by site-directed mutagenesis (Pfu Turbo, Stratagene) at serine 221 within the KIM domain (S221A) to render the peptide constitutively active in terms of its ability to bind to its substrates. Point mutations were also introduced at threonine 231 (T231E) and serine 244 (S244E) in the KIS domain to mimic the phosphorylated form that helps to maintain the stability of STEP [23]. The modified TAT-STEP-Myc peptide was expressed in E. Coli and purified using BD-Talon resin (BD Biosciences, Bedford, MA, USA) according to manufacturer’s protocol. The purity of the eluted recombinant peptide (> 95%) was confirmed by SDS-PAGE. Schematic diagram of the TAT-STEP-myc peptide is presented in Fig. 5A.

Statistical analysis

Data in the text and figures are expressed as mean ± SD or ± SE and each data point represents individual biological replicates. Statistical differences between two groups were analyzed using Student’s t test. Neurological severity score data that were not normally distributed were analyzed using non-parametric Mann-Whitney U test and reported as median with interquartile range. Differences were considered statistically significant when p < 0.05. All analysis were performed using GraphPad Prism software (GraphPad software, version 10, 2023).

Results

Early onset and exacerbation of ischemic brain damage in SHR rats

To evaluate the effect of hypertension on the outcome of ischemic stroke both normotensive (SD) and hypertensive (SHR) rats were subjected to mild focal ischemia induced by 60 min of MCAO, followed by reperfusion for 24 h. Assessment of neurological function at 24 h after reperfusion using the modified neurological severity score shows minimal decline in neurological function in SD rats. In contrast, SHR rats show a severe decline in neurological function (Fig. 1A). Following neurological assessment development of brain lesion was assessed by TTC staining. The representative images of TTC stained brain slices show that in SD rats ischemic brain damage is limited mostly to the striatum. In contrast, a relatively larger infarct size is observed in SHR rats, which encompasses both the striatum and the cortex (Fig. 1B). Quantitative analysis of the TTC stained sections show significant increase in ischemic brain damage in SHR rats (Fig. 1C, D), which is consistent with earlier findings [1,2,3]. To evaluate whether an early onset of injury contributes to the exacerbation of brain damage under hypertensive condition, we further examined the temporal progression of ischemic injury in the acute phase (6–24 h). For this experiment SD and SHR rats were subjected to MCAO (60 min) followed by reperfusion for varying time periods (6, 12–24 h). Coronal brain sections from each reperfusion time point were stained with Fluoro-Jade C, a fluorescent marker for detection of degenerating neurons that has been widely used for the confirmation of neurotoxicity [33]. The representative photomicrographs (Fig. 1E - G) and the corresponding bar diagram (Fig. 1H) show onset of neurodegeneration in both SD and SHR rats within 6 h of reperfusion. However, in SD rats, Fluoro-Jade C-labeled cells are barely visible at 6 h of reperfusion and remains limited to the striatum at 24 h of reperfusion. Whereas in SHR rats, a significant increase in Fluoro Jade C staining is visible within 6 h of reperfusion, scattered in the striatum and cortex, which increases further at 12 h and 24 h of reperfusion encompassing both the striatum and the cortex. Together these observations suggest that in SHR rats even a mild ischemic insult accelerates the progression of neurodegeneration resulting in exacerbation of ischemic brain damage.

Fig. 1

Predisposition to hypertension accelerates the progression of ischemic brain injury. (A - G) SD and SHR rats were subjected to MCAO (60 min) followed by reperfusion for the specified time periods (6, 12–24 h). (A) Neurological severity score was assessed 24 h after MCAO on a 5-point scale and represented as individual data points with median value. (B) Representative photomicrographs of TTC-stained brain slices (2 mm) show the extent of brain damage 24 h after the onset of ischemia. (C) Bar graphs represent quantitative analysis of total infarct volume (mean ± SD, n = 5–6/group). (D) Line graph represents total infarct area within each slice (mean ± SD, n = 5–6/group). (E-G) Representative photomicrographs of coronal brain sections at 6, 12 and 24 h after ischemia stained with Fluoro-Jade C. (H) Bar graphs represent quantitative analysis of infarct volume at 6, 12 and 24 h respectively (mean ± SD, n = 3). Significant difference between means was assessed by student t-test and presented as *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001

Increased dimerization and loss of activity of neuronal tyrosine phosphatase STEP in SHR rats

To evaluate the molecular basis of accelerated brain damage in SHR rats, we next examined the effect of hypertension on the neuroprotectant, STEP. Increased oxidative stress has been shown to cause loss of STEP function through dimerization, and hypertension is associated with increased oxidative stress [7, 25]. Therefore, to examine the effect of hypertension on STEP dimerization cortical tissue from SD and SHR rats were processed for immunoblot analysis with anti-STEP antibody under non-reducing condition. In both SD and SHR rats the existence of monomeric and dimeric forms of STEP₆₁ is evident from the migration (upward shift in electrophoretic mobility) of STEP₆₁ protein band to a position that corresponds to double the molecular weight (100–150 kDa) compared with STEP₆₁ monomer (61 kDa; Fig. 2A). However, a significant increase in the dimerization of STEP₆₁ is observed in SHR rats when compared with the SD controls (Fig. 2A). Immunoblot analysis of the same samples under reducing condition, where disulfide bond formation cannot occur, show equal amount of the monomeric form of STEP₆₁ in both SD and SHR rats (Fig. 2A, lower panel). To further determine whether the increased dimerization of STEP₆₁ in SHR rats has any effect on its phosphatase activity, STEP was immunoprecipitated from the cortex of SD and SHR rats and tyrosine phosphatase activity of STEP was measured using pNPP as a substrate. A significant decrease in STEP activity is observed in the cortex of SHR rats when compared with the SD controls (Fig. 2B). The findings demonstrate that predisposition to hypertension leads to dimerization and loss of activity of STEP₆₁.

Fig. 2

Loss of basal STEP function in SHR rats is associated with sustained increase in neuronal ERK MAPK phosphorylation during ischemia and reperfusion. (A) Cortical tissue lysates from SD and SHR rats with equal amount of protein were processed for immunoblot analysis under non-reducing condition (without β-mercaptoethanol) using anti-STEP antibody to evaluate STEP dimer formation (upper panel). The same lysates were also processed for immunoblot analysis under reducing condition (with β-mercaptoethanol) using anti-STEP antibody to assess total STEP expression level (lower panel). Bar graphs represent quantitative analysis of STEP dimer formation (mean ± SD, n = 6). (B) STEP was immunoprecipitated from equal amount of cortical tissue lysates of SD and SHR rats using anti-STEP antibody and the immune complexes were processed either for tyrosine phosphatase activity assay using pNPP as a substrate or immunoblotting with anti-STEP antibody to ensure equal pull down of STEP. Bar graphs represent quantitative analysis of STEP phosphatase activity (mean ± SD, n = 6) (C) SD and SHR rats were subjected to sham surgery or MCAO (I) for the specified time periods (10 and 60 min). In some experiments 60 min MCAO was followed by reperfusion (RPF) for the specified time periods (3, 6 and 18 h). Tissue lysates with equal amount of protein from the ipsilateral cortex were analyzed using anti-phospho-ERK1/2 (pERK1/2) MAPK antibody (upper panels). Equal protein loading was confirmed by re-probing the blots with anti-ERK2 antibody (lower panels). Bar graphs represent quantitative analysis of pERK levels (mean ± SD, n = 4). (A-C) Significant difference between means was assessed by student t-test and presented as *p < 0.01, **p < 0.001 and ***p < 0.0001. (D) SD and SHR rats were subjected to sham surgery or 60 min MCAO followed by reperfusion for 6 h and then processed for immunohistochemical staining with anti-phospho-ERK1/2 (pERK - red) antibody, a neuron specific antibody, anti-NeuN (green) and nuclear stain DAPI (blue). Arrows in the representative photomicrographs demonstrate localization of phosphorylated ERK in NeuN positive cells in the ipsilateral cortex

Increased inflammatory response in neurons following ischemic insult in SHR rats

To assess whether the loss of phosphatase activity of STEP could prolong the phosphorylation of its substrates following a pathological stimuli, we investigated the temporal profile of post-ischemic phosphorylation of ERK MAPK (a substrate of STEP), whose sustained activation is thought to play a role in mechanisms that triggers neurodegeneration [34]. SD and SHR rats were subjected to MCAO for varying time periods (0 min, 10 min and 60 min), or MCAO for 60 min followed by reperfusion for varying time periods (3 h, 6 h and 18 h). Immunoblot analysis of cortical tissue punches from the ipsilateral hemisphere show that during the initial insult (MCAO, 10 min), ERK MAPK phosphorylation increases significantly in SHR rats when compared to the corresponding SD rats, and remain elevated for the entire time period of the ischemic insult and reperfusion (Fig. 2C). Immunohistochemical analysis of coronal brain sections (ipsilateral cortex) with anti-phospho-ERK MAPK and NeuN antibodies further show that the increase in ERK MAPK phosphorylation in SHR rats is localized primarily in neurons (Fig. 2D). The findings suggest that under hypertensive condition the basal loss of function of endogenous STEP prolongs the duration of ischemia-induced ERK MAPK activation in neurons.

Earlier studies in different cell types have reported that ERK MAPK is involved in the activation of nuclear factor-κB (NF-κB), which upregulates the expression of COX-2, an enzyme that catalyzes the biosynthesis of the proinflammatory mediator PGE₂ [35,36,37,38,39]. Since STEP inhibits ERK MAPK, with the loss of activity of this inhibitory signal under hypertensive condition, the sustained ERK MAPK activation following ischemia could trigger an early onset of NF-κB activation in SHR rats. To test this hypothesis, we evaluated the gene expression profile of the NF-κB p50: p65 heterodimer, the most abundant form of NF-κB activated by pathologic stimuli. Quantitative RT-PCR analysis of mRNA from ipsilateral cortex of SD and SHR rats following MCAO (60 min) and reperfusion (6 h) show that expression of both p50 and p65 subunits are upregulated in SHR rats (Fig. 3A-H). Subsequent studies evaluated whether ischemia-reperfusion in SD and SHR rats have any effect on IκB protein degradation, a seminal step in NF-κB activation [40, 41]. Immunoblot analysis of the cortical lysates from the ipsilateral hemisphere show significant decrease in IκB level in SHR rats (Fig. 3I), indicating increased IκB degradation and NF-κB activation. Immunohistochemical staining of coronal brain sections (ipsilateral cortex) with anti-phospho-NF-κB p65 and NeuN antibodies further show nuclear translocation of active NF-κB primarily in neurons (Fig. 3J). In a parallel series of experiments, we also evaluated the expression of COX-2 in SD and SHR rats. Quantitative analysis of COX-2 gene expression by RT-PCR shows significant upregulation of COX-2 mRNA level in the ipsilateral cortex of SHR rats (Fig. 4A-D). Immunohistochemical staining of coronal brain sections from the ipsilateral cortex further show that upregulation of COX-2 protein level in SHR rats is specifically localized in neurons (Fig. 4E). Elevated COX-2 protein level in SHR rats is also associated with significantly higher levels of PGE₂ release (Fig. 4F). The prolonged ERK MAPK phosphorylation and activation of the NF-κB/COX-2/PGE₂ signaling cascade indicate an early onset of inflammatory response in neurons under hypertensive condition.

Fig. 3

Activation of neuronal NF-κB in SHR rats following ischemia and reperfusion. (A-J) SD and SHR rats were subjected to sham surgery, or MCAO (60 min) followed by reperfusion (RPF) for 6 h. (A-H) Tissue extracts from the ipsilateral cortex were processed to measure mRNA levels of NFκB1 and RelA by quantitative real-time-PCR. Transcript levels were normalized against reference genes β-actin or HPRT. Bar graphs represent relative expression of NFκB1 and RelA genes (mean ± SE, n = 5–6). (I) Tissue lysates with equal amount of protein from the ipsilateral cortex were analyzed using anti-IκB antibody (upper panels). Equal protein loading was confirmed by re-probing the blots with anti-β-tubulin antibody (lower panel). Bar graps represent quantitative analysis of IκB protein levels (mean ± SD, n = 4). (A-I) Significant difference between means was assessed by student t-test and presented as *p < 0.05, **p < 0.01 and ***p < 0.001. (J) Coronal brain sections through the ipsilateral cortex of SD and SHR rats (MCAO 60’/RPF 6 h) were processed for immunohistochemical staining with anti-phospho-NFκB (pNFκB - red) antibody, anti-NeuN (green) antibody and nuclear stain DAPI (blue). Arrows in the representative photomicrographs demonstrate nuclear localization of phosphorylated NFκB in NeuN positive cells in the ipsilateral cortex

Fig. 4

Increase in neuronal COX-2 expression and prostaglandin E2 level in SHR rats following ischemia and reperfusion. (A-F) SD and SHR rats were subjected to sham surgery or MCAO (60 min) followed by reperfusion (RPF) for 6 h. (A-D) Tissue extracts from the ipsilateral cortex (MCAO-60’/RPF 6 h) were processed to measure mRNA levels of COX-2 by quantitative real-time-PCR. Transcript levels were normalized against housekeeping genes β-actin or HPRT. Bar graphs represent relative expression of COX-2 gene (mean ± SE, n = 5–6). (E) Coronal brain sections through the ipsilateral cortex of SD and SHR rats (MCAO-60’/RPF 6 h) were processed for immunohistochemical staining with anti-COX-2 antibody (red), anti-NeuN antibody (green) and nuclear stain DAPI (blue). Arrows in the representative photomicrographs demonstrate localization of COX-2 in NeuN positive cells in the ipsilateral cortex. (F) PGE₂ level was measured by enzyme immunoassay in the supernatants from tissue lysates extracted from ipsilateral cortex of SD and SHR rats (MCAO-60’/RPF 6 h). Bar graphs represent quantitative analysis of PGE₂ level (mean ± SD, n = 3). (A-D, F) Significant difference between means was assessed by student t-test and presented as *p < 0.01

Restoration of STEP signaling in SHR rats attenuates neuroinflammatory response induced by ischemic injury

To further examine whether loss of function of endogenous STEP under hypertensive condition is a key factor in the up regulation of post-ischemic inflammatory response, STEP signaling was restored in SHR rats with intravenous administration of a STEP-mimetic (TAT-STEP-myc peptide). The STEP-mimetic constitutes of the KIM and KIS domain of STEP₆₁ and was rendered cell permeable by fusion to the 11 amino acid protein transduction domain (TAT) of the human immunodeficiency virus-type 1 (Fig. 5A) [18]. The feasibility of delivering the TAT-STEP-myc peptide to the brain, peptide uptake in brain cells in vivo, as well as into neurons in culture, the ability of the peptide to bind to its substrates and the efficacy of the peptide to confer neuroprotection against excitotoxicity in neuronal cultures and ischemic stroke in normotensive rats has been established in several earlier studies [18, 24, 28, 42]. For these experiments SHR rats were subjected to MCAO (60 min) followed by intravenous administration of a single dose of the STEP-mimetic at the onset of reperfusion. Immunoblot analysis of cortical brain lysates from the ipsilateral hemisphere show that restoration of STEP signaling significantly reduces ERK MAPK phosphorylation observed 6 h after reperfusion (Fig. 5B). Analysis of mRNA from ipsilateral cortex at 6 h reperfusion by quantitative RT-PCR also show significant decrease in gene expression of both p50 and p65 subunits of NF-κB following treatment (Fig. 5C-F). Furthermore, significant reduction in the degradation of IκB protein level (Fig. 5G) and inhibition of nuclear translocation NF-κB in neurons at 6 h reperfusion (Fig. 5H) confirms attenuation of NF-κB activation in peptide treated SHR rats. In addition, evaluation of COX-2 gene expression (Fig. 5I, J), COX-2 protein levels in neurons (Fig. 5K) and the release of the pro-inflammatory mediator PGE₂ (Fig. 5L) at 6 h reperfusion show significant reduction in the peptide treated SHR rats. These findings confirm that basal loss of endogenous STEP activity in SHR rats contributes to the enhanced activation of ERK MAPK-NFκB-COX-2 signaling in post-ischemic neurons.

Fig. 5

Treatment with STEP mimetic alleviates post-ischemic inflammatory response in neurons of SHR rats. (A) Diagram of STEP₆₁ indicating the positions of the phosphatase domain, putative proteolytic sites (PEST), transmembrane domain (TM), polyproline rich regions (PP), kinase interacting motif (KIM), kinase specificity sequence (KIS) and known phosphorylation sites. The STEP mimetic (TAT-STEP-myc peptide) was generated from STEP₆₁. The peptide was rendered cell-permeable by fusion to the 11 amino acid protein transduction domain (TAT) of the human immunodeficiency virus-type I at the N-terminus and has a myc-tag at the C-terminus. The serine residue in the KIM domain was mutated to alanine to allow the peptide to bind constitutively to its substrates. The threonine and serine residues in the KIS domain were mutated to glutamic acid to render the peptide resistant to degradation. (B-L) SHR rats were subjected to MCAO (60 min) followed by reperfusion (RPF) for 6 h. TAT-STEP-myc peptide (TAT-STEP) was administered in a subset of SHR rats at the onset of reperfusion. (B) ERK MAPK phosphorylation and (G) IκB protein level were evaluated by immunoblot analysis of tissue extracts from ipsilateral cortex. Blots were re-probed with (B) anti-ERK and (G) anti-β-tubulin antibodies (lower panels). Bar graphs represent mean ± SD (n = 3). (C-F, I, J) Tissue extracts from the ipsilateral cortex were processed to measure mRNA levels of (C, D) NFκB1, (E, F) RelA and (I, J) COX-2 by quantitative real-time PCR. Bar graphs represent relative expression of NFκB1, RelA and COX-2 normalized against reference genes β-actin or HPRT (mean ± SE, n = 6). (H, K) Coronal brain sections through the ipsilateral cortex were processed for immunohistochemical staining with (H) anti-pNFκB (red) antibody, anti-NeuN antibody (green) and nuclear stain DAPI (blue); and (K) anti-COX-2 antibody (red), anti-NeuN antibody (green) and nuclear stain DAPI (blue). Arrows in the representative photomicrographs demonstrate (H) nuclear localization of pNFκB in NeuN positive cells and (K) localization of COX-2 in NeuN positive cells. (I, J). (L) PGE₂ level was measured by enzyme immunoassay using supernatants obtained from ipsilateral cortex. Bar graphs represent quantitative analysis of PGE₂ level (mean ± SD, n = 3). (B-G, I, J, L) Significant difference between means was assessed using student t-test and presented as *p < 0.05, **p < 0.01 and ***p < 0.001

Changes in microglial morphology and morphometric parameters to amoeboid form in SHR rats in response to ischemic injury

Earlier studies have indicated that excessive PGE₂ release in the brain could elicit rapid morphological transformation of brain microglia to the pro-inflammatory form [28, 43]. To investigate changes in microglial morphology in the acute phase after stroke, SD and SHR rats were subjected to MCAO (60 min), and microglial morphology in the ipsilateral cortex was evaluated with anti-Iba-1 antibody, 6 h after reperfusion. A general morphometric analysis of microglial phenotype shows that in both SD and SHR sham rats microglia exhibit a highly ramified morphology with a small cell soma and extensive arborization of processes (Fig. 6A, B), necessary for continuous tissue surveillance, neurotransmission, and maintenance of synaptic integrity. After transient MCAO and reperfusion, microglia in the ipsilateral cortex of SD rats appears to be transitioning towards de-ramified morphology with fewer higher order branches (Fig. 6C). In contrast, post-stroke microglia in ipsilateral cortex of SHR rats displays amoeboid phenotype with large cell bodies and devoid of processes (Fig. 6D). Under pathological conditions, such rapid transition of microglial morphology from highly ramified to amoeboid form is considered to be an objective measure of alteration in microglial function from surveillance mode to pro-inflammatory state [44, 45]. A more detailed assessment of post-stroke microglial morphology by fractal analysis performed on individual microglia (n = 45–50) from the ipsilateral cortex confirms that microglia in SHR rats has less complex branching pattern (fractal dimension) and heterogeneity (lacunarity) when compared to SD rats after MCAO and reperfusion (Fig. 6E, F). Furthermore, significant decrease in overall cell area, cell perimeter and cell circularity in microglia from SHR rats point towards the presence of more amoeboid microglia in post-ischemic SHR rats when compared to corresponding SD rats (Fig. 6G-I). Also, measurement of microglial branching complexity by single cell skeletal analysis (n = 45–50) show significant decrease in the number of branches, branch length, number of junctions and longest shortest path in SHR rats when compared to SD rats (Fig. 6J-M). Additional studies evaluated the gene expression levels of CD68, a transmembrane glycoprotein whose high expression in amoeboid microglia indicates phagocytic activity [46], and TNFα, a microglia derived cytokine that has been implicated in sustaining the pro-inflammatory activation of microglia [47,48,49]. Analysis of cDNA derived from ipsilateral cortex of SD and SHR rats following MCAO (60 min) and reperfusion (6 h) by quantitative RT-PCR shows that gene expression of both CD68 (Fig. 6N-Q) and TNFα (Fig. 6R-U) are upregulated in SHR rats. Collectively, these findings indicate enhanced microglial activation within 6 h of ischemia-reperfusion in SHR rat brain.

Fig. 6

Enhanced morphological transformation of microglia to amoeboid form in SHR rats following ischemia and reperfusion. SD and SHR rats were subjected to sham surgery, or MCAO (60 min) followed by reperfusion (RPF) for 6 h. (A-D) Coronal brain sections through the ipsilateral cortex were processed for immunohistochemical staining with anti-Iba1 antibody. Representative photomicrographs in the left panels show microglial morphology in (A) SD sham, (B) SHR sham, (C) SD MCAO-60’/RPF-6 h and (D) SHR MCAO-60’/RPF-6 h. Amplified images of selected Iba-1 positive microglia (in yellow box) and skeletonized pattern of the microglia are shown in the right panels. (E-M) Microglial cells (n = 45–50 cells) imaged from the ipsilateral cortex of SD and SHR rats following MCAO / reperfusion were processed for assessment of morphological parameters by (E-I) fractal and (J-M) skeletal analysis. Scatter plots with mean ± SD presents the eight parameters: fractal dimension, lacunarity, cell area, cell perimeter, cell circularity, number of branches, number of junctions, maximum branch length and longest shortest path. (N-U) Tissue extracts from the ipsilateral cortex were processed to measure mRNA levels of (N-Q) CD68 and (R-U) TNFα by quantitative real-time PCR. Bar graphs represent relative expression of CD68 and TNFα normalized against reference genes β-actin or HPRT (mean ± SE, n = 6). Significant difference between means was assessed by student t-test and presented as *p < 0.01, **p < 0.001 and ***p < 0.0001

Restoration of STEP signaling in SHR rats mitigates post-ischemic microglial activation

To determine whether restoration of STEP signaling could also attenuate microglial activation, the STEP-mimetic was administered intravenously at the onset of reperfusion in a subset of SHR rats subjected to MCAO (60 min). Immunohistochemical analysis of microglial morphology following staining with Iba-1 antibody show that microglia from the ipsilateral cortex of peptide treated SHR rats, transition towards ramified morphology with smaller cell body and more processes, when compared to the amoeboid phenotype of microglia in untreated SHR rats (Fig. 7A, B). Assessment of microglial phenotype by fractal analysis show significant increase in morphological features directly associated with increased ramification that includes increased complexity of branching pattern, cell area, cell perimeter and cell circularity (Fig. 7C-G). Assessment of microglial features through skeletal analysis further reveals increased number of branches and junctions as well as increased total process length (Fig. 7H-K). In addition, evaluation of gene expression level of CD68 and TNFα by RT-PCR shows significant decrease in their level in the peptide treated SHR rats (Fig. 7L-O). Together these findings imply that treatment with STEP-mimetic enhances the ramified population of microglia in SHR rats to engage them towards surveillance state from a pro-inflammatory state.

Fig. 7

STEP-mimetic alleviates post-ischemic microglial activation in SHR rats. SHR rats were subjected to MCAO (60 min) and reperfusion (RPF) for 6 h. STEP-mimetic (TAT-STEP-myc) was administered in a subset of SHR rats at the onset of reperfusion. (A-B) Immunohistochemical staining of microglia with anti-Iba1 antibody. Representative photomicrographs in the left panels show changes in microglial morphology in SHR rats treated with or without the STEP-mimetic. The right panels show amplified images of a select Iba-1 positive microglia (in yellow box) and skeletonized pattern of microglial morphology. (C-K) Microglial cells (n = 45–50 cells) from the ipsilateral cortex were processed for assessment of morphological parameters by (C-G) fractal and (H-K) skeletal analysis. Scatter plots with mean ± SD presents the eight parameters: fractal dimension, lacunarity, cell area, cell perimeter, cell circularity, number of branches, number of junctions, maximum branch length and longest shortest path. (L-O) Tissue extracts from the ipsilateral cortex were processed to measure mRNA levels of (L, M) CD68 and (N, U) TNFα by quantitative real-time PCR. Bar graphs represent relative expression of CD68 and TNFα normalized against reference genes β-actin or HPRT (mean ± SE, n = 6). (C-O) Significant difference between means was assessed by student t-test and presented as *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001

Restoration of STEP signaling in SHR rats attenuates ischemic brain injury

In subsequent studies SHR rats subjected to MCAO (60 min) were injected intravenously with the STEP-mimetic at the onset of reperfusion. Assessment of neurological function at 24 h after reperfusion shows significantly lower neurological impairment in peptide treated group when compared to the untreated controls (Fig. 8A). Following functional assessment, the rats were processed to evaluate the extent of brain damage by TTC staining. The representative photomicrographs (Fig. 8B) and quantitative analysis of the TTC stained sections (Fig. 8C, D) show significant reduction in ischemic brain damage in the peptide treated SHR rats as compared with the untreated controls. The finding demonstrates that restoration of STEP signaling mitigates the exacerbation of ischemic brain injury in hypertensive rats.

Fig. 8

STEP-mimetic attenuates exacerbation of ischemic brain injury in SHR rats. (A) SHR rats were subjected to MCAO (60 min) and reperfusion (RPF) for 24 h. STEP-mimetic (TAT-STEP-myc) was administered in a subset of SHR rats at the onset of reperfusion. (A) Neurological severity score was assessed 24 h after MCAO on a 5-point scale and represented as individual data points with median value. (B) Representative photomicrographs of TTC-stained brain slices (2 mm) show the extent of brain damage 24 h after ischemia and reperfusion. (C) Bar graphs represent quantitative analysis of the total infarct volume (mean ± SD, n = 6). (D) Line graphs represent area of infarction within each slice. Significant difference between means was assessed by student t-test and presented as *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001

Discussion

Increasing evidence show that post-ischemic inflammation accounts for the secondary progression of brain damage, and the severity of stroke outcome under comorbidities depends on the extent of this inflammatory response. A key finding of the present study is that ischemic insult under hypertensive condition leads to an early onset of inflammatory response in the hyperacute phase (0–6 h), which has not been evaluated earlier. The molecular trigger for this hyperacute inflammatory response seems to be initiated in neurons through the activation of neuronal ERK MAPK and NF-κB signaling pathway, which are known to regulate a range of pro-inflammatory factors. This is associated with enhanced transformation of brain microglia to phagocytic proinflammatory state that is known to be involved in the secondary progression of ischemic brain damage. Since the function of the endogenous neuroprotectant STEP is compromised in neurons under hypertensive condition it raises the possibility that impairment of STEP function is a critical factor for the early onset of post-ischemic inflammatory response and exacerbation of ischemic brain injury in SHR rats. Consistent with this interpretation, restoration of STEP signaling in SHR rats with intravenous administration of the STEP-mimetic attenuates neuronal ERK MAPK/NFκB signaling, microglial transformation to the pro-inflammatory state and exacerbation of ischemic brain injury in SHR rats. Taken together, these findings provide a molecular basis for the severity of stroke outcome under hypertensive condition and further imply that intervention with STEP-peptide mimetic acts as a brake to suppress the early onset of inflammatory response and limit the detrimental impact of inflammation on ischemic brain damage. A schematic representation of this signaling cascade is presented in Fig. 9.

Fig. 9

Schematic diagram representing impairment of STEP function under hypertensive condition causing upregulation of post-ischemic inflammatory response and exacerbation of brain injury. (A) Hypertension diminishes the basal activity of the neuroprotectant STEP in the brain. (B) The impairment of STEP function in SHR rats facilitates prolonged activation of neuronal ERK MAPK (substrate of STEP) following a mild ischemia and reperfusion. This in turn triggers an early onset of neuronal NF-kB activation, cyclooxygenase 2 (COX-2) expression, and prostaglandin E2 (PGE₂) release resulting in upregulation of microglial activation and subsequent exacerbation of ischemic brain injury

An important outcome of the study is the rapid and significant increase in neuronal ERK MAPK activation within 10 min of the ischemic insult, which remains elevated for the time period of the study. Although activation of ERK MAPK is widely associated with neuronal maturation, survival, and long-term potentiation [50,51,52], several earlier studies have indicated that ERK MAPK signaling could also have detrimental effects on neuronal function and survival [53,54,55,56]. It has been postulated that the duration of ERK MAPK activation plays a pivotal role in cell survival and death. While transient activation has been shown to promote a variety of physiological processes, persistent or abnormal ERK MAPK activation contribute to the activation of a range of pathological processes resulting in neurodegeneration [55, 57,58,59]. The duration of ERK MAPK activation is known to be regulated by the sequential activation of a stimulatory and an inhibitory pathway [60]. Tyrosine phosphatase STEP is one such inhibitory pathway, whose rapid activation following excitotoxic stimulation is known to limit the duration of ERK MAPK activation in neurons [15]. As such, under hypertensive condition the sustained ERK MAPK activation observed in our study following ischemic insult could be attributed to the loss of function of endogenous STEP.

Earlier studies in different cell types indicate that ERK MAPK is involved in NF-κB activation through increased phosphorylation and subsequent proteasomal degradation of IkBα resulting in the release and nuclear translocation of active NF-κB dimers [41, 61,62,63]. Such increased NF-κB activation in the brain under pathological condition can either confer neuroprotection or trigger neuroinflammatory response, depending on the composition and time of the activated dimer [64, 65]. NF-κB proteins consist of homo- and heterodimers of the five family members: RelA (p65), RelB, cRel, p50 and p52 [66]. Among the 15 potential dimer associations, only 3 have been extensively studied [67]. Amongst these, the p50: p50 homodimer and the cRel: p50 heterodimer seem to induce anti-inflammatory and pro-survival response [65, 68,69,70,71,72,73]. Whereas the p50: p65 heterodimer is thought to mediate the deleterious effects of NF-κB activation and is considered to play a pivotal role in the progression of neurodegeneration triggered by traumatic insults such as stroke, as well as glutamate excitotoxicity and Aβ-toxicity [65, 71, 72, 74,75,76]. Consistent with this interpretation, our findings show increased transcription of both p50 and p65 subunits, degradation of IκB as well as nuclear translocation of phosphorylated p65 under hypertensive condition, indicative of NF-κB activation. Another variable that seems to impact the outcome of NF-κB activation is the temporal profile of its activation. Early activation of NF-κB within several hours of hypoxic ischemia in a neonatal rat model has been shown to produce more neuronal loss, whereas delayed activation several days later reduced apoptotic neuronal loss [77,78,79]. In this context the early onset of post-ischemic NF-κB activation under hypertensive condition observed in our study further indicate its involvement in inducing the expression of pro-inflammatory mediators leading to neuroinflammation and neurodegeneration.

In most of the earlier studies, the role of NF-κB in neurotoxicity and neurodegeneration has been attributed to its upregulation in microglia and astrocytes, resulting in the release of reactive oxygen species and proinflammatory cytokines. On the other hand, NF-κB function in neurons under disease condition is still not fully understood. Although, it is generally considered to promote neuroprotection, several studies in neuronal cultures indicates a potential role of NF-κB activation in neuronal cell death following excitotoxic insult [41, 65, 75, 80,81,82,83]. Our findings now raise the possibility that under hypertensive condition neuronal NF-κB activation after an insult could be the initial trigger for microglial activation and inflammatory response. This notion is supported by additional findings demonstrating upregulation of neuronal COX-2 expression and increased PGE₂ release following ischemic insult under hypertensive condition. COX-2, a pro-inflammatory mediator downstream of NF-κB activation has been shown to play a profound role in PGE₂ synthesis, which contributes to microglial activation, and is a key element in the pathophysiology of inflammatory disorders [35, 43, 84, 85]. Consistent with this interpretation we also observe transition of microglia from ramified to amoeboid form in the ipsilateral cortex of SHR rats, which along with increased expression of CD68 and TNFα is indicative of a pro-inflammatory microglial phenotype [86,87,88]. In addition to their pro-inflammatory immune response, amoeboid microglia also have the ability to engulf stressed but viable neurons in the peri-infarct zone that transiently express membrane anchored “eat me” signals such as phosphatidylserine [45, 89, 90], which could collectively contribute to the progression and exacerbation of ischemic brain injury observed under hypertensive condition. Although morphological transformation of microglia following ischemic insult has been reported in earlier studies [44, 46, 91], the temporal profile and the extent of such transformation in normotensive rats is distinctly different from our observation in hypertensive rats. Following ischemia and reperfusion under normotensive condition microglial process retraction has been observed at 12 h, and by 24 h they become hypertrophic, with enlarged cell bodies, while by 48 h they become more amoeboid-like, which remain visible in the penumbra even one week after the insult [91]. At this delayed time point the capability of the amoeboid microglia to phagocytose cell debris and myelin debris could play a role in attenuating the detrimental effects of inflammation and create a favorable microenvironment for synaptogenesis, neurogenesis, and network rewiring. Thus, it appears, that the partially de-ramified intermediate morphology of post-ischemic microglia observed in our normotensive SD rats may potentially allow them to participate in injury resolution and thereby limiting the extent of brain injury [92, 93]. The outcome of ischemic insult between normotensive and hypertensive rats, with normotensive rats exhibiting lesser brain damage further supports this hypothesis. Taken together the findings present the novel concept that under hypertensive condition activation of neuronal NF-κB signaling pathway in the hyperacute phase accelerates microglial transition towards pro-inflammatory phenotype and is responsible for the exacerbation of ischemic brain damage observed under hypertensive condition.

Another important outcome of the study is that restoration of STEP signaling following ischemia not only attenuates ERK MAPK-NFκB-COX-2 signaling in neurons but also diminishes activation of microglia to the pro-inflammatory form and reduces ischemic brain injury in hypertensive rats. Since ERK MAPK is a substrate of STEP the findings imply that with the basal loss of endogenous STEP function under hypertensive condition the sustained ERK MAPK activation following ischemia serves as the initial trigger for promoting neuronal NF-κB signaling in the hyperacute phase and is essential for the early onset of microglial activation and subsequent post-ischemic inflammatory response. Further support for this interpretation comes from earlier studies demonstrating that excitotoxic insult in STEP deficient neurons in culture induces NF-κB activation [32] and mild ischemic insult in STEP deficient mice leads to enhanced microglial activation, blood-brain barrier disruption and exacerbation of ischemic brain injury [28]. Collectively these findings clearly demonstrate that STEP plays a critical role in regulating neuronal NF-κB signaling and the loss of function of endogenous STEP under pathological condition is a key contributing factor for the acceleration of post-ischemic neuroinflammatory response and exacerbation of ischemic brain damage. The efficacy of the STEP-mimetic in reducing ischemic brain damage in SHR rats observed in the current study, and in STEP KO mice reported earlier further substantiates our interpretation that loss of endogenous STEP contributes to the exacerbation of ischemic brain injury under hypertensive condition. In addition, the study provides mechanistic insight into the role of endogenous STEP in regulating neuroimmune communication in the brain, which provides a molecular basis for targeting STEP as a potential therapeutic approach to mitigate neuroinflammation in CNS pathophysiology.