Benserazide is neuroprotective and improves functional recovery after experimental ischemic stroke by altering the immune response

Abstract

Stroke is a leading cause of permanent disability worldwide. Despite intensive research over the last decades, key anti-inflammatory strategies that have proven beneficial in pre-clinical animal models have often failed in translation. The importance of neutrophils as pro- and anti-inflammatory peripheral immune cells has often been overlooked in ischemic stroke. However, neutrophils rapidly infiltrate into the brain parenchyma after stroke and secrete an array of pro-inflammatory factors including reactive oxygen species, proteases, cytokines, and chemokines exacerbating damage. In this study, we demonstrate the neuroprotective and anti-inflammatory effect of benserazide, a clinically used DOPA decarboxylase inhibitor, using both in vitro models of inflammation and in vivo mouse models of focal cerebral ischemia. Benserazide significantly attenuated PMA-induced NETosis in isolated human neutrophils. Furthermore, benserazide was able to protect both SH-SY5Y and iPSC-derived human cortical neurons when challenged with activated neutrophils demonstrating the clinical relevance of this study. Additional in vitro data suggest the ability of benserazide to polarize macrophages towards M2-phenotypes following LPS stimulation. Neuroprotective effects of benserazide are further demonstrated by in vivo studies where peripheral administration of benserazide significantly attenuated neutrophil infiltration into the brain, altered microglia/macrophage phenotypes, and improved the behavioral outcome post-stroke. Overall, our data suggest that benserazide could serve as a drug candidate for the treatment of ischemic stroke. The importance of our results for future clinical trials is further underlined as benserazide has been approved by the European Medicines Agency as a safe and effective treatment in Parkinson's disease when combined with levodopa.

Keywords: Benserazide; Functional improvement; Human iPS neuronal cells; Ischemic stroke; NETosis; Neuroinflammation.

Figure 1

Benserazide reduces PMA-induced inflammatory reactions of phagocytes, especially of neutrophils. (a) Benserazide reduced the chemiluminescence signal of PMA-stimulated human full blood samples in a dose-dependent manner. PMA stimulation (1 µg/ml) of full blood samples resulted in a significant higher luminescent reaction when compared with benserazide co-stimulated samples. ****p < 0.0001; technical n = 5. (b, c) Benserazide treatment markedly reduced NETosis in PMA-stimulated human neutrophils (400 nM PMA, 9 min incubation before imaging), as shown by a decreased SYTOX green signal. The addition of 5 µM benserazide to PMA-challenged or -unchallenged neutrophils significantly inhibited NETosis (in PMA-challenged neutrophils: p = 0.038; in control cells (no PMA added): p < 0.0001). However, a lower concentration of benserazide (2.5 µM) had no protective effect when used on PMA-stimulated neutrophils. In line, neutrophils stimulated with benserazide (2.5 µM) only had no inducing effect on NETosis when compared with unstimulated control neutrophils (cntrl normalized to 100%) either. If not stated differently, data are in comparison to PMA-stimulated neutrophils; *p = 0.038, **p = 0.0076, ***p = 0.0007, **** p < 0.0001; technical n = 14–31. (c) Representative IncuCyte images show the occurrence of NETosis as detected by SYTOX green positive signals; one illustrative image per treatment regime and time point (9 & 300 min after PMA-stimulation +/- Benserazide treatment at 2.5 or 5 µM); scale bar (red) 200 µm. (d) Benserazide treatment decreased the toxicity of PMA-activated neutrophils on SH-SY5Y neuronal cells significantly. Exposure of SH-SY5Y cells to PMA-stimulated neutrophils caused robust neuronal cell death. Importantly, the addition of 25 or 50 µM benserazide decreased the neuronal death substantially. **p < 0.01; technical n = 6–12. (e) Benserazide has no direct neurotoxic effect on human iPSC-derived neurons. Treatment of human iPSC-derived neurons with different concentrations of benserazide (2.5, 5, or 10 µM) did not decrease but rather promote cell viability (5 µM benserazide p = 0.0547), as shown by the resazurin assay. Technical n = 6–8. (f) Benserazide improved neuronal cell viability after stimulation with PMA-activated neutrophils in vitro. Stimulation of human iPSC-derived neurons with control neutrophils (unstimulated), PMA-stimulated neutrophils respectively, caused massive reduction in neuronal cell viability (p < 0.0001). A PMA-accelerated decrease of neuronal cell viability was alleviated by benserazide (p < 0.006). **p = 0.006; ***p = 0.0002; ****p < 0.0001; technical n = 6–12; all statistics: One-way ANOVA with Tukey’s multiple comparison test; SEM.

Figure 2

Benserazide alters protein expression of Arg1 and HO-1 in LPS-challenged RAW264.7 macrophages. (a) Representative images of Western blots show Arg1 and HO-1 protein expression of RAW 264.7 macrophages stimulated for 24 and 48 h with LPS (50 ng/ml) in presence of 0, 2.5, or 5 µM benserazide. At 24 h, the addition of 2.5 and 5 µM of benserazide caused a strong protein expression of both Arg1 (b, p < 0.05) and HO-1 (d, p < 0.01) when compared with LPS-stimulated macrophages. At 48 h, Arg1 (c) was still significantly increased in presence of benserazide (p < 0.01). HO-1 (e), however, was significantly reduced at 48 h in presence of 5 µM benserazide when compared with LPS-stimulation only (p < 0.05). Actin: loading control; at 24 h: N = 4/group, at 48 h: N = 4/group (2/cntrl); bands were quantified and normalized to β-actin; One-way ANOVA with Tukey’s multiple comparison test, *p < 0.05, **p < 0.01, SEM.

Figure 3

Benserazide is protective against ischemia-induced cell-loss. (a) After permanent ischemia, vehicle-treated mice showed significantly higher infarction volumes than mice subjected to benserazide treatment. *p = 0.031; N = 11/10. (b) Benserazide-treated mice showed significantly smaller lesion volumes at one day post-transient ischemia when compared with vehicle-treated mice. Relative infraction volumes were calculated according to Shuaib’s formula. Two-tailed student’s t-test, *p = 0.0458, SEM, N = 12/15.

Figure 4

Benserazide ameliorated neurological and motor functional deficits post-stroke. (a) Benserazide-treated mice show a significant better outcome in general and functional neurological deficits post-stroke when compared with vehicle-treated mice (p = 0.0425, N = 13/15) and (b) the overall body weight loss within the first 7 days post-transient ischemia was notably reduced in benserazide-treated mice (p = 0.031, N = 14/6). Using the CatWalk gait analysis, mice subjected to benserazide treatment showed improved functional motor outcome post-stroke when compared with vehicle-treated mice in the following categories: kinetic motor function, namely SwingSpeed of the RH paw (c, p < 0.05), StandIndex (mean) (d, p < 0.05), and MinIntensity of the LF paw (e, p < 0.05). All two-tailed student’s t-test, *p < 0.05, SEM, benserazide: N = 10–18, vehicle: N = 6–15.

Figure 5

Benserazide reduces astrogliosis and MPO-immunoreactivity post-stroke without altering the general microglia/macrophages activation. (a) One-day after pMCAO the signal of MPO, a histopathological marker for neutrophils, was significantly lower in the benserazide-treated group when compared with the vehicle-treated group (p = 0.0003). (b–d) MPO⁺ cells were quantified from a region of interest (ROI) in the border zone of the infarction (b); representative images MPO⁺ cells of vehicle- (c) or benserazide- (d) treated mice. (e) Even though, the Iba1-immunoreacivity was increased in the lesion area when compared with the corresponding contralateral area of both treatment groups (p = 0.0035, p = 0.0005 respectively), no benserazide-induced effect was detected (p > 0.05). (f) In the peri-ischemic area, the GFAP-reactivity was markedly decreased in benserazide-treated mice when compared with vehicle-treated mice at 7 days after tMCAO (p = 0.0043). (g–h) illustrative images of the GFAP-immunoreactivity in vehicle- (g) and benserazide- (h) treated ischemic mouse brain slices; white boxes indicate ROIs. (i) At 3 dpi (tMCAO), the signal of Arg1⁺ cells was significantly higher in benserazide-treated animals when compared to vehicle-treated animals (p = 0.0389); representative images of Arg1-immunoreactivity from vehicle- (j) and benserazide- (k) treated ischemic mouse brains; white boxes indicate ROIs. Scale bars (c, d) 100 µm; (g, h, j, k) 1000 µm; (a, f, i) two-tailed student’s t-test with or without Welch’s correction, (e) One-way ANOVA with Tukey’s multiple comparison test, ** p < 0.005, *** p < 0.0005; SEM, benserazide: N = 7–10, vehicle: N = 6–9; (c, d, g, h, j, k) staining intensities were increased for illustration purposes using ZEN software (Zeiss).

Figure 6

Benserazide treatment altered gene expression in the post-ischemic brain. The gene expression of tissue samples from peri-ischemic brain area at 3 days post-permanent ischemia was analyzed. (a) Tnf and (b) Hmox1 were significantly reduced in the peri-ischemic area of mice treated with benserazide when compared with vehicle-treated mice (Tnf: p < 0.001; Hmox1: p < 0.005). However, gene expression levels of (c) PADi4 (PAD4), (d) Aif1 (Iba1), or (e) GFAP differed insignificantly between benserazide- and vehicle-treated animals in the peri-ischemic area. Two-tailed student’s t-test, **p <0.005, ***p < 0.001, SEM, N = 6–7