Neuroprotection Comparison of Rosmarinic Acid and Carnosic Acid in Primary Cultures of Cerebellar Granule Neurons

Abstract

Neurodegenerative disorders such as amyotrophic lateral sclerosis (ALS), Alzheimer's disease, and Parkinson's disease, are characterized by the progressive loss of neurons in specific regions of the brain and/or spinal cord. Neuronal cell loss typically occurs by either apoptotic or necrotic mechanisms. Oxidative stress and nitrosative stress, along with excitotoxicity and caspase activation, have all been implicated as major underlying causes of neuronal cell death. Diverse nutraceuticals (bioactive compounds found in common foods) have been shown to have neuroprotective effects in a variety of in vitro and in vivo disease models. In the current study, we compared the neuroprotective effects of two polyphenolic compounds, rosmarinic acid and carnosic acid, which are both found at substantial concentrations in the herb rosemary. The capacity of these compounds to rescue primary cultures of rat cerebellar granule neurons (CGNs) from a variety of stressors was investigated. Both polyphenols significantly reduced CGN death induced by the nitric oxide donor, sodium nitroprusside (nitrosative stress). Rosmarinic acid uniquely protected CGNs from glutamate-induced excitotoxicity, while only carnosic acid rescued CGNs from caspase-dependent apoptosis induced by removal of depolarizing extracellular potassium (5K apoptotic condition). Finally, we found that carnosic acid protects CGNs from 5K-induced apoptosis by activating a phosphatidylinositol 3-kinase (PI3K) pro-survival pathway. The shared and unique neuroprotective effects of these two compounds against diverse modes of neuronal cell death suggest that future preclinical studies should explore the potential complementary effects of these rosemary polyphenols on neurodegenerative disease progression.

Keywords: antioxidants; neuronal apoptosis; nutraceuticals; polyphenols; rosemary.

Figure 1

Structural comparison of rosmarinic acid (A) and carnosic acid (B).

Figure 2

Rosmarinic acid and carnosic acid each protect cerebellar granule neurons (CGNs) from nitrosative stress. (A) CGNs were co-incubated for 24 h in serum-free culture medium containing 25 mM KCl (Control; Con) alone, or with rosmarinic acid (ROS; 50 μM) or carnosic acid (CAR; 10 μM) and sodium nitroprusside (SNP; 100 μM), or SNP alone. Following incubation, CGNs were fixed and stained with Hoechst dye to visualize the nuclei. Decolorized (black & white) panels are shown to emphasize nuclear morphology and gray panels show the bright field images of the same fields. Scale bar indicates 10 µm. (B) Quantitative assessment of cellular apoptosis for CGNs in untreated controls, SNP alone, and SNP plus ROS or CAR. Cells were quantified by counting as either living or apoptotic based on nuclear morphology, and the percentage of cells showing apoptotic nuclei (either condensed or fragmented morphology) was determined. Data are expressed as the means ± SEM, n = 4. ^### indicates p < 0.001 compared to control, *** indicates p < 0.001 compared to SNP alone as determined using one-way ANOVA with a post hoc Tukey’s test.

Figure 3

Rosmarinic acid, but not carnosic acid, protects CGNs from excitotoxicity. (A) CGNs were co-incubated for 24 h in serum-free culture medium containing 25 mM KCl (Control; Con) alone or with rosmarinic acid (ROS; 20 μM) or carnosic acid (CAR; 10 μM) and glutamate/glycine (G/G; 100 μM/10 μM final concentrations), or G/G alone. Following incubation, CGNs were fixed and stained with Hoechst dye to visualize the nuclei. Decolorized (black & white) panels are shown to emphasize nuclear morphology and the gray panels show the bright field images of the same fields. Scale bar indicates 10 µm. (B) Quantitative assessment of cellular apoptosis for CGNs in untreated controls, G/G alone, and G/G plus ROS or CAR. Cells were quantified as either living or apoptotic based on nuclear morphology, and the percentage of cells showing apoptotic nuclei (either condensed or fragmented morphology) was determined. (C) MTT cell viability assay was used to confirm the protective effects of ROS against G/G excitotoxicity. CGNs were co-incubated with 100 μM glutamate and 10 μM glycine (G/G), G/G + ROS and percentage of viability was calculated against control treatment. Data are expressed as the means ± SEM, n = 4. ^### indicates p < 0.001 compared to control, * indicates p < 0.05 compared to G/G alone, *** indicates p < 0.001 as determined using one-way ANOVA with a post hoc Tukey’s test.

Figure 4

Carnosic acid, but not rosmarinic acid, protects CGNs from caspase-dependent apoptosis. (A) CGNs were co-incubated for 24 h in serum-free culture medium containing 25 mM KCl (Control; Con). Treated cells had 25 mM KCl medium replaced with a non-depolarizing, low potassium medium (5 mM KCl; 5K apoptotic medium) either alone or with carnosic acid (CAR; 20 μM). Following incubation, CGNs were fixed and stained with Hoechst (blue) and β-tubulin antibody (green) to visualize the nuclei and microtubule network, respectively. Scale bar indicates 10 µm. (B) Quantitative assessment of cellular apoptosis for CGNs in untreated controls, 5K apoptotic medium, and 5K plus ROS or CAR. Cells were quantified as either living or apoptotic based on nuclear morphology, and the percentage of cells showing apoptotic nuclei (either condensed or fragmented morphology) was determined. (C) MTT cell viability assay was used to confirm the protective effects of CAR against caspase-dependent apoptosis. CGNs were co-incubated with 5K, 5K + CAR and percentage of viability was calculated against control treatment. Data are expressed as the means ± SEM, n = 4. ^### indicates p < 0.001 compared to control, *** indicates p < 0.001 compared to 5K apoptotic condition alone as determined using one-way ANOVA with a post hoc Tukey’s test.

Figure 5

Carnosic acid protects CGNs from caspase-dependent apoptosis through the PI3K pro-survival pathway. (A) CGNs were incubated for 24 h in serum-free culture medium containing 25 mM KCl (Control; Con). Treated cells had the plating medium replaced with a non-depolarizing, low potassium medium (5 mM KCl; 5K apoptotic medium) either alone or co-treated with carnosic acid (CAR; 20 μM) or with CAR + wortmannin (Wort; 100 nM). Following incubation, CGNs were fixed and stained with Hoechst dye to visualize the nuclei. Decolorized (black & white) panels are shown to emphasize nuclear morphology and the gray panels show the bright field images of the same fields. Scale bar indicates 10 µm. (B) Quantitative assessment of cellular apoptosis for CGNs in untreated controls, 5K, 5K + CAR, and 5K + CAR + Wort. Cells were quantified by counting as either living or apoptotic based on nuclear morphology, and the percentage of cells showing apoptotic nuclei (either condensed or fragmented morphology) was determined. Data are expressed as the means ± SEM, n = 4. ^### indicates p < 0.001 compared to control, *** indicates p < 0.001 compared to 5K apoptotic medium, ^xxx indicates p < 0.001 compared to 5K + CAR, as determined using one-way ANOVA with a post hoc Tukey’s test.

Figure 6

Carnosic acid protects CGNs from caspase-dependent apoptosis independently of the AKT or MEK/ERK pro-survival pathways. CGNs were incubated for 24 h in serum-free culture medium containing 25 mM KCl (Control; Con). Treated cells had the plating medium replaced with a non-depolarizing, low potassium medium (5 mM KCl; 5K apoptotic medium) either alone or co-treated with carnosic acid (CAR; 15 μM) alone or (A) CAR + AKT inhibitor (AKT inh; 10 μM) or (B) CAR + MEK inhibitor PD98059 (PD; 10 μM). Following incubation, CGNs were fixed and stained with Hoechst dye to visualize nuclei. (A,B) Quantitative assessment of cellular apoptosis for CGNs in untreated controls, 5K, 5K + CAR, 5K + CAR + AKT inh, and 5K + CAR + PD. Cells were quantified by counting as either living or apoptotic based on nuclear morphology, and the percentage of cells showing apoptotic nuclei (either condensed or fragmented morphology) was determined. Data are expressed as the means ± SEM, n = 4. ^### indicates p < 0.001 compared to control, *** indicates p < 0.001 compared to 5K apoptotic medium alone, as determined using one-way ANOVA with a post hoc Tukey’s test.