Neuroprotection by herbal formula FBD and its active compounds

Abstract

FBD, a Chinese herbal formula, composed of Fu Ling [the sclerotium of the fungus of Poria cocos (Schw.) Wolf (Polyporaceae)], Bai Zhu [the rhizome of Atractylodes macrocephala Koidz.(Compositae)], and Danggui [the root of Angelica sinensis (Oliv.) Diels (Umbelliferae)], has been found beneficial in treating brain ischemia-reperfusion (I/R). In this work, oral pretreatment with supercritical CO₂ extract (FBD-CO₂, 37. 5 mg/kg) twice daily for 3.5 days, significantly attenuated brain infarction (p < 0.05), blocked neuron specific enolase (NSE) efflux (p < 0.05) in mice subjected to repetitive 10 min of common carotid artery occlusion following 24 h reperfusion, whether coupled with aqueous extract (FBD-H₂O, 150 mg/kg) or not. Except atractylenolide I and pachymic acid, atractylenolide II, III, levistolid A, and ferulic acid isolated from FBD-CO₂ alone protected neuron-like PC12 cells obviously and concentration-dependently against I/R-like insults (p < 0.05 ~ 0.01) induced by sodium dithionite (10 mM), monosodium glutamate (2 mM), KCl (0.2 M), or hydrogen peroxide (0.2 mM) at 1-100 μM. Even though at sub-active concentrations (< 1 μM), the combination of the six components contained in FBD-CO₂ (10 μg/mL), protected markedly PC12 cells (p < 0.01), in a synergistic fashion. Moreover, intraperitoneal treatment with dual doses of 37. 5 mg/kg FBD-CO₂ and 1. 2 mg/kg combination reduced both infarct size (p < 0.01 and p > 0.05, respectively) and NSE efflux (p < 0.05 and p > 0.05, respectively). Therefore, neuroprotection by FBD on I/R induced neuronal injury originated partially from the synergies of its compounds atractylenolide II, atractylenolide III, levistolid A and ferulic acid.

Keywords::

• Angelica sinensis
• Atractylodes macrocephala
• ischemia-reperfusion
• Poria cocos

Introduction

Traditional Chinese medicine (TCM) has a long history in stroke therapy, and more than 100 herbs are in ordinarily clinical use. As documented in Pharmacopeia of the People’s Republic of China, Fu Ling (Poria, or the sclerotium of the fungus of Poria cocos (Schw.) Wolf (Polyporaceae)), Bai Zhu [white atractylodes rhizome, or the rhizome of Atractylodes macrocephala Koidz. (Compositae)], and Danggui [Chinese angelica root, or the root of Angelica sinensis (Oliv.) Diels (Umbelliferae)] are three essential Chinese medicinal plants, which are clinically used to modulate psychoneural, gastrointestinal, and gynecological disorders, respectively (Adams et al., 2007). Recently, research has focused upon their neuroprotective actions, because the three herbs were being frequently applied to cure stroke in Asia (Gong & Sucher, 1999).

Pharmacological investigation implied that the combination of Fu Ling, Bai Zhu, and Danggui (FBD) had a predominant potential for stroke treatment. Total extracts of FBD prevented ischemia-reperfusion (I/R) induced brain injury in rodents, in part through antineurotoxic and antioxidant activities (Lin et al., 2005; Dong et al., 2006; Zhou et al., 2006; Wu et al., 2007). Neuroprotection of FBD originated principally from its liposoluble factions, because in vitro I/R-like insults in neurons, astrocytes and neuron-like PC12 cells were ameliorated by the total extracts of FBD or its supercritical carbon dioxide extract (FBD-CO₂) (Wang et al., 2004; Zhang et al., 2005; Liu et al., 2007).

Six small compounds, including atractylenolide I, II, III, levistolid A, ferulic acid, and pachymic acid have been identified from bio-fluids of rats orally taking FBD-CO₂ (Cao et al., 2007; Luo et al., 2005; Wang et al., 2006). However, active components in FBD or FBD-CO₂ remained unclear. Up to now, few psychoneural activities were available on these compounds except ferulic acid. As phospholipase A2 inhibitor (Cuella et al., 1996; Giner et al., 2000), and DNA topoisomerases inhibitor (Li et al., 2004), pachymic acid from Fu Ling inhibited tumor (Gapter et al., 2005; Kaminaga et al., 1996), and free-radical-induced hemolysis (Sekiya et al., 2003). Of the three sesquiterpenoids from Bai Zhu, atractylenolide I and III inhibited pro-inflammatory cytokines production and uterine contraction (Li et al., 2007; Zhang et al., 2000). Being an N-methyl-d-aspartate (NMDA) receptor antagonist and antioxidant (Kanski et al., 2002; Ogiwara et al., 2002; Zhang et al., 2003), ferulic acid from Danggui protected brain against glutamate or amyloid beta-induced neurotoxicity and cerebral I/R (Jin et al., 2005; Yu et al., 2006). To identify potential active components and underlie their pharmacological relationships to FBD or FBD-CO₂, the work assessed neuroprotective activities of its six components on neuronal injury in vivo and in vitro.

Materials and methods

Reagents and chemicals

Nimodipine, memantine, monosodium glutamate (MSG), N, N-dimethylformamide, triphenyltertrazolium chloride (TTC), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), vitamin E (Vit E) were from Sigma (St. Louis, MO). Dulbecco’s modified Eagle medium (DMEM) was from Gibco-BRL (Gaithersburg, MD). Enzyme-linked immunosorbent assay (ELISA) kit for neuron-specific enolase (NSE) was from Biosource International (Camarillo, CA).

Preparation of extracts of FBD

Three herbal materials used in this work, and the formula of FBD were reported previously (Lin et al., 2005). The mixed herbal materials were dried and powdered at 60ºC, macerated in 95% ethanol (v/w, 1:1) for 30 min. FBD-CO₂, a brown liquid with density 0.85–0.9 at 60ºC, was obtained at 25. 0 MPa and 38ºC and at a CO₂ flow of 110 kg/h, with a yield of 2.75%. The residue was then extracted with water for 30 min to gain dried brown powder (FBD-H₂O), with a yield of 9.75%. The total extracts include FBD-CO₂ and FBD-H₂O.

Isolation of atractylenolide I, II, III, levistolid A, ferulic acid, and pachymic acid

Atractylenolide I, II, III from Bai Zhu, levistolid A and ferulic acid from Danggui, pachymic acid from Fu Ling were isolated from FBD-CO₂ and identified as reported previously with purity > 98% (Luo et al., 2005). The contents of the six compounds (3.22% in total) in FBD-CO₂ were atractylenolide I 0.55%, II 0.88%, III 0.68%, levistolid A 0.003%, ferulic acid 0.39%, and pachymic acid 0.69%, at a ratio of 1:1.6:1.2:0.005:0.7:1.3, quantified by HPLC (Figure 1).

Figure 1.  Chemical structures of atractylenolide I, II, III, ferulic acid, levistolid A, and pachymic acid.

I/R insults in mice brain

Male ICR mice (26–32 g) were provided by China Pharmaceutical University (Nanjing, China). Bilateral common carotid artery occlusion (CCAO) was performed as previously described (Lin et al., 2005; Zhen & Doré, 2007). In brief, bilateral CCA were separated, and occluded for 10 min at 38ºC, with 0.15 mL blood lost from tail. After cerebral blood flow comeback for 10 min, another 10 min occlusion followed. Sham-operated mice as sham-operated control did not undergo CCAO. The animal procedures in the research process were in strict accordance with the China National Institutes of Healthy Guidelines for the Care and Use of Laboratory Animals.

Pretreatment of extracts of FBD

Clinically, a single formula of FBD consisted of 10 g Poria cocos, 5 g Atractylodes macrocephala and 3 g Angelica sinensis. In this work, the dose of 187. 5 mg/kg of the total extracts of FBD twice per day was calculated according to the herbal formula dose of 18 g per day for a person with body weight of 60 kg, and its total extraction yield 12.5%. Extracts of FBD were dissolved with saline (containing 0.5% Tween 80 and 2% alcohol). Mice were randomized into five groups: I/R group and sham-operated group, 10 mL/kg saline, per oral (p.o.); FBD-pretreated groups, 187. 5 mg/kg FBD, 150 mg/kg FBD-H₂O, 37. 5 mg/kg FBD-CO₂, p.o.; FBD extracts were orally administered twice daily for 3.5 days prior to CCAO.

Measurement of brain infarction and circulating NSE

For TTC staining, 2 pieces of 2-mm thick coronal sections per brain were sliced. After incubation in 2% TTC for 30 min and fixation in 10% formalin for 45 min at 37ºC, brain slices were captured with a digital camera and measured by Zeiss AxioVs 40 image analysis system (Oberkochen, Germany). Brain infarction was presented as following equation:

% brain infarction = white infarct area/slice area × 100%

NSE efflux in serum, another specific biomarker for neuronal injury was assayed by ELISA, with sensitivity of 1 ng/mL.

I/R-like insults in PC12 cells

Neuron-like rat pheochromocytoma PC12 cells were provided by the Institute of Cell Biology (Shanghai, China). The cells were suspended in DMEM supplemented with 10% heat-inactivated newborn calf serum (NCS), benzylpenicillin (100 kU/L) and streptomycin (100 mg/L), and incubated at 37ºC in 5% CO₂.

PC12 cells were exposed to: 1) Na₂S₂O₄ (10 mM), an O₂ scavenger for 1.5 h to induce hypoxic injury; 2) MSG (2 mM) in a Mg²⁺ free Earle’s solution for 4 h to induce neurotoxic injury; 3) H₂O₂ (0.2 mM) for 1 h to induce oxidative injury; or 4) KCl (0.2 M) for 4 h to induce calcium overload injury, then treated with atractylenolide I, II, III, levistolid A, ferulic acid or pachymic acid at 0.1–100 μM, or atractylenolide I (0.24 μM), II (0.38 μM), III (0.27 μM), levistolid A (0.008 μM), ferulic acid (0.20 μM), and pachymic acid (0.13 μM) contained in FBD-CO₂ (10 μg/mL) alone or in combination. Memantine (1 μM), an NMDA antagonist, nimodipine (1 μM), an L-type Ca²⁺ channel blocker, and Vit E (10 μM), a free radical scavenger were set as three positive controls; and the vehicle 0.1% dimethyl sulphoxide (DMSO) artificial cerebrospinal fluid (ACSF) was negative control (Lin et al., 2005; Cao et al., 2006; Goldshmidt et al., 2001).

After 20 h, MTT colorimetric assay was performed to observe the cell viability in PC12 cells and to calculate the neuroprotection by the six compounds. Briefly, MTT solution (0. 5 mg/mL) was added to each culture well. After incubation for an additional 4 h, the formazan crystals were dissolved by addition of 50 μL DMSO, and measured spectrophotometrically at dual wavelengths, 570 nm/650 nm.

Statistical analysis

SPSS 12.0 software and Origin 7.0 software were applied to analyze experimental data, and results were expressed as mean ± SD. All data were evaluated with analysis of variance (ANOVA) following by Dunnett’s t-test for multiple comparisons and p < 0.05 indicates that the difference was statistically significant.

Results

Neuroprotective effects of extracts of FBD in vivo

Twenty-four hours after cerebral repetitive I/R, large NSE efflux and diffused brain infarction were observed in I/R mice. As compared with the sham-operated mice, oral pretreatment with 187. 5 mg/kg FBD or 37. 5 mg/kg FBD- CO₂, attenuated brain infarction remarkably by 40.6% (p < 0.01) and 43.9% (p < 0.05), respectively, and significantly reduced circulating NSE level (p < 0.05). By contrast, FBD-H₂O protected against brain I/R to a lesser extent (Figure 2).

Figure 2.  Effects of FBD, FBD-CO2 and FBD-H2O on brain infarction and circulating NSE efflux in ICR mice subjected to cerebral repetitive I/R. Each column represents mean ± SD of 6 mice. NSE, neuron specific enolase; I/R, ischemia-reperfusion; #p<0.05, ##p<0.01 vs the sham-operated group; *p<0.05, **p<0.01 vs the saline-pretreated I/R group.

Neuroprotective activities of the six compounds of FBD in vitro and in vivo

At concentrations of 1–100 μM, atractylenolide II (Figure 3A), III (Figure 3B), levistolid A (Figure 3C) and ferulic acid (Figure 3D) except atractylenolide I and pachymic acid, inhibited PC12 cells I/R-like insults with concentration-dependent manners. All the actions of the four compounds were significant, but relatively weak with inhibitions less than 20%, even though at 100 μM, with respect to those of nimodipine 26.6–36.2% (Figures 4A and 4D), memantine 38.1% (Figure 4B), and VitE 25.2% (Figure 4C). Herein, the six compounds had no toxicity on innate PC12 cells, at 0.1–100 μM.

Figure 3.  Neuroprotective activities of atractylenolide II, atractylenolide III, levistolid A, ferulic acid from FBD-CO2 on PC12 cells I/R-like insults. Each dot represents mean ± SD of 6 wells. I/R, ischemia-reperfusion; *p<0.05, **p<0.01 vs the vechicle-treated PC12 cells.

Figure 4.  Neuroprotective activities of atractylenolide I, II, III, levistolid A, ferulic acid and pachymic acid combination on PC12 cells I/R-like insults. Values are means ± SD of three different experiments. The A570/650 in control was 0.64 ± 0.02. I/R, ischemia-reperfusion; Ctl, Control; Vh, vehicle; Com, Combination; CO2, FBD-CO2; VE, Vitamin E; ##p < 0.01 vs the control (Ctl); **p<0.01, ***p < 0.001 vs the vehicle (Vh)-treated cells; ††p < 0.01 vs the FBD-CO2-treated cells.

Neither atractylenolide I, II, III, levistolid A, ferulic acid, nor pachymic acid at sub-active concentrations (< 1 μM) had obvious influence on the PC12 cells insults (Table 1). However, once in combination the six components protected additively PC12 cells against all the insults (p < 0.01), though their synergies were still weaker than those of FBD-CO₂ (p < 0.01) (Figure 4 and Table 1).

Table 1.  Neuroprotections (%) of atractylenolide I,II, III, levistolid A, ferulic acid and pachymic acid alone or in combination on PC12 cells I/R-like insults induced by Na₂S₂O₄, MSG, H₂O₂ and KCl in assay I–IV in vitro.

Assay no.	Assay I	Assay II	Assay III	Assay IV
Atractylenolide I	1.8	0.4	0.3	0.7
Atractylenolide II	1.5	0.8	0.0	1.1
Atractylenolide III	2.1	6.6	0.6	1.1
Levistolid A	0.6	0.4	0.0	0.4
Ferulic acid	1.8	0.8	1.3	0.7
Pachymic acid	2.1	0.4	0.0	0.4
Predicted combination	9.9	9.4	2.2	4.4
Actual combination	16.4	21.8	12.6	11.0
Ratio (actual/predicted)	1.7	2.3	5.7	2.5
FBD-CO2	34.3	39.9	27.4	30.9
Ratio (actual/FBD-CO2)	0.5	0.5	0.5	0.4

Neuroprotections (%) were means of three different experiments by MTT assay.

In addition, intraperitoneal treatment with dual doses of 37. 5 mg/kg FBD-CO₂ and 1. 2 mg/kg compound combination (atractylenolide I, 0. 21 mg/kg; II, 0. 33 mg/kg; III, 0. 26 mg/kg; levistolid A, 0. 01 mg/kg; ferulic acid, 0. 15 mg/kg; and pachymic acid, 0. 26 mg/kg, i.p.) administered at 30 min prior to and 5 h after CCAO reduced brain infarction by 52.6% and 40%, respectively, and NSE efflux to different extents. No difference existed between the two treatments (p > 0.05, Figure 5).

Figure 5.  Effects of FBD-CO2 and its six compounds combination on brain infarction and circulating NSE efflux in ICR mice subjected to cerebral repetitive I/R. Each column represents mean ± SD of 6 mice. NSE, neuron specific enolase; I/R, ischemia-reperfusion; #p<0.05, ##p<0.01 vs the sham-operated group; *p<0.05, **p<0.01 vs the saline-pretreated I/R group.

Discussion

In the present work we demonstrated again that total extracts of FBD might prevent and treat I/R-induced neuronal injury. It was likely that neuroprotection of FBD or FBD-CO₂ was derived partially from its compounds, atractylenolide II, III, levistolid A and ferulic acid. As evidence, (i) FBD or FBD-CO2 attenuated brain infarction and NSE efflux; (ii) Except atractylenolide I and pachymic acid, atractylenolide II, III, levistolid A, and ferulic acid isolated from FBD-CO2 alone protected PC12 cells against I/R-like insults; (iii) Once in combination, the six compounds may synergistically protected PC12 cells, and ameliorate brain infarction and NSE efflux, similar to FBD-CO2.

It is well documented that repetitive CCAO results in histological and neurochemical abnormality. Our results indicate that pretreatment with FBD or FBD-CO₂ offers protection against neuronal damage induced by CCAO, as they reduced the infarct size and NSE release from disrupted neurons (Figure 2), in harmony with other studies (Yan et al., 2007; Cho et al., 2007).

The significant protection of FBD-CO₂ on brain I/R injury probably at least lay in the synergistic actions of atractylenolide I, II, III, levistolid A, ferulic acid, and pachymic acid, rather than those of single compounds. By contrast, less neuroprotection was elicited by FBD-H₂O because none of these components was contained in it, except some ferulic acid. During the early stage of brain I/R, energy depletion, lipid peroxidation, depolarization or excited amino acid release consequently resulted in acute neuronal necrosis and delayed neuronal death (Watanabe et al., 1995). In accordance with the findings in vivo, except atractylenolide I and pachymic acid, atractylenolide II, III, levistolid A or ferulic acid attenuated significantly PC12 cells against I/R-like insults, similar to memantine, nimodipine, or Vit E (Figures 3 and 4). However, neither atractylenolide I, II, III, levistolid A, ferulic acid, nor pachymic acid might exclusively account for the whole actions of their combination or FBD-CO₂, whether in vivo or in vitro. At sub-active concentrations (< 1 μM), no compound had obvious impact on the PC12 cells insults. In contrast, once coupled with one another, the six components contained in FBD-CO₂ (10 μg/mL), protected markedly PC12 cells against all the insults (Figure 4). The actual actions (11%–21.8%) of the combination were 1.7–5.7 times over the predicted sums (2.2%–9.9%, Table 1).

The exact mechanism by which FBD alleviated neuronal injury was not very clear in this work, because the neuroprotective actions of the compounds combination accounted for 40%–50% of those of FBD-CO₂ in vitro, but 76%–80% of those of FBD-CO₂ in vivo (Figures 4 and 5). It was strongly attributed to the presence of multiple activities by multiple components. On one hand, various biological activities of the six compounds were implicated in the neuroprotections of their combination or FBD-CO₂. Antiplatelet activity of ferulic acid (Wang et al., 1988), and anti-inflammatory activities of atractylenolide I, III and pachymic acid could contribute to the effects of their combination or FBD-CO₂ on thrombus or inflammation-induced neuronal injury (Giner Larza et al., 2000; Li et al., 2007; Sekiya et al., 2003). On the other hand, some other active compounds from FBD might further be additive to the combination. Recently, more than ten prototype components with psychoneural properties, including butylphthalide (Xu & Feng, 2000), ligustilide (Kuang et al., 2006), and butylidenephthalide (Ko et al., 1997), were identified in rabbit orally receiving Danggui (Lin et al., 1998; Liu et al., 2004; Wang et al., 2005). More bioanalytical and pharmacological tests are needed to uncover the mode and mechanism of the Chinese herbal formula.

To sum up, neuroprotection of FBD or FBD-CO₂ on neuronal I/R injury originated partly from the synergies of its neuroprotective compounds atractylenolide II, III, levistolid A and ferulic acid. This helped to better understand its clinical application for ischemic cerebrovascular disorders.

Acknowledgements

The authors thank Yi Cao and Lan Luo for experimental assistance.

Decleration of interest: This work was supported by various grants from the National New Drug Foundation of China (969010538), from the National Natural Science Foundation of China (30271604 & 30500683), and from the Doctoral Innovative Foundation of Jiangsu Province of China (200593).