Abstract

Ethnopharmacological relevance: Mercury sulfides (HgS) are frequently included in Ayurveda, Tibetan and Chinese medicines to assist the presumed therapeutic effects,but the thnopharmacology remains elusive. The present study examined the protective effects of α-HgS-containing Hua-Feng-Dan and β-HgS-containing 70 Wei- Zhen-Zhu-Wan (70W, Rannasangpei) against Parkinson’s disease mice induced by lipopolysaccharide (LPS) plus 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP).

Method: A single injection of LPS (5 mg/kg. ip) was given to adult male C57BL/6 mice,and 150 days later, the low dose of MPTP (15 mg/kg, ip, for 4 days) was given to produce the “two-hit” Parkinson’s disease model. Together with MPTP treatment, mice were fed with clinically-relevant doses of Hua-Feng-Dan (0.6 g/kg) and 70W (0.2 g/kg) for 35 days. Rotarod test was performed to examine muscle coordination capability. At the end of the experiment, brain was transcardially perfused with paraformaldehyde, the substantia nigra was sectioned for microglia (Iba1 staining) and dopaminergic neuron (THir staining) determination. Colon bacterial DNA was extracted and subjected to qPCR analysis with 16S rRNA probes.

Results: The low-grade, chronic neuroinflammation produced by LPS aggravated MPTP neurotoxicity, as evidenced by decreased motor activity, intensified microglia activation and loss of dopaminergic neurons. Both Hua-Feng-Dan and 70W increased rotarod activity and ameliorated the pathological lesions in the brain. In gut microbiomes examined, LPS plus MPTP increased Verrucomicrobiaceae, Methanobacteriaceae,Pronicromonosporaceae, and Clostridaceae species were attenuated by Hua-Feng-Dan and 70W.

Conclusions: α-HgS-containing Hua-Feng-Dan and β-HgS-containing 70W at clinical doses protected against chronic LPS plus MPTP-induced toxicity to the brain and gut,suggesting HgS-containing traditional medicines could target gut microbiota as a mechanism of their therapeutic effects.

Keywords: Mercury sufides (HgS); Hua-Feng-Dan; 70 Wei-Zhen-Zhu-Wan (70W,Rannasangpei); LPS; MPTP; Neuroinflammation; Dopaminergic neuron; gut microbiome

1. Introduction

Minerals have been included in traditional remedies since ancient times. In Pharmacopeia of Ayurveda, 15 kinds of minerals are listed in 8% recipes (Joshi et al., 2017). In Pharmacopeia of China, over 10 kinds of minerals are included in 7% of the recipes (Pharmacopeia Committee, 2015). Among the minerals used in traditional remedies, mercury sulfides (HgS) are of utmost importance. Mercury (Hg) is a toxic substance, and the therapeutic effects, the risk assessment, and pharmacology of HgS-containing medicines are important issues that still need to be addressed (Liu et al.,2018; Liu et al., 2019).

Among mercury compounds only HgS is used in oral medication at appropriate dosages,in sharp contrast to other inorganic mercurials (i.e., HgCl2) and organic mercurial compounds (i.e., MeHg) (Huang et al., 2013; Kamath et al., 2014; Liu et al., 2018; Zhou et al., 2009). Αlpha-HgS (also called cinnabar, Zhusha) is included in over 40 Chinese medicine formulas (Pharmacopeia Committee, 2015), mainly used for brain diseases.Hua-Feng-Dan (also called WSHFD) is such an example, the recipe contains α-HgS (~10%) and realgar (~10%), alone with herbs Arisaema erubescens, Nepeta cataria, Bombyx batryticatus, Buthus martensii Karsch, Aconitum oreanum, Atractylodes japonica and others (Chen et al., 2020; Zhang et al., 2012). Beta-HgS, mainly in the form of Zuotai (a mineral mixture with 54% β-HgS) (Li et al., 2018), is included in many Tibetan medicines. 70W Zhen-Zhu Wan (70W, also called Rannasangpei, Padma-28) is such an example (Du et al., 2013; Schwabl and Vennos, 2015). 70W is composed of 70 herbs and minerals, such as pearl, Hong-sik, Dalbergia odorifera, Nine stone,Saffron, Bezoar, Musk and Zuotai, and is listed in the 2015 edition of Pharmacopoeia of China (Pharmacopeia Committee, 2015). 70W is effective clinically for cerebrovascular diseases (Wan et al., 2009) and experimentally against vascular dementia in rats (Wu et al., et al., 2016a) and CCl4-induced liver injury in mice (Nie et al., 2017).

The brain-gut-microbiota axis is now recognized to play an important role in brain diseases (Martin et al., 2018; Perez-Pardo et al., 2018; Spielman et al., 2018). The absorption of HgS (0.2%) in the gastrointestinal tract is quite low compared to HgCl2 (15%) and MeHg (95%) (Liu et al., 2008). We hypothesize that HgS could affect gut microbiota to exert beneficial effects. Indeed, oral α-HgS and Zuotai are quite different from HgCl2 and MeHg in producing gut microbiota alterations (Zhang et al., 2019), and α-HgS-containing HFD is effective in ameliorating LPS plus rotenone-induced gut microbiota disturbance (Chen et al., 2020). Thus, to examine gut microbiota might provide a clue to pharmacological basis of HgS used in traditional medicines.

The low-grade inflammation produced by a single injection of lipopolysaccharides (LPS) has been used to induce progressive neurodegeneration in the brain along with motor dysfunction that recapitulate neuropathological and symptomatic features of Parkinson’s disease (PD) patients (Qin et al., 2013). A single LPS-induced low-grade neuroinflammation could potentiate neurotoxicity of rotenone (Chen et al., 2020; Huang et al., 2017; Li et al., 2019), but also potentiate neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) (Byler et al., 2009; Gao et al., 2003). We have recently demonstrated that α-HgS-containing Hua-Feng-Dan is effective in protecting against LPS plus rotenone-induced dopaminergic neuron loss and activation of microglia (Chen et al., 2020). The goal of the present study is to use chronic LPS plus MPTP-induced PD mice to examine the neuroprotective effect of α-HgS-containing Hua- Feng-Dan, and β-HgS-containing 70W, focusing on gut microbiota as a mechanism of neuroprotection.

2. Materials and Methods
2.1 Animals

Adult male C57BL/6J mice weighing 20-22 g were obtained from the Experimental Animal Center of the Third Military Medical University (Chongqing, China; Certificate No. SCXK 2012–0011). Mice were kept in the accredited animal facilities at Zunyi Medical University with 21 ± 2 。C, light on from 8:00 to 20:00, had free access to standard feed and drinking water. The experimental protocol followed Chinese Guidelines of Animal Care and Welfare and was approved by the Animal Care and Use Committee of Zunyi Medical University (2015-07) (Chen et al., 2020)

2.2 Reagents

Lipopolysaccharide (Escherichia coli serotype 0111:B4; LPS, L4130) and 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine (MPTP, M0896) were purchased from Sigma (St.Louis MO). Hua-Feng Dan was provided by Hua-Feng Dan Pharmaceuticals as described previously (Zhang et al., 2012; Chenet al., 2020). 70 Wei Zhen-Zhu-Wan (70W, also called Rannasangpei) was provided by Tibetan Medicine Manufacture Factory as described previously (Nie et al., 2017), based on the 2015 edition of Pharmacopoeia of China (Lot number Z20110561). Polyclonal mouse anti-Tyrosine
Hydroxylase (TH) antibody (MAB318) was purchased from MABEMD Milipore (Temecula, CA); Polyclonal rabbit anti Iba-1 antibody (019-19741) was purchased from Wako (Richmond, VA). TIANamp Stool DNA Kit was purchased from Tiangen Biotechnology Company (Beijing, China). All other chemicals are of reagent grade.

2.3 Drug Treatment

Mice were given a single injection of LPS (5 mg/kg, ip), based on LPS-induced chronic neuro-inflammation in C57BL/6 mice (Qin et al., 2013). Five months later, mice received the threshold dose of MPTP (15 mg/kg, ip) for 4 consecutive days. Starting from MPTP treatment, mice were given standard rodent chow or specifically prepared feed (smashed the rodent chow to a powder, then mixed with Hua-Feng-Dan or 70W, and 60。C overnight to dry) (Chen et al., 2020). The clinical doses of 70W and Hua-Feng-Dan were used (Chen et al., 2020; Du et al., 2013; Nie et al., 2017). After approximately 10-fold conversion to clinical dose, the dose of Hua-Feng-Dan was 0.6 g/kg, and the dose of 70W was 0.2 g/kg. Mice without receiving LPS served as controls and MPTP alone controls, respectively. Animals were fed with drug-containing diet for 35 days and the body weights were recorded weekly.

2.4 Rotarod measurement

Rotarod test measures motor coordination in mice in an accelerating rotarod (Chengdu Technology & Market, ZB-200, Chengdu, China). Mice were trained for three days before experiments. The initial speed was 10 rpm/min, and increased 5 rpm per min.The maximum time was 300 s per trial. The time at which the mice fall off the rod is recorded (Li et al., 2019). If the mice remained on the rod for more than 300 s, the stay time was recorded as 300 s. From 14 days after MPTP administration, mice were subjected to Rotarod test weekly, and the last two trials on day 31 and day 35 were averaged for calculation.

2.5 Immunohistochemistry and cell count in the SN of C57BL/6 mice

Mice (n=3) were euthanized using 6% choral hydrate and transcardially perfused with phosphate buffered solution (PBS), followed by perfusion Ellman’s Reag with 4% paraformaldehyde (PFA). Brains were further post-fixed in 4% PFA for 24 h, then transferred to ice-cold 10% sucrose solution and incubated at 4 °C overnight. The 10% sucrose was then replaced with ice-cold 30% sucrose, and brains were equilibrated completely at 4 °C, as evidenced by their sinking to the bottom of the container. Brains were then sectioned immediately on a freezing sliding microtome and processed for immune-staining as described previously (Song et al., 2019). The free-floating sections were immune-blocked with 4% goat serum in 0.25% triton/PBS for 2 hours and then incubated with Iba-1 (1:2000) antibody overnight at 4°C. On the second day, the sections were washed by 1% bovine serum albumin (BSA) in 0.25% triton/PBS before the incubation with anti- TH antibody (1:2000) overnight at 4 °C. The double-labeled immunofluorescence pictures were taken under the confocal microscope by using Alexa-488 (green) and Alexa-594 (red) conjugated secondary antibodies (1:1,000) to visualize the TH immune-reactive (THir) or Iba-1-positive (Iba1+) cells. The number of THir and Iba1+ cells was counted as described previously (Li et al., 2019). Atotal of 8 sections were sampled at 105 µm intervals for the substantia nigra region. Samples were counted in a double-blind manner. Data were expressed as percentage to saline-injected controls (Song et al., 2019).

2.6 Bacterial DNA isolation

Colon contents (150-200 mg) were collected in the lysis buffer and incubated for 10 min at 80 。C to disrupt bacteria and centrifuged. The inhibit EX was used to absorb other materials in the supernatant, and protein kinase K was used to remove residual proteins, followed by DNA column purification. The resulted purified DNA had 260/280 ratio >1.9, and diluted to 20 ng/µL for qPCR analysis (Chenet al., 2020; Zhang et al.,2019).

2.7 Real-time qPCR

The CFX96 Touch™ Real-time PCR system (Bio-Rad Laboratories, Hercules, CA) was used for qPCR analysis as described previously (Chenet al., 2020; Zhang et al., 2019).The bacterial primers are listed in Supplemental Table 1. The Ct values of 16S rRNA expression were calculated by the 2−ΔΔCt method and normalized to the total bacteria of the same sample and expressed as percentage of Controls.

2.8 Statistical Analysis

The data are expressed as the mean ± SEM and statistical significance was assessed with One-way ANOVA followed by Bonferroni’s t test with the Statistical Package for Social Sciences (SPSS) software (version 17.0). The criteria of significance were set at p < 0.05.

Results

3.1. Animal body weight and general health

On hundred fifty days after LPS injection, mice were treated with MPTP (15 mg/kg, ip for 4 consecutive days), with and without Hua-Feng-Dan (HFD) or 70W in the feed. Figure 1 showed animal weekly body weights when MPTP was given. Mice in LPS+MPTP group had two mice dead (2/11, 18% mortality), and all the mice in other groups survived and were healthy. In general, no apparent abnormalities in animal activities were observed except for slightly decreased activities in LPS+MPTP group.

Figure 1. Animal body weights. Mice were given a single injection of LPS (5 mg/kg, ip), and 150 days later challenged with MPTP (15 mg/kg, ip for 4 days). The clinical doses of Hua-Feng-Dan (0.6 g/kg) and 70W (0.2 g/kg) were given via feed for 35 days starting from MPTP administration.

3.2. Rotarod activity

The behavioral activities were examined weekly after 14 days of MPTP injection, and the Rotarod results on d31 and d35 were averaged to make assessments. As shown in Figure 2, LPS alone, MPTP alone, and LPS+MPTP decreased the rotarod activity compared with Control group, but there was no difference among these three groups.Compared to LPS+MPTP group, Hua-Feng-Dan (HFD) and 70W significantly increased the time on the rod by 15% and 18%, respectively. The spontaneous activities were also measured; however, no significant differences were found among groups (data not shown).

Figure 2. Behavioral test. Rotarod tests were performed on day 31 and day 35 of the treatments. Data are mean ± SE of 9-10 mice. #Significantly different of LPS, MPTP,and LPS+MPTP groups from Controls, p < 0.05; *Significantly different of 70W and Hua-Feng-Dan groups from LPS+MPTP, p < 0.05.

3.3 THir neuron loss with increased neuroinflammation

Loss of dopaminergic neurons in the Substantia nigra is a main characteristic in PD patients and PD animal models (Li et al., 2019), which is associated with activation of microglia (Chen et al., 2020). As illustrated in Figure 3, 150 days after a single injection of LPS and 4 injections of MPTP at low doses, the loss of dopaminergic neurons (THir cells, red) was evident (Fig. 3A). The numbers of THir cells were reduced by 34%, 18% and 38% after LPS, MPTP, and LPS+MPTP, respectively (Fig. 3B). 70W (94% of control) and Hua-Feng-Dan (HFD, 90% of control) restored neuron loss as compared to LPS+MPTP group. Activated microglia had hypertrophied, amoebic morphology and intensified Iba1+ staining (green). The numbers of Iba1+ cells were increased 35-50% after LPS, MPTP, and LPS+MPTP, while Hua-Feng-Dan and 70W reduced such increases.

Figure 3. Neurotoxicity and microglia activation. A. Representative images of immunohistochemistry; B. THir and Iba1+ cell counts. Data are mean ± SE of 9-10 mice. #Significantly different of LPS, MPTP, and LPS+MPTP groups from controls, p < 0.05;*Significantly different of 70W and Hua-Feng-Dan groups from LPS+MPTP, p < 0.05.

3.4. Alteration of gut microbiome

Colon bacteria DNA was analyzed via real-time PCR using 16S rRNA probes. Figure 4 and 5 show the results 16 genus/species from 6 phyla and 13 families in a style of Phylum_Family_Probe.
Figure 4 shows Verrucomicrobia, Bacteroidetes, and Euryarchaeota phyla. The expression of Verrucomicrobia_Verrucomicrobaceae (probe AJ400275 andAY271254) was markedly increased 15-30 fold by LPS, MPTP, and LPS+MPTP, which was prevented following the treatment of Hua-Feng-Dan (HFD) and 70W, similar to Hua- Feng-Dan protection on LPS plus rotenone-treated rats (Chen et al., 2020). However,another probe (HM453208) did not show the difference; Bacteroidetes_Prevotellaceae (probe EU728784) was increased 2-fold by LPS+MPTP, but further increased by Hua-
Feng-Dan (5-fold) and 70W (2.8-fold). However, in another family Bacteroidetes_Porphymonadaceae (probe provetella), no difference was evident among groups; Bacteroidetes_Bacteroidaceae (probe KV751184) was increased markedly 10- fold by MPTP, and 2.3-fold by LPS, LPS+MPTP. Hua-Feng-Dan and 70W ameliorated such increases. Bacteroidetes_Porphymonadaceae_Parabacteroides sp. (probe KP06661) was increased 8-10 fold by all treatment.

Euryarchaeota_Methanobacteriaceae (probe NZ_CCXU01000080) was increased 2.5- fold by LPS, MPTP, and LPS+MPTP, but such increase was prevented by Hua-Feng-Dan and 70W.Figure 4. Gut microbiome analysis of phylum Verrucomicrobia, Bacteroidetes, and Euryarchaeota with selected 8 families. Data are mean ± SE of 9-10 mice. #Significantly different of LPS, MPTP, and LPS+MPTP groups from Controls, p < 0.05; *Significantly different of 70W and Hua-Feng-Dan groups from LPS+MPTP, p < 0.05.Figure 5 showed phylum Actinobacteria, Firmutes, and Proteobacteria.

Actinobacteria_Actinobacterias (probe Act920F3) was mildly increased 2-fold in all groups; Actinobacteria_pronicromonosporaceae (probe KP076663) was increased 3.4-fold by MPTP and LPS+MPTP, which was decreased by Hua-Feng-Dan (HFD), and ameliorated by 70W; Firmutes_Clostridaceae (probe KU321269) was increased 1.7-fold by MPTP and LPS+MPTP, which was decreased by 70W and ameliorated by Hua-Feng-Dan; Firmutes_Lactobacillaceae (probe HM079525) was increased by LPS,MPTP, and LPS+MPTP for 10-fold, which was ameliorated by 70W (4.6-fold); Another
Firmutes_Lactobacillaceae (probe KX247777) was also increased by LPS+MPTP by 3.8-fold, which was ameliorated by Hua-Feng-Dan, similar to Hua-Feng-Dan protection on LPS plus rotenone-treated rats (Chen et al., 2020). Proteobacteria_Enterobactaceae (probe Eco1457F) was increased 2.5 to 3-fold by LPS, MPTP, and LPS+MPTP, 70W further increased it to 5.7-fold, but Hua-Feng-Dan markedly increased it 50-fold;However, another Proteobacteria_Enterobactaceae (probe DQ836271) was increased 3-fold by LPS+MPTP but was decreased by Hua-Feng-Dan.

Figure 5. Gut microbiome analysis of phylum Actinobacteria, Firmutes, and Proteobacteria with selected 8 families. Data are mean ± SE of 9-10 mice. #Significantly different of LPS, MPTP, and LPS+MPTP groups from Controls. *Significantly different of 70W and Hua-Feng-Dan groups from LPS+MPTP, p < 0.05.

Discussion

The present study clearly demonstrated the beneficial effects of HgS-containing HFD and 70W in protecting against LPS+MPTP-induced neurodegeneration, as evidenced by increased rotarod activity, decreased microglia activation, and reduced dopaminergic neuron loss. Hua-Feng-Dan and 70W were also effective in restoring LPS+MPTP-induced gut microbiota alterations, such as LPS+MPTP-increased expression of Verrucomicrobiaceae, Methanobacteriaceae, Pronicromonosporaceae, and Clostridaceae. These results shed light on the ethnopharmacological basis of HgS-
containing traditional medicine Hua-Feng-Dan and 70W.Chronic neuroinflammation and neuron death form a vicious cycle in neurodegenerative diseases (Qin et al., 2013). The low-grade inflammation induced by LPS, even after several months, could aggravate neurotoxic effects of MPTP (Byler et al., 2009),rotenone (Chen et al., 2020; Huang et al., 2017; Li et al., 2019) and norepinephrine (NE)-depleting toxin DSP-4. Microglia activation with increased proinflammatory factors plays key roles in the process. Figure 3 demonstrated that the beneficial effects of Hua- Feng-Dan appear to be due, at least in part, to suppress microglia activation and rescue dopaminergic neuron loss, similar Youth psychopathology to the Hua-Feng-Dan protection against LPS plus
rotenone-induced neurodegeneration in PD rats (Chen et al., 2020) The Microbiota-gut-brain axis in neurodegenerative disorders is increasingly recognized.MPTP (Lai et al., 2018; Perez-Pardo et al., 2018; Sun et al., Environment remediation 2018) and rotenone (Johnson et al., 2018; Yang et al., 2017) are reported to alter gut microbiota. Bacterial DNA sequencing is the major technique to study microbiome, as shown by the differential effects of mercury compounds on gut microbiota (Zhang et al., 2019).However, qPCR with specific primers for 16S bacterial RNA has also been applied to examine the effect of rotenone on the microbiome (Chen et al., 2020). These results are briefly discussed below.

Increased Verrucomicrobiaceae is frequently seen in PD patients worldwide (Boertien et al., 2019). In the present studies, LPS, MPTP, LPS+MPTP increased Verrucomicrobiaceae 15-30 fold with two PCR probes, which was decreased by Hua- Feng-Dan and 70W (Fig. 4), consistent with Hua-Feng-Dan reduction of the increased Verrucomicrobia in LPS plus rotenone-induced PD rats (Chen et al., 2020).Methanogens are anaerobic prokaryotes that utilize hydrogen to reduce a variety of methyl compounds into methane (Chaudhary et al., 2018). In the current study, a probe for Methanobacteriaceae detected 2.5 fold increases by LPS, MPTP, and LPS+MPTP, which was ameliorated by Hua-Feng-Dan and 70W. Pronicromonosporaceae belongs to the phylum of Actinobacteria. MPTP was reported to increase Actinobacteria phylum 4- fold (Sun et al., 2018). In the current study a probe for Pronicromonosporaceae showed a 3.4-fold increase by MPTP and LPS+MPTP, which was decreased by Hua-Feng-Dan and 70W. Firmicutes/Bacteroidetes ratio is often used to assess the bacterial composition changes. In the rotenone-induced PD model, increases in Firmicutes and decreases in Bacteroidetes were observed (Yang et al., 2017). In the present study,three of Firmicutes probes (Fig. 5, Clostridaceae and Lactobacillaceae) showed increases after LPS and MPTP, which was reduced by Hua-Feng-Dan and 70W. Thus, some alterations of gut microbiome are in the line with the literature, and coincide with the neuroprotective effects of Hua-Feng-Dan and 70W .

Different probes could produce different results. In the present study, two probes for Verrucomicrobiaceae showed increases by LPS and MPTP, while the third probe showed no change (Fig. 4). This is the case for Proteobacteria_Enterobacteriaceae one probe (Eco1457) showed a 5-fold increase after 70W and a 50-fold increase after HFD.However, when using another probe (DQ836271), different results was obtained (Fig. 5). Interestingly, Proteobacteria_Enterobacteriaceae (Probe Eco1457 and U02524) showed marked increases when 70W was given to normal mice (Zhang et al., 2020).Enterobacteriaceae was decreased 50% by LPS plus rotenone, and HgS-containing Hua-Feng-Dan restored the expression (Chen et al., 2020). Thus, caution should be
taken in interpreting microbiome alterations.

We have performed toxicity comparisons of α-HgS (Lu et al., 2011; Shi et al., 2011) and β-HgS (Liu et al., 2016; Wu et al., 2016b) with HgCl2 and MeHg, and found that their acute and chronic toxicity are substantially less than environmental mercury compounds, and total mercury content alone are insufficient for risk assessment (Liu et al., 2018). It should also be noted that mercury sulfide is an important ingredient in Hua-Feng-Dan.Hua-Feng-Dan recipes with reduced or removed HgS showed reduced neuroprotective effects both in neuron/microglia co-cultures (Zhang et al., 2012), and in LPS plus rotenone PD rats (Chen et al., 2020). In polyherb-metallic preparations, HgS is presumed to assist the therapeutic effects of other ingredients (Liu et al., 2018), and the traditional recipe interactions could also reduce Hg toxicity (Xia et al., 2018).Chemical forms of mercury are the major determinant of their disposition, efficacy and toxicity (Liu et al., 2018; Liu et al., 2019). It should be mentioned that patients taking HgS-containing traditional medicines, mercury toxicity to the liver and kidney was mild and reversible, with normal neurological and cardiovascular functions (Du et al., 2013; Li et al., 2014; Sallon et al., 2017; Vickers et al., 2001). The therapeutic effects and clinical toxicity in humans are of utmost importance to evaluate HgS-containing traditional medicines.In summary, the present study showed that (1) low-grade inflammation induced by chronic LPS could potentiate MPTP neurotoxic effects and gut microbiota disturbance; (2) α-HgS-containing Hua-Feng-Dan and β-HgS-containing 70W protects against LPS plus MPTP-induced neurotoxicity, and (3) their therapeutic effects may be partially due to modulation of gut microbiome.