Introduction

Increase in average life expectancy over the past century have caused a shift in the main causes of illness and death, from communicable diseases (infectious and parasitic diseases) to age-related non-communicable diseases, including cardiovascular and neurodegenerative diseases, cancer, and type-2 diabetes. As far as neurodegenerative diseases are concerned, Alzheimer’s disease (AD) and other dementias (dementia with Lewy bodies, vascular dementia, and frontotemporal dementia), Parkinson’s disease (PD) and PD-related disorders, are the most common debilitating neurodegenerative pathologies. World Health Organization (WHO) estimates that about 36 million people suffer from dementia, a number which will double by 2030 and triple by 2050 (World Health Organization 2012). Over 10 million people around the world suffer from Parkinson’s disease, especially in developed countries, and an increase in the risk of Parkinson’s disease in people who live in low and medium income countries is expected in the upcoming years, further increasing the incidence of this pathology (Schapira 2005). Thus, considering that at present there are no proven therapies for these incurable pathologies, their prevention represents one of the major challenges of the twenty first century. Growing evidence suggests that the brain undergoes morphological (e.g., brain volume, fractional anisotropy) and neurochemical (e.g., decreased levels of cholinergic and dopaminergic neurotrasmitters) changes as part of the healthy aging process, and as a result of age-associated neuronal dysfunctions such as reduced dendritic branching, progressive loss of synapses, protein misfolding/aggregation, impaired calcium homeostasis and decreased brain derived neurotrophic factor (Dickstein et al. 2007; Erickson et al. 2010; Hajieva et al. 2009). These modifications, in turn, seem to be associated with oxidative stress and chronic inflammation (Johnson and Johnson 2015). In fact, the presence of toxic proteins induces the accumulation of reactive oxygen (ROS) and nitrogen (RNS) species, which increase oxidative stress leading to apoptosis and neuronal loss.

Recent findings have shown that age-associated brain changes occur at a much faster rate under pathological conditions, leading to the development of neurodegenerative disorders characterized by early brain damage due to the progressive dysfunction, degeneration and death of neurons (Skaper et al. 2017; Surmeier et al. 2017). As old age is the most important risk factor for neurodegenerative diseases, neuroprotective interventions aimed at delaying age-related brain injury could help delay the development of these pathologies.

Traditional and folk medicines recommend a number of plant extracts for neuroprotective purposes (Essa et al. 2012; Howes and Houghton 2012; Kim et al. 2010; Li et al. 2006; Zhang and Liu 2006). Unfortunately, there is little to no experimental or clinical data to support the use of these plant compounds. Plant extracts used as neuroprotective agents include certain Paeonia species (P. lactiflora Pall and P. veitchii Lynch), which have been studied for potential neuroprotective effects along with their major component, paeoniflorin. This review reports the available data on the neuroprotective effects of paeoniflorin. In addition, we will also discuss the sources, chemistry, bioavailability, traditional uses, and toxicity of this interesting bioactive compound.

Source

The roots of P. lactiflora Pall. (Ranunculaceae), a perennial herb which grows in China, India and Japan, commonly known as white peony, contain many bioactive components including terpenes (monoterpene and triterpene), hydrolysable tannins (galloyl-glucose derivatives), steroids, carboxylic acids (benzoic acid and gallic acid) and phenolic compounds (Li et al. 2009).

The root of the wild plant is called Radix Paeoniae Rubra (chishao), whereas once peeled and boiled in water it is known as Radix Paeoniae Alba (baishao), and these possess different pharmacological activities and clinical applications (Wang et al. 2008). As the main active compound in the plant, the Chinese Pharmacopoeia recommended paeoniflorin as a phytochemical marker in the evaluation of product quality (Wu et al. 2009). According to the Japanese and Chinese Pharmacopoeia, Radix Paeoniae must contain not less than 2.0% paeoniflorin (Commission 1992; Society of Japanese Pharmacopoeia 1996).

Structure

Paeoniflorin, the β-glucoside of paeoniflorigenin (C23H28O11), is a monoterpene glycoside with a cage-like pinane skeleton (Fig. 1) which is near-unique among natural products (Corey and Wu 1993; Kaneda et al. 1972; Kim and Ha 2009). It is a neutral compound with a molecular weight of 428.47, and it shows a good water solubility (log P = 2.88) (Liu et al. 2006a, b; Martey et al. 2013).

Fig. 1
figure 1

Chemical structure of paeoniflorin

Full size image

Bioavailability

According to a recent study, paeoniflorin has a low bioavailability, and this limits its clinical efficacy (Wu et al. 2009). This is a consequence of a number of concurring reasons, including poor absorption and permeation in the gastrointestinal tract, first-pass metabolism occurring in gut or liver, p-glycoprotein-mediated efflux, hydrolysis and decomposition by intestinal bacterial microflora (Chan et al. 2006; Wang et al. 2008). Paeoniflorin low permeability is also a result of its low partition coefficient and low lipophilicity (Chan et al. 2006). Co-administration with sinomenine leads to a significant increase in its bioavailability (Liu et al. 2005a) through inhibition of p-glycoprotein (Liu et al. 2006a). Efflux, mediated by p-glycoprotein at the apical membrane in Caco-2 cells, is responsible for pumping out paeoniflorin. So the use of a p-glycoprotein inhibitor, such as cyclosporine-A, increase the absorption of the compound. Much of the paeoniflorin is hydrolyzed by lactase phlorizin hydrolase (LPH), the brush border glucosidase. Inhibition of LPH using gluconolactone significantly decreases hydrolysis, leading to the lower absorption of paeoniflorin in the upper regions of the small intestine, where LPH usually exhibits more activity. The aglycone, paeoniflorgenin, has a permeability 48 times greater than that of the glucoside form. These findings suggest that the glucoside is substantially removed from upper regions of the small intestine, hydrolyzed and then absorbed as an aglycone (Liu et al. 2006a).

The bioavailability of paeoniflorin is higher when administered as an herbal extract component, rather than when administered in its pure form. Phytochemical studies revealed that other compounds, such as paeonoside, oxypaeoniflorin and suffruticoside, are present in the plant, and these may be degraded by intestinal bacteria. The presence of different derivatives utilizable by bacteria results in the higher bioavailability of paeoniflorin in herbal extract (Wu et al. 2009). Moreover antibacterial activity of gallotannins and a monoterpene glycoside of the herb have been demonstrated in a previous study, suggesting that these compounds may inhibit bacterial metabolization when present in extracts, and consequently enhance the paeoniflorin bioavailability (An et al. 2006; Kang et al. 2008). Co-administration of antibiotics with paeoniflorin increases the bioavailability of the compound through inhibition of intestinal bacteria (He et al. 2003).

The pure compound has been found in the plasma of rats within 5 min after oral administration and it could be detected in plasma over 360 min following oral administration of pure paeoniflorin dosed at 30 mg/kg. Plasma elimination of the compound is rapid, with a T1/2 of 1.19 ± 0.33 h. Poor absorption of the compound results in its low bioavailability, with intestinal bacteria degrading the unabsorbed portion (Takeda et al. 1997). Intravenous injection and oral administration of paeoniflorin with a plant decoction showed a logarithmic plasma concentration versus time dependence, indicating a phase of rapid absorption followed by a subsequent flat slope period of elimination. These results revealed that a higher amount of paeoniflorin was being absorbed by oral administration of Radix Paeoniae Rubra (wild grown root) (Wang et al. 2008).

Pharmacokinetic parameters of a formulation of frozen dry powder of paeoniflorin were evaluated following intravenous administration in rat (40 mg/kg). The T1/2 and AUC0-∞ were 0.739 ± 0.232 h and 43.75 ± 6.9 µg h mL−1, respectively (Cheng et al. 2006). However, these parameters were lower (T1/2: 2.91 ± 1.56 h and AUC0–∞: 1.38 ± 0.19 µg h mL−1) for oral administration of the compound at a dosage of 53.36 mg kg−1 (Wang et al. 2006), indicating that the compound may have low oral bioavailability. Therefore, intravenous administration of paeoniflorin may be a better therapy for achieving pharmacological effects. Furthermore, sulfonated monoterpenes available in sulfur-fumigated white peony root may have improved bioavailability and delayed absorption, as found in mice (Cheng et al. 2010). Paeoniflorin is widely metabolized to its aglycone–paeoniflorgenin– and to paeonimetabolins I and II, probably by intestinal microflora (Shu et al. 1987) (Fig. 2).

Fig. 2
figure 2

Chemical structures of paeoniflorin metabolites

Full size image

Taking into account all this information on the low bioavailability of paeoniflorin, its bioavailability could be improved through co-administration with other phytochemicals or compounds.

Traditional medicine use of Radix Paeoniae

The dried roots of P. lactiflora are used extensively in traditional folk medicine, especially Chinese Traditional Medicine, for treating a number of human ailments (World Health Organization 1999). In particular, Radix Paeoniae is used against menstrual disorders such as dysmenorrhoea, amenorrhoea, and menstrual pains (Commission 1992). In addition, Radix Paeoniae is indicated for the treatment of skin allergies (i.e., atopic eczema, urticaria and angioedema). It is worth noting that Radix Paeoniae has been used against certain types of dementia and to improve cognitive function (World Health Organisation 1989).

Toxicity of Radix Paeoniae

Radix Paeoniae is considered a traditional remedy with low toxicity. In fact, in a murine model system, the lethal dose, 50% (LD50) calculated after intraperitoneal injection is 230 mg/kg. Moreover, at a concentration of 4 mg/mL, the ethanolic extract does not show mutagenic activity in an Escherichia coli PQ37 genotoxicity assay (SOS chromotest) (Chang 1989; Xu et al. 2002). Nevertheless, considering that Radix Paeoniae is suggested as an abortion-inducing agent in traditional medicine, its use is strongly advised in pregnancy and as well as during lactation (European Medicines Agency 2016).

Neuroprotective effects of paeoniflorin

The discovery and study of natural bioactive compounds in herbal medicine with neuroprotective properties has raised considerable interest over recent decades for the isolation, identification and development of neuroprotective candidates for use in the treatment of several neurological disorders (such as apomorphine and rivastigmine) (Kumar and Khanum 2012; Lobo et al. 2010). The neuroprotective effects of paeoniflorin are summarized in Table 1. One of the earliest studies on the neuroprotective effects of paeoniflorin was published in 1994 by Ohta et al. (1994), which showed that daily doses (0.01 mg/kg) to old (25 months) and young (5 months) Fischer 344 rats improved the learning ability of the old group with no influence on the learning ability of the young group. The authors concluded that paeoniflorin would be useful for treatment of age-related dementia and cognitive dysfunction. Paeoniflorin has also shown influence in alleviating cerebral ischemia and inflammatory pain (Zhong et al. 2009; Zhang and Liu 2006; Tao et al. 2017) and in improving cognitive impairment caused by depression and Parkinson’s-like behavior in animal studies (Cao et al. 2010; Mao et al. 2010). Liu et al. (2005b) showed that paeoniflorin might be used as promising anti-stroke drug in transient ischemia model rats when subjected to a 1.5 h occlusion of the middle cerebral artery. In fact the administration of the compound (2.5 and 5 mg/kg, s.c.) have decreased in a dose-dependent manner both neurological impairment and the histologically-measured infarction volume, with a paeniflorin-induced neuroprotection related to the activation of adenosine A1 receptor (A1R). Similar results were also obtained in permanent ischemia model treating animals with 2.5, 5.0 and 10.0 mg/kg s.c. These results are interesting but need verification, since the applied model systems were not performed following the ARRIVE criteria and the results showed protection also in subcortical brain areas that are usually resistant to protective approaches.

Table 1 In vitro and in vivo neuroprotective effects of paeoniflorin
Full size table

While its role in neuroprotection has been reported in several works, the mechanisms by which it exerts these effects are yet to be fully clarified. Recently, Wang et al. (2013a) described the notable neuroprotective effects of this monoterpene against glutamate-induced cell damage on PC12 cells. In particular, at the highest tested concentration of 100 μM it was found to increase cell vitality, phosphorylation of protein kinase B (AKT) and of glycogen synthase kinase-3β, while decreasing the accumulation of reactive oxygen species, nuclear and mitochondrial apoptotic alteration and B cell lymphoma 2 (Bcl-2)/Bax ratio. Moreover, paeoniflorin, but not its isomer albiflorin (both present in the root), is responsible for intracellular Ca2+ overload and the expression of calcium/calmodulin protein kinase II. Paeoniflorin is also able to protect PC12 cells from 1-methyl-4-phenylpyridinium-induced damage and α-synuclein by induction of the autophagy pathway (Cao et al. 2010; Sun et al. 2011; Wang et al. 2013b). In particular, it reduces calcium influx and lactate dehydrogenase release, activating α-synuclein degradation and, finally, upregulating the mitogen activated protein kinase/extracellular signal–regulated kinase (MAPK/ERK) pathway and LC3-II protein. These effects lead to a decrease in the apoptotic pathway, an activation of acid-sensing ion channels, the maintained integrity of the mitochondrial membrane and the modulation of autophagic vacuoles.

Tsai et al. (2005) try also to highlight the mechanisms by which phloretin affects neuronal or neuroendocrine functions. They notice a dose-dependent effect of the monoterpene glycoside on inhibition of L-type Ca2+ channels in NG108-15 cells in a mechanism unlinked to the binding to adenosine receptors. Moreover at a concentration of 30 μM it slight decreases voltage-dependent Na+ current and delayed rectifier K+ current. So inhibition of these channels (although paeoniflorin is structurally different from prototypical inhibitors) can be one of ionic mechanisms utilized for functional activity of this molecule in neurons.

The protective effects of paeniflorin have been analyzed also on SH-SY5Y cell injury induced by Aβ25–35, a cell line widely used as a typical model for Alzheimer’s disease. The pretreatment of the cells with paeniflorin (2, 10, 50 μM) results in decreases of mitochondrial membrane potential, ROS production and increases of Bax/Bcl-2 ratio, cytochrome c release and activation of caspases (caspase-3 and caspase- 9), typical features of cytotoxicity due to Aβ25–35 (Wang et al. 2014).

In a recent report, Wu et al. (2013) suggested that paeoniflorin may have the potential to promote the survival of neural stem/progenitor cells under treatment with hydrogen peroxide. The onset of oxidative stress is a critical factor in avoiding the activation of apoptotic processes in these cells and paeoniflorin shows dose-dependent activity. A pretreatment with paeoniflorin (100, 200 and 400 μg/mL) was found to markedly decrease cell death caused by 200 μM hydrogen peroxide. Its action induces a decrease in procaspase-3, a balance of Bcl-2 and Bax expression and increases phosphatidylinositol 3 kinase/protein kinase B phosphorylation. This latter element is likely responsible for the protective activity. In fact, its selective inhibition with LY294002 eliminates the positive effects of paeoniflorin. In an in vivo study, paeoniflorin pretreatment (2.5 and 5 mg/kg for 11 days) and post-treatment (2.5 and 5 mg/kg once a day following administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine for 3 days) protects striatal nerve fibers and tyrosine hydroxylase (TH)-positive neurons in a rat model of PD with the activation of the adenosine A1 receptor (Liu et al. 2006b). Sub-chronic treatment (2.5, 5.0 and 10.0 mg/kg, subcutaneously, twice a day for 11 days) reduces unilateral striatal lesion due to 6-hydroxydopamine in a dose dependent-manner without influencing the activity of dopamine D1 receptors or dopamine D2 receptors (Liu et al. 2007).

Human and animal studies have also pointed out the involvement of neuroinflammation in the loss of neurons in several neurodegenerative diseases (Hunot and Hirsch 2003), such as Parkinson’s disease, which is characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta. One of the main elements that bring to the onset of brain inflammation is the activation of glia, particularly microglia (Liu and Hong 2003), through the release of proinflammatory and neurotoxic factors, such as reactive oxygen species, tumor necrosis factor-alpha and interleukin 1 beta, eicosanoids, reactive nitrogen species and excitatory amino acids (Liu and Hong 2003). Liu et al. (2006b) reported that subcutaneous administration of paeoniflorin (2.5 and 5.0 mg/kg) for 11 days could reduce the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced toxicity in a mouse model of Parkinson’s disease by inhibition of neuroinflammation, through activation of the adenosine A1 receptor. This monoterpene glycoside exhibits also multiple effects on experimental autoimmune encephalomyelitis (Zhang et al. 2017). The treatment of C57BL/6 mice with 5.0 mg/kg/d results in a decrease of onset and clinical symptoms of experimental autoimmune encephalomyelitis, with a concomitant decrease of Th17 cells infiltrated in the central nervous system and in the spleen. This latter process can be a consequence of the inhibition of IKK/NF-κB and JNK signaling pathways, which results in a suppression of the expression of co-stimulatory molecules and the production of interlukin-6 by dendritic cells. In a recent work (Li et al. 2017), pretreatment with paeoniflorin (20 mg/kg or 40 mg/kg for 4-week) reverses the depressive-like behavior and the abnormal inflammatory cytokine levels in the serum, medial prefrontal cortex, ventral hippocampus and amygdala in mice treated with interferon-alpha at 15 × 106 IU/kg. The neuroprotective influence of paeoniflorin on cell and animal model systems are summarized in Fig. 3.

Fig. 3
figure 3

Neuroprotective effects of paeoniflorin on cell and animals model systems

Full size image

Conclusion and future prospects

Many food and medicinal plant substances have been classified as beneficial supplements in the maintenance of human health. Paeoniflorin is a monoterpene glycoside occurring in Radix Paeoniae which is used as a low-toxicity traditional remedy against many human ailments. Chinese Traditional Medicine prescribes Radix Paeoniae also as a neuroprotective agent. Recent in vitro studies confirmed that paeoniflorin shows its neuroprotective effects mainly through anti-inflammatory and anti-apoptotic properties and by its potential to promote neuronal survival.

Therefore, due to its low toxicity and health-promoting effects on neurons, paeoniflorin seems to be a promising compound for neurodegenerative disorder prevention and treatment. One of the major challenge derives from its low bioavailability, as well as from the need to clarify whether the reported protective effects in vivo actually result from direct or indirect effects on the brain. However, further studies are required to reach a definitive recommendation on the use and beneficial effects of paeoniflorin in human healthcare. In particular, results published so far encourage the design of further studies on paeoniflorin, focused on the evaluation of the most effective doses, on clinical trials and on the molecular mechanisms underlying its neuroprotective effects to show its clinical efficacy or to point to the opposite direction.