A Classic Herbal Formula Guizhi Fuling Wan for Menopausal Hot Flushes: From Experimental Findings to Clinical Applications

A classic herbal formula Guizhi Fuling Wan (GFW) has been used for managing menopausal hot flushes (MHFs), but the evidence across different study types has not been systematically summarized. This project investigated the clinical effects, phytochemistry, pharmacodynamics, and potential mechanisms of actions of GFW on the causative target proteins potentially driving MHFs. Twenty English and Chinese databases were searched for relevant clinical and experimental studies. A total of 12,988 studies were identified, of which 46 were included. Seven clinical studies demonstrated GFW had no statistically significant changes in the frequency and severity of MHFs; however, it could improve peripheral blood flow in the fingertips, jaw, and toes. Thirty-five studies on phytochemistry identified 169 chemical compounds of GFW. Four experimental studies revealed GFW’s therapeutic effects (e.g., normalize calcitonin gene-related peptide (CGRP) level) and potential target protein/cytokine (estrogen receptor beta (ESR2) with genetic variation, CGRP receptor, and interleukin-8) on MHFs. Therapeutic effects across different study types were inconsistent, possibly due to the dose difference and genotype variety of ESR2 in the human population. Further clinical and experimental studies, as well as biochemical investigation on the mechanisms of actions of GFW, are recommended.


Introduction
A menopausal hot flush (MHF) is a sudden and transient onset of erythema and warmth or burning sensation on the face and skin of the neck which appears during the transition to (pre-menopausal and peri-menopausal), and through, menopause (menopausal and post-menopausal) [1,2]. The onset commonly lasts for seconds to five minutes, but its persistence duration is unpredictable [3]. MHFs significantly impact the quality of life of both the sufferers and their families [3]. Its mechanism has not been fully investigated. The sudden reduction in estrogen level, the narrowed central thermoneutral zone, changes of the certain neurotransmitter synthesis (e.g., noradrenaline and serotonin), and peripheral vascular reactivity are linked to MHF onset [4][5][6][7][8][9]. Menopausal hormone therapy is the most popular and effective MHF management [10,11]. However, adverse effects, such as nausea, dizziness, dry mouth, and contraindication, with hormone-dependent diseases (e.g., breast cancer) exist [12]. Thus, more than half of all middle-aged menopausal women have taken non-hormonal therapies [13], such as Chinese herbal medicine (CHM).
Guizhi Fuling Wan (GFW, also known as Keishibukuryogan) is a classic Chinese herbal formula that has been used for MHFs in modern clinical practice [14,15]. In ancient China, there was no specific A total of 12988 records were identified, and seven of them met the inclusion criteria. Figure 1 illustrates the selection process of the included studies. Two RCTs [14,15], one parallel controlled study [27], and two case series [28,29] were included in this review. Another two studies [22,23] were claimed as controlled trials; however, two intervention groups were incomparable. Thus, the treatment groups of these two studies were considered as case series studies, and only data from GFW groups were extracted for analyses in this review. The characteristics of the included studies are summarized in Table 1.
Biomedicines 2019, 7, x FOR PEER REVIEW 3 of 30 A total of 12988 records were identified, and seven of them met the inclusion criteria. Figure 1 illustrates the selection process of the included studies. Two RCTs [14,15], one parallel controlled study [27], and two case series [28,29] were included in this review. Another two studies [22,23] were claimed as controlled trials; however, two intervention groups were incomparable. Thus, the treatment groups of these two studies were considered as case series studies, and only data from GFW groups were extracted for analyses in this review. The characteristics of the included studies are summarized in Table 1.    [24], quality of parallel controlled studies was evaluated by ROBINS-I ("Risk Of Bias In Non-randomised Studies -of Interventions") [25], and case series were assessed according to the Instrument for Evaluating the Quality of case series in Chinese herbal medicine [26].

RCTs
One RCT [15] reported that both low dosage (7.5 g/day) and high dosage (12.5 g/day) GFW could significantly reduce MHF frequency and severity when comparing between before and after the 12-week intervention. When compared to placebo, there was no statistical significance. However, the reduction may be dose-dependent, as the results from the high dosage group demonstrated a greater reduction, which provided a direction for future research on the relationship between GFW dosage and clinical effects. Another RCT [14] reported that the peripheral blood flow in postmenopausal females with MHFs in the GFW group (n = 67) was significantly decreased compared to the menopausal hormone therapy group (n = 64) under the jaw (mean difference (MD) -3.56, 95% confidence interval (CI) -5.17 to -1.95) and in the middle fingertip (MD -7.10, 95% CI -10.99 to -3.21) at the end of one-month treatment. Opposite effects on the blood flow in the third toe were observed in two groups: GFW increased the blood flow in the toe, whereas menopausal hormone therapy caused a decrease.

Parallel Controlled Study
The parallel controlled study [27] reported positive results, concluding that the severity of MHFs in 73.7% of the participants improved from severe to mild or moderate after six-month GFW intake. It was significantly higher than the control group. No significant difference in estradiol (E 2 ) and follicle-stimulating hormone (FSH) levels was observed between groups. Results of serum cytokine level indicated that serum monocyte chemotactic protein-1 (MCP-1) level in women treated with keishibukuryogan decreased significantly (-16.3%) compared to the no treatment group (3.8%). Further analysis of the GFW responder group (n = 28) indicated that concentrations of serum interleukin (IL)-8 and macrophage inflammatory protein (MIP)-1β were reduced significantly, whereas those in the non-responders were increased dramatically. Sub analysis results on menopausal transition status showed that GFW decreased the IL-8 and MIP-1β levels in perimenopausal GFW responders and MCP-1 in postmenopausal GFW responders.

Case Series
One case series study [29] revealed that the improvement of MHF severity was related to genetic variation of the polymorphic dinucleotide (CA) repeat of the estrogen receptor beta (ESR2) gene on chromosome 14, which include three types: two short alleles (SS), two long alleles (LL), and a short and long allele (SL) [30,31]. MHF severity in participants with LL genotype improved significantly after the intervention. FSH level changes also depended on genotype: FSH levels decreased in participants with SS genotype but increased in participants with SL or LL genotypes. E 2 level decreased in all genotype groups. Another case series [28] reported that the plasma calcitonin gene-related peptide (CGRP) level in eight post-menopausal participants significantly decreased after the four-week GFW intervention when compared to baseline data (MD -2.88, 95% Cl -4.07 to -1.69). CGRP is a potent vasodilator neuropeptide, and it is hypothesized to be a specific vasodilation neuropeptide of MHFs, which significantly rose in concentration during MHF onset [32]. However, GFW's effects on changes of frequency before and after intervention in three case series had no statistically significant difference (standardized mean difference (SMD) -1.15, 95% CI -2.49 to 0.20) [22,23,28].

Experimental Studies on Phytochemistry and Pharmacodynamics of GFW on MHFs
We searched the 20 English and Chinese databases, detailed in Section 2, to identify experimental studies that have investigated the chemical compounds of GFW. Studies were excluded if they did not provide the following details: (1) plants with valid plant specimen voucher numbers, ingredients and ratios for raw herbs (decoction), or (2) company names and batch/lot numbers for purchased patent products (pills, granules, or capsules).
For pharmacodynamics, the same set of 20 databases was searched to identify experimental studies that have investigated GFW on MHFs. To be eligible to be included, the studies must provide the details as specified in (1) and (2) as above. Also, experiments were required to have (3) utilized cell or animal models related to MHFs; (4) compared GFW's effects with the blank model control group (i.e., no treatment); (5) evaluated the pharmacodynamic effects on MHFs. Studies involving any co-interventions were excluded.
The structures of chemical compounds were drawn using BIOVIA Draw 2019 [33] or Avogadro [34]/Discovery Studio Visualizer 2019 [35]. Molecular weight was calculated by Lenntech Molecular Weight Calculator (https://www.lenntech.com/calculators/molecular/molecular-weightcalculator.htm). The quality of in vivo studies was assessed based on 20 domains specified in the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines [36].
As a result, 35 studies on phytochemistry, and four studies on pharmacodynamics met the inclusion criteria. Figure 1 illustrates the selection process of the included studies.

Pharmacodynamic Effects of GFW on MHFs
Characteristics and results of included in vivo studies are summarized in Table 3. The results from the quality assessment of included in vivo experimental studies are presented in Figure 3. The results suggested that GFW had a similar action to E2 on managing skin temperature by restoring the plasma level of CGRP: it increased the ovariectomy-induced CGRP reduction [66,67] and reduced CGRP-induced elevation of skin temperature in GnRH (gonadotropin-releasing hormone) analogtreated rats [67,68]. A dose-dependent skin temperature change was detected, showing that a significant inhibition effect on elevated skin temperature was observed at a dose of 1000 mg/kg [67,68]. However, GFW had no significant effects on CGRP concentrations and CGRP mRNA levels in the dorsal root ganglia [66], which suggested minor effects on CGRP synthesis in an ovariectomized rat model. GFW had no significant effects on CGRP concentrations in the spinal cord [66], which suggested that it may have an inconsequential influence on CGRP's effects on the central nervous system in the ovariectomized rat model.

Pharmacodynamic Effects of GFW on MHFs
Characteristics and results of included in vivo studies are summarized in Table 3. The results from the quality assessment of included in vivo experimental studies are presented in Figure 3. The results suggested that GFW had a similar action to E 2 on managing skin temperature by restoring the plasma level of CGRP: it increased the ovariectomy-induced CGRP reduction [66,67] and reduced CGRP-induced elevation of skin temperature in GnRH (gonadotropin-releasing hormone) analog-treated rats [67,68]. A dose-dependent skin temperature change was detected, showing that a significant inhibition effect on elevated skin temperature was observed at a dose of 1000 mg/kg [67,68]. However, GFW had no significant effects on CGRP concentrations and CGRP mRNA levels in the dorsal root ganglia [66], which suggested minor effects on CGRP synthesis in an ovariectomized rat model. GFW had no significant effects on CGRP concentrations in the spinal cord [66], which suggested that it may have an inconsequential influence on CGRP's effects on the central nervous system in the ovariectomized rat model. metabolites. The results demonstrated that GFW showed no estrogenic activity and low ER βdependent estrogenic activity before or after metabolization with a concentration at 100 µg/mL. Findings from two in vivo studies concluded that E2 could significantly reduce the levels of pituitary LH (luteinizing hormone) and FSH, as well as the weights of uterus and ovaries, whereas GFW did not influence those factors [67,68]. Thus, GFW did not confer estrogen activity on plasma and might exert pharmacodynamic effects against MHFs via other pathways.  One in vitro experimental study [69] utilized an ESR-dependent cell proliferation bioassay and an ESR-dependent reporter assay to investigate the potential estrogenic activity of GFW and its metabolites. The results demonstrated that GFW showed no estrogenic activity and low ER β-dependent estrogenic activity before or after metabolization with a concentration at 100 µg/mL. Findings from two in vivo studies concluded that E 2 could significantly reduce the levels of pituitary LH (luteinizing hormone) and FSH, as well as the weights of uterus and ovaries, whereas GFW did not influence those factors [67,68]. Thus, GFW did not confer estrogen activity on plasma and might exert pharmacodynamic effects against MHFs via other pathways.

Therapeutic Effects
A well-designed RCT is considered as the gold standard when evaluating the effects of an intervention [70]. However, its limitations have triggered extensive discussion as RCTs may be unnecessary, inappropriate, inadequate, or even impossible in the clinic, and methodological problems can cause bias in the results [71][72][73]. Besides, CHM involves a holistic therapeutic approach. However, most of the published RCTs on CHM were designed to only investigate the therapeutic aspects (i.e., Chinese medicine intervention + western medicine diagnosis) instead of the holistic therapeutic approach (i.e., Chinese medicine intervention + Chinese medicine diagnosis) [74]. The importance of integration of various study designs was highlighted to accommodate the evaluation of various research questions and for the outcome of interest [75,76]. Therefore, the therapeutic effects of a range of study designs were systematically reviewed (Figure 4).

Therapeutic Effects
A well-designed RCT is considered as the gold standard when evaluating the effects of an intervention [70]. However, its limitations have triggered extensive discussion as RCTs may be unnecessary, inappropriate, inadequate, or even impossible in the clinic, and methodological problems can cause bias in the Guizhi Fuling Wan  As shown in Figure 4, the study types and outcome measures varied across all the included studies. Most outcomes were measured in different study designs. GFW demonstrated overall no statistically significant changes in the frequency and severity of MHFs between before and after treatment and no statistically significant differences between GFW and control groups. However, dose-dependent skin temperature changes were detected in both clinical [15] and experimental studies [69,70]. The effective dose suggested in the in vivo study was 1000 mg/kg [69,70], which was four times the dose in the RCT (12.5 g/day) when the participant weighed in at 50 kg [15]. Thus, more dose variation should be considered in further RCTs. The influence of GFW on FSH and E 2 was unclear based on the included studies due to the limitation of study design. Results of experimental studies indicated that GFW had no effects on LH, FSH, and E 2 in Sprague-Dawley rats and no estrogenic effects on the rat liver S9 fraction. However, case series studies pointed out that GFW might target a specific human ESR2 beta genotype. GFW was found to affect CGRP level in both case series and in vivo studies [28,66,67,69]. GFW had similar effects to E 2 on normalizing CGRP level but without affecting its synthesis in the dorsal root ganglia. Only peripheral blood flow and cytokines were reported in one study design, which should be further investigated in various studies.

Potential Multi-Targeting Actions
Three potential target proteins/cytokine could be proposed based on findings from the above-mentioned clinical and experimental studies: estrogen beta receptor, IL-8, and CGRP receptor.
There are significant associations between the ESR2 CA dinucleotide repeat length and other menopausal-related symptoms. For example, low bone mineral density has been reported previously in pre-and post-menopausal women [31,77,78]. This genetic variation was characterized by Tsukamoto and co-workers in 1998 [79]. Recently, ESR2 CA dinucleotide repeat length was reported to be closely related to clinical effects of GFW on MHFs and that it had better performance on participants with the LL genotype [29]. Women with the SS genotype had significantly higher values of circulation estrone concentration (highest) when compared to those harboring the LL genotype (lowest) after adjustment for other confounding factors, including age, hormone replacement status, and circulating levels of sex hormone-binding globulin [31]. Thus, genetic variation should be considered as an essential factor of using GFW for MHFs to achieve precision treatment. The prevalence of the SS and SL genotypes was approximately 47.79% (108/226) and 22.56% (51/226), respectively [31]. The sum of these two genotypes was much higher than that of the LL genotype (29.65%, 67/226). Thus, patients without targeting genotype might contribute to the negative results in the RCT. It is suggested that future research focus on the three genotypes of ESR2 for precision medicine.
IL-8, also known as CXCL8, is a cytokine potentially involved in thermoregulation and was detected to be higher in premenopausal and menopausal women with MHFs than in those without MHFs [80,81]. GFW could significantly reduce concentrations of serum IL-8, whereas those in the non-responders were increased dramatically [27]. Thus, GFW might selectively influence the synthesis of IL-8. However, the underlying mechanism of this selective effect in different MHF sufferers is not fully understood.
CGRP is a widely expressed sensory neuropeptide, which plays a major role in modulating metabolism, inflammatory response, and blood pressure, as well as contributing to nerve development and function [82][83][84][85][86]. In the sympathetic nervous system, CGRP is believed to be the specific vasodilator responsible for MHFs [4,84]. Recently, Liang et al. [86] have determined the structure of the human CGRP receptor in complex with CGRP, with the Gs-protein heterotrimer at 3.3 Å global resolution, which makes it possible for future investigation of the potential herb-target relationship between GFW and MHF-related target protein.

Safety
Inorganic elements that may have vital biological activities in the human body were traced in GFW, including calcium, sodium, magnesium with a dose over 300 µg/g, followed by iron, manganese, and strontium with a dose over 10 µg/g [65]. The dosage of toxic inorganic elements detected in the included studies included beryllium, arsenic, cadmium, antimony, mercury, thallium, and bismuth, which were under 0.03 µg/g and were lower than the maximum limit level listed in the Chinese Pharmacopoeia [65]. GFW is believed to be a safe herbal product as only a small number of minor gastrointestinal symptoms were reported as adverse effects: two studies reported adverse events, including diarrhea (n = 15) [15] and abnormal feeling in the gastrointestinal tract (n = 2) [27]. They were the main causes leading to drop out. Additionally, an in vivo study on Sprague-Dawley rats indicating that GFW was not carcinogenic after a continuous 24-month intervention [87]. Results from another in vitro experiment [68] on estrogen-dependent human breast cancer (MCF-7) cells showed that GFW at concentrations of 10.6-10.4 mg/mL did not activate the proliferation of MCF-7 cell, which suggested that GFW did not exhibit estrogen activity. Therefore, GFW might be a potential therapeutic option for MHFs in women who are undergoing breast cancer treatment.

Implications for Future Research
In total, 169 compounds from GFW and three potential target proteins/cytokine (estrogen beta receptor, IL-8, and CGRP receptor) were identified based on findings from the above-mentioned clinical and experimental studies. However, at this stage, a clear understanding of their interactions could not be obtained based on current literature. The bioactivities of the phytochemicals and their mechanisms of actions on MHF were not fully investigated. Although quality control is improving during the last two decades, evidenced by the increased quantitative index from one in the 2005 edition to two or three in the 2015 edition of the Chinese Pharmacopeia [17], not all the herbal ingredients have a quantitative index. The clinical effects of CHM are often regarded as the result of the multi-targeting interaction of various phytochemical compounds. Thus, more research on the mechanism of actions of CHM on a condition at the molecular level (ligand-target interaction) and the therapeutic dose of the bioactive compounds are warranted.

Conclusions
This study identified 169 compounds from GFW pills, tablets, and capsules. Therapeutic effects, including frequency and severity of MHFs, peripheral blood flow, hormones, neurotransmitter, and cytokines, were not consistent across different study types. ESR2 with genetic variation, CGRP receptor, and IL-8 were identified to be related to the mechanisms of actions of GFW on MHFs. The differences in therapeutic effects could be potentially due to dose variations among clinical and experimental studies and ESR2 gene differences in rats and homo species. Further clinical and experimental studies, as well as biochemical investigation on the mechanisms of actions of GFW, are warranted. It is recommended that the dose-dependent effects and ESR2 expression at the participant recruitment stage are considered in future investigation.