identified compounds by means of HPLC, LC-MS, etc.

**Table 8.** Acetophenone glycosides from *Stachys* spp.


#### **Table 9.** Lignans from *Stachys* spp.



spp.





**Table 11.** Phenylpropanoid glucosides from *Stachys* spp.




**Table 12.** *Cont.*

**Table 13.** Diterpenes from *Stachys* spp.




**Table 14.** Triterpene derivatives, Phytosterols and Phytoecdysteroids from *Stachys* spp.

**Table 15.** Megastigmane derivatives from *Stachys* spp.


#### *4.1. Flavonoids*

The genus *Stachys* consists a rich source of flavonoids. Accumulating studies have reported the several types of flavonoids occurring in *Stachys* spp., including flavones (Tables 2 and 16), poly-methylated flavones (Tables 3 and 17), flavonols (Tables 4 and 18), flavanones (Tables 5 and 19) and one biflavonoid (Tables 6 and 20).

Regarding the flavone derivatives (Tables 2 and 16), 18 flavone 7-O-acetylallosylglucosides were mentioned in the most species of subgenus *Stachys* (31 species). The flavone 7-O-glucosides were also found in many species through the two subgenera. Marin et al. (2004) reported that tricetin 3 ,4 ,5 -trimethyl-7-O-glucoside (**62**) consists a chemotaxonomic marker for the subgenus *Betonica* [5]. Precisely, selgin 7-O-glucoside (**59**), tricin 7-O-glucoside (**61**) and tricetin 3 ,4 ,5 -trimethyl-7-Oglucoside (**62**) were identified from the leaves of three species of the latter subgenus; *S. alopecuros* (section Betonica), *S. o*ffi*cinalis* (section Betonica) and *S. scardica* (section Macrostachya) [5]. Furthermore, derivatives of apigenin *p*-coumaroyl glucosides and chrysoeriol *p*-coumaroyl glucosides were reported in *Stachys* species, though some *p*-coumaroyl glucosides (not determined) were also identified [5,75]. To be mentioned that chrysoeriol 7-*O-*glucoside (**43**), chrysoeriol *p*-coumaroyl glucosides (**46,47**) and chrysoeriol 7-O-[6-O-acetyl-allosyl]-(1→2)-glucoside (stachyspinoside) (**44**) were mainly isolated from wild Greek taxa of the subgenus *Stachys* [3,77,98,99,102], apart from the Greek species *S. ionica* [20], *S. tetragona* [100] and the cultivated species *S. iva* [56]. Nazemiyeh et al. (2006) investigated the phytochemical profile of the stems of *S. schtschegleevii*, reporting four flavonoids, among them were also two *p*-coumaroyl derivatives of apigenin and chrysoeriol [74]. Moreover, flavone 7-O-mannosylglucosides were reported from the two species *S. atherocalyx* (section Eriostomum) and *S. spectabilis* (section Olisia) [72,89,90]. Few flavone C-glucosides were mentioned in the species *S. aegyptiaca* (subg. *Stachys*; sect. Ambleia), *S. o*ffi*cinalis*(subg. *Betonica*; sect. Betonica), and *S. scardica* (subg. *Betonica*; sect. Macrostachya) [5,68,104]. Zinchenko (1973) reported the existence of two derivatives of methoxybaicalein, namely palustrin (**63**) and palustrinoside (**64**), from the species *S. palustris* of subgenus *Stachys*(section Stachys) [104]. Notably, the subterranean organs of *S. annua* were investigated and the isolation of two flavone derivatives was reported, namely 4 -O-methyl-isoscutellarein (**12**) and 4 -O-methyl-isoscutellarein-7-O-(6-O-acetyl)allopyranosyl-(1→2)-glucopyranoside (**21**) [95].

Furthermore, our survey revealed the presence of poly-methylated flavones in the genus *Stachys* (Tables 3 and 17). Precisely, six species and four subspecies from subgenus *Stachys*, as well as one species from subgenus *Betonica*, are found to contain poly-methylated flavones. The most common representative was xanthomicrol (**69**) which was mentioned in seven *Stachys* species and subspecies of different sections from the subgenus *Stachys* [20,68,74,77,78,102,107]. In the stems of the species *S. schtschegleevii*, apart from xanthomicrol (**69**), was also found circimaritin (**66**) [74].

A few studies mentioned the existence of flavonols in *Stachys* spp. (Tables 4 and 18), mainly in species occurred in Greece. Afouxenidi and colleagues (2018) isolated kaempferol (**91**) from the *n*-butanol residue of the aerial parts of *S. tetragona* [100], which was also identified in the aerial parts of *S. cretica* subsp. *smyrnaea* [81]. Moreover, isorhamnetin (**92**) was isolated from the methanol extract of the aerial parts of *S. swainsonii* subsp. *swainsonii* and *S. swainsonii* subsp. *argolica* [102]. A study conducted by Marin et al. (2004) identified the presence of quercetin 3-O-rutinoside (**93**) and isorhamnetin 3-O-glucoside (**94**) from the aerial parts of *S. palustris* [5].

In addition, three flavanones were isolated from three species of the genus *Stachys*(Tables 5 and 19). Eriodictyol (**95**) was mentioned in *S. cretica* [108] and in one subspecies of *S. swainsonii* [102], while naringenin (**96**) was isolated from the aerial parts of the species *S. aegyptiaca* [104]. A flavanone rutinoside, known as hesperidin (**97**), was identified as one of the major compounds of the aerial parts of *S. cretica* subsp. *smyrnaea* [81].

Of great interest is the isolation of a rare diflavone ester of μ-truxinic acid, namely stachysetin (**98**). It is well-known that diglycoside flavone esters of dicarboxylic acids are rare compounds in plant kingdom. Stachysetin was firstly isolated from the ethanol extract (70% v/v) of the aerial parts of *S. aegyptiaca* [69]. Then, Murata and co-workers (2008) reported it in the methanol residue (80% v/v) of the aerial parts of *S. lanata* [82]. In a current study carried out by Pritsas et al. (2020), stachysetin was isolated from the methanol: aqueous (5:1) extract from the flowering aerial parts of the cultivated *S. iva* (Tables 6 and 20) [56]. Up to now, there is no report of this secondary metabolite in the species of the subgenus *Betonica*. The presence of this rare natural compound in the sections Ambleia, Eriostomum and Candida of the subgenus *Stachys* might be considered as a chemotaxonomic marker among the two subgenera and of the genus *Stachys*.


isolated from *Stachys* spp.

> **Table 16.**

Chemical structures of flavones



#### *Medicines* **2020** , *7*, 63



#### *Medicines* **2020** , *7*, 63


**Table 17.** Chemical structures of poly-methylated flavonoids from *Stachys* spp.


**Table 18.** Chemical structures of flavonols from *Stachys* spp.



**Table 19.** Chemical structures of flavanones from *Stachys* spp.

**Table 20.** Chemical structure of biflavonoid from *Stachys* spp.

#### *4.2. Phenolic Derivatives; Acetophenone Derivatives*

Regarding the phenolic derivatives of genus *Stachys*(Tables 7 and 21), mainly chlorogenic acid (**103**) was appeared in nine *Stachys* species; *S. candida* [78], *S. iva* [56], *S. cretica* (*S. cretica* subsp. *smyrnaea* [81], *S. cretica* subsp. *mersinaea* [108], *S. cretica* subsp. *vacillans* [112]), *S. lanata* [82], *S. tmolea* [85], *S. thirkei* [84], *S. recta* [14], *S. palustris* [104] and *S. o*ffi*cinalis* [111]. The isomers of chlorogenic acid (**102**, **104**, **105**) also reported in *S. atherocalyx* [110], *S. recta* [14] and *S. palustris* [23,104]. Caffeic (**108**) and *p*-coumaric (**106**) acids were found in two *Stachys* spp. [104,110]. Moreover, Kirkan (2019) identified vanillic (**100**) and syringic (**101**) acids from the aerial parts of *S. cretica* subsp. *vacillans* [112]. Though, 4-hydroxybenzoic acid (**99**) was reported from *S. tmolea* [85]. Arbutin (**107**) was also identified in the aerial parts of *S. germanica* subsp. *salviifolia* [109]. One study also reported the presence of acetophenone derivatives from the roots of *S. lanata*, namely androsin (**109**), neolloydosin (**110**) and glucoacetosyringone (**111**) (Tables 8 and 22) [82]. The isolation of the latter compounds might be attributed to the different investigated plant parts (roots).

**Table 21.** Chemical structures of phenolic derivatives from *Stachys* spp.

Glc: glucose.

#### *4.3. Lignans*

Lignans are types of polyphenols with diverse structures. Although these bioactive compounds were presented in Lamiaceae family [149], a few studies reported their existence in plants of genus *Stachys*. Specifically, three lignans categorizing into two furanofuran-type derivatives (sesamin and paulownin) and one benzofuran-type lignan (urolignoside) were reported in two species of the subgenus *Stachys* (Tables 9 and 23). Laggoune et al. (2016) isolated sesamin (**112**) and paulownin (**113**) from the aerial parts of *S. mialhesii* [103], while urolignoside (**114**) was isolated from the aerial parts of *S. tetragona* [100]. Given that up to now there is no study reported the presence of lignans in the subgenus *Betonica*, the identification of lignans might be considered as a chemotaxonomic difference between the two subgenera *Stachys* and *Betonica*.

**Table 23.** Chemical structures of lignans from *Stachys* spp.

#### *4.4. Phenylethanoid Glycosides; Phenylpropanoid Glucosides*

The present review unveiled 29 phenylethanoid glycosides in 17 *Stachys* species (Tables 10 and 24). Acteoside or verbascoside (**118**) was the most abundant found in 16 *Stachys* spp. of all sections through this survey. Additional phenylethanoid glycosides isolated and identified from this genus includes martynoside, leucosceptoside A and lavandulifoliosides. Lavandulifolioside A (or stachysoside A) (**129**) was firstly isolated from the methanol extract of the aerial parts of *S. lavandulifolia* in 1988 [115], while in 2011 Delazar et al. (2011) isolated lavandulifolioside B (**130**) from the same plant, for the first time [12]. Moreover, three phenylethanoid glycosides were reported from the aerial parts of *S. byzantina* (section Eriostomum), including verbascoside (**118**), 2 -O-arabinosyl verbascoside (**122**) and aeschynanthoside C (**133**) [35]. Among them, the first and the last compound has been isolated only from the specific species. A survey conducted by Murata and co-workers (2008) reported ten phenylethanoid glycosides from different plant parts [82]. In the aforementioned study, leonoside B (or stachysoside D) (**134**) and martynoside (**135**) were mentioned from the aerial parts of *S. lanata*, while from the roots of the specific species were reported eight phenylethanoid glycosides, namely rhodioloside (**115**), verbasoside (**116**), 2-phenylethyl-D-xylopyranosyl-(1→6)-D-glucopyranoside (**117**), verbascoside (**118**), isoacteoside (**119**), darendoside B (**120**), campneoside II (**121**) and campneoside I (**136**). It is remarkable to point out that compounds **115**, **117** and **120** haven t been reported in other *Stachys* species. This might be attributed to the fact that the plant material was roots. Another study carried out by Karioti et al. (2010) focused on the phenolic compounds from the aerial parts of *S. recta*, and reported many phenylethanoid glycosides from its aerial parts, including acteoside (**118**), isoacteoside (**119**), β-OH-acteoside (**121**), betunyoside E (**127**), campneoside I (**136**), forsythoside B (**137**), β-OH-forsythoside B methyl ether (**138**) [14]. Furthermore, lamiophloside A (**141**) was isolated with some other phenylethanoid glycosides from the aerial parts of *S. tetragona* [100]. Of great interest is that our survey revealed that this constituent is mentioned only in the specific species. Two rare phenylethanoid glycosides, parviflorosides A-B (**142**–**143**) were isolated from the whole plant of *S. parviflora* [120]. These two compounds are characterised by the presence of a third saccharide (rhamnose) linked to the proton H-2 of glucose, comparing to others common phenylethanoid glycosides where the connection of the third saccharide is in proton H-3 of glucose. Of great interest is that *S. parviflora* is now considered as the monotypic genus *Phlomidoschema* (only *P. parviflorum* (Benth.) Vved.) [2]. Furthermore, leonoside A (or stachysoside B) (**139**) was isolated with other three phenylethanoid glucosides from the whole plant of *S. riederi* [114]. To be mentioned that phenylethanoid glycosides were reported in both subgenera of genus *Stachys*.

Apart from phenylethanoid glucosides, Murata et al. (2008) mentioned two phenylpropanoid glucosides in the roots of *S. lanata* (subg. *Stachys*; sect. Eriostomum), coniferin (**144**) and syringin (**145**) (Tables 11 and 25) [82]. It is worth to mention that the isolation of phenylpropanoid glucosides only from the specific plant, might be assigned to the different studied plant material (roots).




**Table 25.** Chemical structures of phenylpropanoid glucosides from *Stachys* spp.

#### *4.5. Iridoids*

Iridoids are among the major chemical compounds found in genus *Stachys*. According to Tundis et al. (2014), iridoids are considered as good chemotaxonomic markers of this genus [3]. Accumulating phytochemical studies have reported diverse types of iridoids [3]. The present review summarises all these studies, exemplifying 38 *Stachys* species which their iridoid cargo has been investigated (Tables 12 and 26). Harpagide (**148**; 31 species) and its acetyl derivative; 8 acetyl-harpagide (**150**; 28 species) are of common occurrence in genus *Stachys* and might be considered as characteristic iridoids of these plants. Furthermore, ajugol (**146**; 18 species), ajugoside (**147**; 18 species), melittoside (**166**; 17 species), monomelittoside (**165**; 4 species) and 5-allosyloxy-aucubin or 5-O-allopyranosyl-monomelittoside (**167**; 4 species/1 subsp.) were also mentioned in various species. Allobetonicoside (**161**) was firstly isolated from the aerial parts of *S. o*ffi*cinalis* [127] and then from the aerial parts of *S. glutinosa* [122] and of *S. macrantha* [117]. The latter study also mentioned the isolation of cinnamoyl-harpagide derivative, macranthoside (**156**), for the first time. To be mentioned that Jeker et al. (1989) also isolated 6-O-acetylmioporoside (**155**) from the aerial parts of *S. o*ffi*cinalis* [127]. In addition, two species revealed the presence of 8-*epi*-loganic acid (**157**), 8-*epi*-loganin (**159**) and gardoside (**160**) [20,56], as well as 7-O-acetyl-8-*epi*-loganic acid (**158**) was only mentioned from the aerial parts of *S. spinosa* [98]. Of note, Iannuzzi et al. (2019) isolated from the leaves of *S. ocymastrum* (syn. *S. hirta* L.) five iridoids which haven t been documented in other species, namely 6β-acetoxyipolamiide (**172**) 6β-hydroxyipolamiide (**173**), ipolamiide (**174**), ipolamiidoside (**175**) and lamiide (**176**) [123]. A study conducted by Háznagy-Radnai (2006) examined the phytochemical profiles of *Stachys* spp. growing in Hungary, reporting the iridoid content of ten taxa [124]. Murata and co-workers (2008) isolated five new esters of monomelittoside from the aerial parts and roots of *S. lanata* [82]. In particular, stachysosides E (**168**), G-H (**170–171**) were found in roots, while stachysosides E (**168**) and F (**169**) were discovered from the aerial parts of the specific species. It is important to be mentioned the detection of a new iridoid diglycoside, 4 -O-β-D-galactopyranosyl-teuhircoside (**162**), which was isolated from the flowering aerial parts of *S. alopecuros* subsp. *divulsa* [119]. Muñoz et al. (2001) reported the presence of 5-desoxy-harpagide (**151**) and 5-desoxy-8-acetyl-harpagide (**152**) from the aerial parts of *S. grandidentata* [129]. Notably, this review unveiled some differences in iridoids among subgenera *Stachys* and *Betonica*. Firstly, it was observed that there is no report for the presence of monomelittoside or melittoside derivatives in the subgenus *Betonica*. Secondly, reptoside (**153**) was found in two species of subgenus *Betonica* (*S. macrantha* and *S. o*ffi*cinalis*) and not in the plants of subgenus *Stachys*.


*Medicines* **2020** , *7*, 63

**Table 26.**

Chemical structures of iridoids from

*Stachys* spp.

*Medicines* **2020** , *7*, 63

#### *4.6. Diterpenes*

A landmark study for diterpenes of genus *Stachys* is the review article of Piozzi and Bruno (2011), including all the reported diterpenoids from roots and aerial parts of *Stachys* spp [21]. Up to now, several types of diterpenes have been mentioned, comprising types of *neo*-clerodane, labdane, rosane and *ent*-kaurene skeleton (Tables 13 and 27). The most common type is the *neo*-clerodane skeleton derivates, as they were found in various species. *S. aegyptiaca* has thoroughly studied for its phytochemical profile. A study conducted by Hegazy et al. (2017) reported the discovery of three new *neo*-clerodane diterpenoids from the aerial parts of the aforementioned plant, namely stachaegyptins A-C (**190**–**192**) [106]. One year later, two new compounds were mentioned; stachaegyptins D-E (**193**–**194**) [131], while in a current work stachaegyptins F-H were isolated (**195**–**197**) [133]. Moreover, stachysperoxide (**189**) was isolated from the *S. aegyptiaca* [132]. These stachaegyptin derivatives and stachysperoxide reported only in the species *S. aegyptiaca* and might be a characteristic chemical compound of the specific plant of the section Ambleia. Derkach (1998) mentioned the compounds annuanone (*cis*-clerodane type) (**181**), stachylone (**182**) and stachone (**183**) in five species of the subgenus *Stachys*; *S. atherocalyx*, *S. inflata*, *S. iberica* and *S. sylvatica* [134]. Other *neo*-clerodane type diterpenes which were found in many species are roseostachenone (**184**), roseostachone (**185**), roseostachenol (**186**) and roseotetrol (**187**). Ruiu and co-workers (2015) explored the aerial parts of *S. glutinosa*, isolating roseostachenone and the new *neo*-clerodane diterpene, 3α,4α-epoxyroseostachenol (**188**) [107]. Furthermore, labdane type derivatives were occurred in the genus *Stachys*. Fazio et al. (1994) investigated the aerial parts of *S. mucronata* and isolated three labdane skeleton compounds; ribenone (**198**), ribenol (**199**) and 13-*epi*-sclareol (**200**) [57]. The latter compound has also been found in *S. rosea* [141]. Paternostro et al. (2000) studied the aerial parts of *S. plumosa*, determining the following labdane type derivatives (+)-6-deoxyandalusol (**201**), 13-*epi*-jabugodiol (**202**) and (+)-plumosol (**203**) [144]. The compound (+)-6-deoxyandalusol were also found in *S. distans* and *S. ionica* [139]. Some *ent*-kaurene derivatives were reported in *S. aegyptiaca* [130], *S. lanata* [135] and *S. sylvatica* [142]. Moreover, one abietane diterpenoid, horminone (**211**), was isolated from the aerial parts of *S. mialhesii* [103]. It is noteworthy to be underlay the presence of two rare rosane type diterpenes in the aerial parts of *S. parviflora*, namely stachyrosanes 1 (**212**) and 2 (**213**) [134]. In addition, six diterpene lactone derivatives, i.e., betolide (**214**), betonicolide (**215**) and betonicosides A-D (**216**–**219**) were found in the species *S. o*ffi*cinalis* [143,145] and *S. scardica* [143] of the subgenus *Betonica*.

In the context of chemotaxonomic significance, it could be observed that species of subgenus *Stachys* product mainly *neo*-clerodane and labdane type derivatives, while the plants of subgenus *Betonica* biosynthesized diterpene lactone derivatives. Thus, the latter derivatives might be recognised as characteristic chemotaxonomic markers of subgenus *Betonica*. Another important chemotaxonomic point is reported by Piozzi et al. (2002), mentioning that (+)-6-deoxyandalusol has been determined only in three *Stachys* species of eastern part of the Mediterranean region [139].

**Table 27.** Diterpenes from *Stachys* spp.

**Table 27.** *Cont.*

**Table 27.** *Cont.*

#### *4.7. Triterpene Derivatives, Phytosterols and Phytoecdysteroids*

Triterpene derivatives and phytosterols are major secondary metabolites of Lamiaceae family. In genus *Stachys*, five phytosterol derivatives (**220**–**224**) were found in *S. byzantina* [17,35], *S. annua* [95],

*S. spinosa* [99], *S. tetragona* [100], *S. palustris* [146] and *S. alopecuros* subsp. *divulsa* [119] (Tables 14 and 28). Furthermore, the triterpenoids; ursolic (**226**) and oleanolic (**227**) acids were only reported from the section Olisia (subg. *Stachys*) [95,99,100]. Kotsos et al. (2007) isolated an oleanolic lactone derivative (**228**) of the aerial parts of *S. spinosa* [99]. It is noteworthy to be mentioned the presence of saponin derivatives in genus *Stachys* (Tables 14 and 28). The first saponins isolated from this genus were from the water extract of the whole plant of *S. riederi*, including 8 stachyssaponins (I-VIII, **231–238**) [147]. Afterwards, stachyssaponins A-B (**229–230**) were found from the methanol extract of the aerial parts of *S. parviflora* [63].

Few *Stachys* spp. include phytoecdysteroids (Tables 14 and 28). Ramazanov and co-workers (2016) isolated five phytoecdysteroids from *S. hissarica* [67], namely 20-hydroxyecdysone (**239**), polipodin B (**240**), integristeron A (**241**), 2-desoxy-20-hydroxyecdysone (**242**) and 2-desoxyecdyson (**243**).

**Table 28.** Triterpene derivatives, Phytosterols and Phytoecdysteroids from *Stachys* spp.

**Table 28.** *Cont.*

Glc: Glucose, Xyl: Xylose, Rha: Rhamnose, Ara: Arabinose.

#### *4.8. Other Chemical Categories*

Notable among the above-mentioned classes of compounds are the megastigmane derivatives from *Stachys* spp. (Tables 15 and 29). Takeda and colleagues (1997) isolated from the aerial parts of *S. byzantina* five bioactive compounds from this group, including byzantionosides A-B (**244**,**245**), icariside B2 (**246**), (6R, 9R)- and (6R, 9S)-3-oxo-α-ionol glucosides (**247**) and blumeol C glucoside (**248**) [148]. Furthermore, vomifoliol (**249**) and dehydrovomifoliol (**250**) were reported from the aerial parts of *S. lanata*, while citroside A (**251**) was isolated from the roots of this species [82]. This study also mentioned the presence of sugar ester (cistanoside F) from the roots of *S. lanata* [82]. At this point, we should note that few studies reported some oligosaccharides from *Stachys* spp. [3]. For instance, stachyose is a tetrasaccharide which consists one of the most common oligosaccharides in genus *Stachys* and shows beneficial effects for the gastrointestinal system as it can be directly consumed [3,23,119,150]. Precisely, the species *S. sieboldii* is a major source of this constituent [27,151,152]. Stachyose is an oligosaccharide, which can be directly consumed for the benefit of gastrointestinal system [150]. Furthermore, Yin and colleagues (2006) mentioned that the bitter taste of some *Stachys* species, such as *S. annua* and *S. balansae*, might be attributed to their bitter diterpene derivatives, like stachylone [22,151].

**Table 29.** Chemical structures of megastigmane derivatives from *Stachys* spp.

#### **5. Pharmacological Activities**

This section includes the most interesting pharmacological data of the last five years (from 2015 to 2020). Many studies exemplified the great antimicrobial, antioxidant and cytotoxic effects of the essential oils of these plants [3,15]. Tundis et al. (2014) described in detail the biological studies (in vitro and in vivo) of the essential oils, extracts and compounds [3]. Thus, in the present review, we focused on the current available pharmacological researches of the extracts and isolated compounds from *Stachys* spp. as they are presented in Table 30.


**Table 30.** Pharmacological activities of *Stachys* spp.






134

#### *5.1. Antioxidant Activity*/*Cytoprotective*

Tundis et al. (2015) evaluated five extracts (*n*-hexane, dichloromethane, methanol, methanol with Soxhlet apparatus and ethanol 70% extract) from the aerial parts of *S. lavandulifolia* for their antioxidant activity, using β-carotene bleaching test, 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulphonic acid (ABTS), 1,1-Diphenyl-2-picrylhydrazyl (DPPH), and Ferric Reducing Antioxidant Power (FRAP) assays [116]. The most polar extracts, ethanol 70% and methanol extracts, showed the highest radical scavenging activity against ABTS radical (IC50 values of 19.9 and 22.8 μg/mL, respectively), whereas the methanol extract Soxhlet apparatus was the most active in the DPPH method (IC50 of 25.0 μg/mL). In the β-carotene bleaching test, the methanol and ethanol extract demonstrated the stronger activity after 30 min of incubation (IC50 = 29.3 and 33.0 μg/mL, respectively) and the IC50 values were of 60.3 and 34.6 μg/mL, respectively after 60 min of incubation. Moreover, they studied the antioxidant activity of bioactive secondary metabolites; arbutin (**107**), acteoside (**118**), monomelittoside (**165**), melittoside (**166**), 5-allosyloxy-aucubin (**167**), and stachysolone (**177**), reporting that in both DPPH and ABTS assays the most active compounds was arbutin (**107**) with IC50 values of 62.5 and 45.7 μg/mL, respectively [116]. Another work investigated the antioxidant activity of three extracts of *S. guyoniana*, through β-carotene–linoleic acid, DPPH, ABTS, CUPric Reducing Antioxidant Capacity (CUPRAC) and metal chelating assays [155]. The chloroform extract had the highest antioxidant activity (IC50 = 2.3 ± 1.27 μg/mL) in β-carotene–linoleic acid and in ABTS method (IC50 = 7.29 ± 0.23 μg/mL). The *n*-butanol extract showed the better antioxidant capacity in DPPH test (IC50 = 2.91 ± 0.14 μg/mL) compared to other extracts and to the reference compound α-tocopherol (IC50 = 7.31 ± 0.17 μg/mL), as well as it demonstrated highest activity in CUPRAC method (A0.50 = 0.15 ± 0.05 μg/mL) and in metal cheating assay (inhibition at 100 μg/mL: 48%). In addition, Laggoune et al. (2016) demonstrated the great antioxidant properties in vivo of *S. mialhesii* [103]. Particularly, the *n*-butanol extract of the specific plant showed IC50 value of 0.047 mg/mL in DPPH assay, while the IC50 value of the isolated compound isoscutellarein-7-O-[6-O-acetyl]-β-D-allopyranosyl-(1→2)-β-D-glucoside (**15**) was 0.066 mg/mL and the reference compound quercetin was 0.012 mg/mL. Notably, they also mentioned that the extract (up to 10 g/kg) did not show any toxicity in mice during 24 h after administration. In addition, the antioxidant activity of three subspecies of *S. cretica* (i.e., *S. cretica* subsp. *mersinaea*; *S. cretica* subsp. *smyrnaea*; *S. cretica* subsp. *vacillans*) were investigated in different works [81,108,112]. The antioxidant capacity of the methanol extract of *S. parviflora* was measured, exhibiting an IC50 value of 76.87 ± 0.57 μg/mL (DPPH method) and of 188.47 ± 0.76 μg/mL (β-carotene bleaching test; BCB), while the standard compound, butylated hydroxytoluene (BHT), had stronger activity in both tests (DPPH test: IC50 = 1.23 ± 0.02 μg/mL; BCB test: 34.31 ± 0.40 μg/mL) [64]. Guo et al. (2018) examined the antioxidant activity of five fractions from the 70% ethanol extract of tubers of *S. a*ffi*nis* by DPPH assay and superoxide radical scavenging activity [28]. The ethyl acetate fraction showed extremely high antioxidant activity in DPPH method (IC50 = 0.85 ± 0.04 μg/mL) with α-tocopherol as positive control (IC50 = 18.68 ± 0.51 μg/mL). They reported that this great antioxidant activity was attributed to the high content in phenolics and flavonoids of this fraction and confirmed the use of this plant as a natural antioxidant. Another work studied the antioxidant activity of the extracts and fractions of the same *Stachys* species on reactive oxygen species (ROS) production induced by H2O2 in HT-1080 cells [29]. In particular, the *n*-hexane fraction decreased H2O2-induced ROS and oxidative stress-induced DNA damage, as well as it increased glutathione (GSH) production. The species *S. mucronata* demonstrated strong anti-radical activity due to the high content in polyphenols [156]. A recent study conducted by Aminfar et al. (2019) described a chemometric-based approach in order to classify *S. lanata* by Gas Chromatography-Mass Spectrometry (GC-MS) fingerprints and to correlate their chemical constituents with their antioxidant capacity [35]. They identified eight antioxidant markers which could also serve as volatile markers. In addition, Elfalleh and co-workers (2019) demonstrated the differences of the antioxidant properties of the extracts of *S. tmolea*, reporting that water extract exhibited highest activity than methanol extract, using DPPH, ABTS, CUPRAC, FRAP, phosphomolybdenum and ferrous ion chelating methods [85]. A survey

conducted by Hwang et al. (2019) demonstrated that the ethanol extract of *S. riederi* var. *japonica* exhibited antioxidant effects on ultraviolet A (UVA)-irradiated human dermal fibroblasts (HDFs), through suppression of ROS generation [160]. The antioxidant activity of the methanol extract of the Lebanese species *S. ehrenbergii* was measured by ABTS radical cation decolorization assay and the methanol extract showed an IC50 value of 52 ± 7.5 mg/mL [154]. Furthermore, the chemical profile and some biological activities of three herbal teas in Anatolia were examined [84]. Among them, the methanol extract of *S. thirkei* showed strongest antioxidant capacity, through β-carotene (IC50 = 47.79 ± 0.59 μg/mL), DPPH (IC50 = 49.31 ± 0.38 μg/mL), ABTS (IC50 = 13.34 ± 0.02 μg/mL) and CUPRAC (absorbance%: 1.88 ± 0.02 μg/mL) assays. Sadeghi et al. (2020) assessed the the antioxidant properties of hydroalcoholic extract of *S. pilifera* on nephrotoxicity induced with cisplatin (CP) in vivo (in rats), showing that the specific extract restored the antioxidant capacity, as well as it had renoprotective activity [19].

#### *5.2. Cytotoxicity and Antiproliferative Activity*

Venditti et al., (2017) investigated the cytotoxic activity and the anti-reactive oxygen species activity of the ethanol extract from tubers of the Chinese artichock (*S. a*ffi*nis*) [27]. Regarding the cytotoxicity, the specific extract didn t demonstrate any activity in K562, SH-SY5Y and Caco-2 cell lines, even at the highest concentrations (1.0 mg/mL). The cytotoxic activity of extracts and isolated flavonoids from the aerial parts of *S. lavandulifolia* were studied by Delnavazi et al. (2018) through the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay [13]. The dichloromethane extract showed the highest cytotoxic activity in brine shrimp lethality test (BSLT) (LD50 = 121.8 ± 5.6 μg/mL), while as a positive control was used podophyllotoxin (LD50 = 3.1 ± 0.6 μg/mL). Afterwards, they explored the cytotoxic activity of isolated flavonoids in three cancer cell lines (MDA-MB-231, HT-29 and MRC-5), using as reference compound tamoxifen. All the nine isolated flavonoids moderated the cytotoxicity activated on the studied cell lines. However, chrysosplenetin (**84**) was reported as the most active compound in the first two cell lines. In MRC-5 cell line, apigenin (**1**) exhibited the greatest activity. It is remarkable to point out that the specific study also mentioned the selective activity against cancer cells, reporting that chrysosplenetin (**84**), kumatakenin (**79**) and viscosine (**78**) exhibited higher selective toxicity against MDA-MB-231 cell line than tamoxifen. At this point, we should underlie that the great cytotoxic activity of these compounds is attributed to their substitutions with (poly)-methylated groups which increase this effect. Another study evaluated the methanol extract, the alkaloid and the terpenoid fractions of *S. pilifera* for their cytotoxic and antiproliferative activity in vitro (HT-29 cell line), indicating great results [45]. The terpenoid fraction was found to have the best cytotoxic activity compared to the other fractions and as reference compound was used cisplatin. Moreover, they investigated the antiproliferative activity, studying the effects on the activity of caspase-8 and caspase-9, Nuclear factor-κB (NF-κB) and Nitric Oxide (NO), reporting that the extract/fractions increased the activity of caspase-8/-9 and decreased NF-κB and subsequently NO level. Of note, they compared their results with previous data of cytotoxic activity in vitro of other *Stachys* species such as *S. acerosa, S. benthamiana, S. floridana, S. lavandulifolia, S. obtusicrena, S. persica, S. pubescens* and *S. spectabilis*. Three isolated compounds from the extract (CH2Cl2:MeOH 1:1) of the aerial parts of *S. aegyptiaca* were investigated for the cytotoxic activity in HepG2 cell line, using MTT assay [132]. Precisely, the IC50 values of stachaegyptin D (**193**), stachysolon monoacetate (**178**) and stachysolon diacetate (**180**) were 94.7, 63.4 and 59.5 μM, respectively, with stachysolone diacetate being the most active. In another study, the cytotoxic effect of the ethanol extract of *S. riederi* var. *japonica* on UVA-irradiated HDFs was evaluated at different concentrations for 48 h by MTT assay, showing no or little cytotoxicity [160]. Shakeri et al. (2019) mentioned that the methanol extract of *S. parviflora* demonstrated no cytotoxic activity toward the cancer cell lines, namely A2780, HCT, and B16F10 in all tested concentrations (>100 μg/mL) [64]. Moreover, the genotoxic activity of the extracts from four different plants were investigated by Slapšyte˙ and colleagues (2019) [157]. They reported that all the plant extracts induced DNA damage, using

the comet assay, whereas the extract of *S. o*ffi*cinalis* induced the increase of sister chromatid exchange value. The methanol extract of the Lebanese species *S. ehrenbergii* was investigated for its antioxidant and cytotoxic activity [154]. The cytotoxicity was examined by MTT assay where the methanol extract showed the highest cytotoxicity (IC50 = 420 ± 104 μg/mL) at a concentration of 3000 mg/mL.

#### *5.3. Polycystic Ovary Syndrome (PCOS)*

In Iran, *S. sylvatica* is used for the treatment of women with polycystic ovary syndrome (PCOS). A current study carried out by Alizadeh et al. (2020) evaluated the hydroalcoholic extract of this plant in a rat model of PCOS [47]. It was observed that the extract at the dose of 500 mg/kg increased gonadotropins FSH and LH (5.95 ± 0.02 mIU/mL; 6.48 ± 0.09 mIU/mL) and reduced the level of estrogen (0.9 ± 0.07 mIU/mL) compared to the PCOS group (FSH level: 1.69 ± 0.08 mIU/mL; LH level: 6.29 ± 0.04 mIU/mL; estrogen level: 1.42 ± 0.05 mIU/mL), causing the ratio of LH/FSH to be close to 1:1 (6.48/5.59). According to the literature, this ratio LH/FSH is almost 1:1 in normal cases, while in PCOS women is higher e.g., 2:1 or 3:1. They also mentioned that these great results of the extract of *S. sylvatica* could be correlated to the flavonoid content of the plant. Previous studies showed that flavonoids could decrease the level of estrogen and could also act as GABA receptor agonists, regulating gonadotropins. Given that women with PCOS showed high concentrations of inflammation factors, they assumed that the extract could act as anti-inflammatory and antioxidant agent as flavonoids and iridoids demonstrated antioxidant and anti-inflammatory properties.

#### *5.4. Anticholinesterase and Anti-Alzheimer's Activity*/*Neuroprotective Activity*

The aqueous extract from the tubers of *S. sieboldii* ("chorogi") was studied in vivo in mice model for its neuroprotective potential [152]. Specifically, the study examined the effects of chorogi's extract on celebral ischemia and scopolamine-induced memory impairment, using as positive control the extract of *Gingko biloba*, proving that *S. sieboldii* improves the learning and memory dysfunction correlated with ischemic brain injury. Another work examined the cholinesterase inhibitory activity of *S. lavandulifolia* extracts and isolated compounds [116]. Specifically, the most active extract against anticholinesterase (AChE) was the *n*-hexane extract with an IC50 value of 13.7 μg/mL. However, the dichloromethane extract was the most effective against butyrylcholinesterase (BChE) (IC50 = 143.9 μg/mL) where its major constituent, stachysolone (**177**), inhibited the activity of this enzyme with a percentage of inhibition of 50% at 0.06 mg/mL. Among the studied polar extracts, the methanol extract exhibited a selective inhibitory activity against AChE with an IC50 value of 211.4 μg/mL and the isolated compounds, arbutin (**107**) and 5-allosyloxy-aucubin (**167**), showed a percentage of inhibition of 50 and 23.1% at 0.06 mg/mL, respectively, against AChE. Notably, the other constituents of this species were inactive at the maximum concentration tested of 0.25 mg/mL. Ferhat et al. (2016) examined the AChE activity of *n*-butanol, the ethyl acetate and the chloroform extracts of the aerial parts of *S. guyoniana*, demonstrating that the *n*-butanol extract (IC50 = 5.78 ± 0.01 μg/mL) was a little less active than the used standard drug against Altzheimer's disease; galantamine (IC50 = 5.01 ± 0.10 μg/mL). Furthermore, they exhibited that this extract inhibited the BChE, having an IC50 value of 39.1 ± 1.41 μg/mL which was better than the standard (IC50 = 39.10 ± 1.41 μg/mL) [155]. Moreover, the anti-Alzheimer's activity of two subspecies of *S. cretica* (*S. cretica* subsp. *smyrnaea; S. cretica* subsp. *mersinaea*) were evaluated in different works [81,108]. In addition, the potential effects of 20% ethanol extract of *S. sieboldii* was evaluated against oxidative stress induced by H2O2 in SK-N-SH cells and memory enhancement in ICR mice [162]. This study showed that the daily intake of the extract (dose: 500 mg/kg) through dietary supplementation produced memory enhancing effects in animals. Recently, Ertas and Yener (2020) reported that the acetone extract of *S. thirkei* demonstrated good activity against AChE and BChE with a percentage of inhibition of 52.46 ± 1.26% and 75.04 ± 1.91%, respectively [84].

#### *5.5. Anti-tyrosinase Activity*

The anti-tyrosinase activity of the ethanol and methanol Soxhlet apparatus extracts of the aerial parts of *S. lavandulifolia* exhibited the best activity with IC50 values of 33.4 ± 0.8 and 42.8 ± 1.1 μg/mL [116]. They underlay that the specific extracts were characterized by the phenolic compounds, acteoside (**118**) and arbutin (**107**), which are recognised as tyosinase inhibitors. Moreover, they evaluated the anti-tyrosinase activity of the isolated iridoids among which monomelittoside (**165**) and melittoside (**166**) showed IC50 values of 119.6 ± 2.2 and 163.1 ± 3.1 μg/mL respectively, while 5-allosyloxy-aucubin (**167**) inhibited the enzyme with a percentage of 22.4% at a concentration of 200 μg/mL. In addition, current works investigated the anti-tyrosinase activity of three subspecies of *S. cretica* (*S. cretica* subsp. *smyrnaea; S. cretica* subsp. *mersinaea; S. cretica* subsp. *vacillans*), reporting that the ethyl actetate extract was the most effective in the first two susbspecies (2.45 mg KAEs/g; 16 mg KAEs/g, respectively) [81,108]. Though, the methanol extract of *S. cretica* subsp. *vacillans* had the higher activity against tyrosinase (314.04 ± 2.05 mg KAE/g extract) [112].

#### *5.6. Anti-diabetic Activity*

Bahadori et al. (2018) evaluated the anti-diabetic activity of the extracts of *S. cretica* subsp. *smyrnaea* [81]. Specifically, the methanol extract demonstrated strong anti-diabetic activity against α-amylase (61.4 mg ACEs/g dry plant) and α-glucosidase (47.8 mg ACEs/g dry plant), following by ethyl acetate extract. They assumed that the above good properties were attributed to the phenolic constituents of the methanol extract since the anti-glucosidase activity is associated with caffeic acid, *trans*-cinnamic acid, and vanillin, whereas the amylase inhibitory activity is related to kaempferol and *p*-hydroxybenzoic acid. A year later, the anti-diabetic activity of the extracts of *S. cretica* subsp. *mersinaea* was studied, reporting that the ethyl acetate extract had best activity against α-amylase (396.50 mgACEs/g), while the methanol extract exerted strong activity against α-glucosidase (734 mg ACEs/g) [108]. Furthermore, the α-amylase inhibition of the methanol and water extract of *S. cretica* subsp. *vacillans* was evaluated, with the methanol extract exhibited stronger activity (433.99 ± 5.10 mg ACE/g extract) [112]. Currently, Pritsas et al. (2020) studied the anti-diabetic activity in silico of 17 isolated compounds from the cultivated *S. iva*, mentioning that stachysetin (**98**) interacted with five out of ten proteins implicated in diabetes [56]. This is the only study reported a pharmacological activity of this rare compound.

#### *5.7. Antimicrobial Activity*

Regarding the antibacterial activity, the *n*-butanol extract of *S. guyoniana* showed strong activity against *Staphylococcus aureus* (MIC = 32 ± 0.90 μg/mL) and *Enterobacter aerogenes* (MIC = 32 ± 0.70 μg/mL), while it was not active against *Pseudomonas aeruginosa* and *Morganella morganii* [155]. The ethyl acetate extract demonstrated the best inhibition against *Escherichia coli* (MIC = 64 ± 0.60 μg/mL), whereas it didn t show any activity against *P. aeruginosa* and *M. morganii*. Shakeri et al. (2019) reported the antimicrobial activity of the methanol extract of the aerial parts of *S. parviflora* which exerted the highest activity against the Gram-positive bacterium, *Bacillus cereus*, with a MIC of 0.12 mg/mL [64]. Furthermore, the antimicrobial activity of extracts of *S. thirkei* against different microorganisms were studied according to inhibition zone diameter and MIC value [84]. The acetone and methanol extract demonstrated good activity against *S. aureus*, *Streptococcus pyogenes* and *E. coli*. Intriguingly, *S. thirkeis* extracts were not active against *P. aeruginosa* (Gram-negative bacterium) and *Candida albicans* (yeast).

#### *5.8. Hepatoprotective*

The hepatoprotective property of the ethanol extract of *S. pilifera* was studied in carbon tetrachloride (CCl4)-induced hepatotoxicity in rats and indicated that this extract could act as hepatoprotective agent [158]. They assumed that this property might be also related to the strong antioxidant activity of the species. Later, Mansourian et al. (2019) exhibited the hepatoprotective and antioxidant activity of hydroalcoholic extract of *S. pilifera* on hepatotoxicity induced by acetaminophen (APAP) in male rats [159]. Precisely, the extract reduced hepatotoxicity by decreasing liver function markers/enzymes, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) and protein carbonyl (PCO) compared to the APAP group. It also diminished the oxidative stress through inhibiting protein oxidation and inducing the activity of glutathione peroxidase (GPX) enzyme. So, they assumed that this great activity was attributed to the antioxidant activity of this plant.

#### *5.9. Others*

Ruiu et al. (2015) investigated the phytochemical profile of the dichloromethane extract of *S. glutinosa* and studied the binding affinity to μ and δ opioid receptors (MOR and DOR) [107]. The extract showed an interesting binding affinity for MOR (Ki values of 10.3 μg/mL) and DOR (Ki values of 9.0 μg/mL), while xanthomicrol (**69**) demonstrated the strongest opioid binding affinity to both opioid receptors (Ki for MOR = 0.83 μM, Ki for DOR = 3.6 μM) with the highest MOR selectivity with a ratio Ki (DOR)/Ki (MOR) = 4.4. Notably, they reported that the existence of a further hydroxy group at the 3 position like in sideritoflavone (**70**) reduced the binding affinity for MOR (Ki = 18.5 μM), whereas the replacement of this group with a methoxy moiety, as in 8-methoxycirsilineol (**71**), eliminated the affinity for MOR (Ki > 50 μM). Furthermore, they evaluated the antinociceptive activity of xanthomicrol in an animal model (in mice) of acute pain (tail-flick test). In another study, the *n*-butanol extract of *S. mialhesii* exhibited significant anti-inflammatory activity *in vivo*, reducing the weight of edema: 52.03% induced by carrageenan in the rat's paw, whereas indomethacin (dose: 5 mg/kg; decrease 83.36%) was used as a reference drug [103]. In the same study, the *n*-butanol extract exerted antinociceptive effect at dose-dependent manner. Ramazanov et al. (2016) evaluated the wound healing activity of the extract of *S. hissarica* on rats, showing that the extract improved the healing process of linear skin wounds at an oral dose of 10 mg/kg [67]. Of note, the wound healing activity of the extract was more effective than the known drug methyluracil (2,4-dioxo-6-methyl-1,2,3,4 tetrahydropyrimidine), especially in case of alloxan induced diabetic animals. A study carried out by Iannuzzi et al. (2019) studied the antiangiogenic activity in two in vivo models (zebrafish embryos and chick chorioallantoic membrane assays) of the isolated compounds of the leaf extract of *S. ocymastrum*. The isolated compounds with the best antiangiogenic activity in both assays were β-hydroxyipolamiide (**173**) and ipolamiide (**174**) [123]. Recently, Lee et al. (2020) studied the anti-obesity and anti-dyslipidemic property of the roots powder of *S. sieboldii* in rats, following a high-fat and high-cholesterol diet (HFC) [161]. This powder demonstrated the anti-adipogenic and lipid-lowering effects through enhancing lipid metabolism.

Taken together all the above pharmacological studies, we could observe that these findings confirmed most of the traditional medicinal uses of *Stachys* spp. However, the present review unveiled that there are still species pharmacologically uncharted.

#### **6. Clinical Studies**

Through our literature survey, four clinical studies for the species *S. lavandulifolia* were revealed. The first clinical study carried out by Rahzani et al. (2013) reported the effects of the aqueous extract of the specific plant (dose; infusion from 3 g aerial parts of plant, twice daily) on the oxidative stress in 26 healthy humans, underlying that the participants demonstrated a significant reduction in oxidative stress [163]. In parallel, another randomized clinical trial (33 women) examined the effects of *S. lavandulifolia* and medroxyprogesterone acetate (MPA) in abnormal uterine bleeding (AUB) in PCOS [164]. This study exemplified that the infusion of the aerial parts of wood betony (dose; 5 g of plant in 100 mL boiling water; duration 3 months) showed a reduction of AUB, recommending its consumption for the treatment of AUB related to PCOS. They also mentioned that this result might be attributed to the flavonoid content of the plant and mainly to apigenin. In addition, Monji et al. (2018) evaluated on a clinical trial the therapeutic effects of standardized formulation of *S. lavandulifolia* on primary dysmenorrhea, indicating that the standardized capsules of plant's extract could diminuish

the menstrual pain, and might be recommended as an auxiliary therapy or an alternative remedy to nonsteroidal antiinflammatory drugs (NSAIDs) with fewer side effects in primary dysmenorrhea [165]. Recently, a double-blind randomized clinical study mentioned the analgesic activity of the herbal tea of *S. lavandulifolia* (10 g in 200 cc of boiling water) in 50 patients with migraine [166], showing the capability of this herbal tisane to decrease and also improve the pain intensity in these patients. In addition, Ashtiani et al. (2019) considered that the therapeutic properties of this plant associated with its rich phytochemical profile which include iridoids, flavonoids and phenylethanoid glucosides [166].

To sum up, the above clinical studies confirm the ethnomedicinal uses of *S. lavandulifolia* as a traditional medicine. Although these promising results, more clinical studies should be performed for obtaining data for diverse *Stachys* spp. As a future prospective, further studies should strengthen the research of bioavailability, dosage, toxicity and potential drug interactions in order to endorse the observed pharmacological activities of these plants.

#### **7. Toxicity**

*S. lavandulifolia* is popularly claimed as an abortifacient agent by Iranian women. The effect of its hydroalcoholic extract on fertility was investigated, revealing that the extract had a dose dependent abortifacient activity. Thus, its use during pregnancy may cause abortion and consequently, the plant should be considered as contraindicated or be used with caution [167]. In addition, the nephrotoxicity of the same extract was studied on male Wistar rats and a mild degeneration of renal tubular epithelial cell after one month was observed, while in the second month the histologic lesions were significantly more. However, further studies need to evaluate renal complications of this plant in human [168]. Moreover, the acute and subchronic toxicological evaluation of *S. lavandulifolia* aqueous extract in rats indicated that the high dose (2 g/kg) did not produce any symptoms of toxicity and there was no significant difference in body weights between the control and treatment groups of the animals [169].

#### **8. Conclusions**

In the present review, we attempted to describe in detail all the current knowledge and research advances of genus *Stachys*, focusing on pointing the significance of this genus as herbal supplement and medicine.

Taken together with all the analyzed studies in the current review, we categorized the used literature data into four categories according to their general characteristics; ethnobotanical (no of used studies: 48), phytochemical (no of used studies: 91), pharmacological (no of in vitro studies: 22, no of in vivo studies: 8 and 2 in silico study), clinical studies (no of used studies: 4) and reviews (no of used studies: 4). The general characteristics of the analyzed studies in the current review are showed in Table 31.


**Table 31.** General characteristics of the analyzed studies in the current review.

\* N.B. It could be found more than one type of data in the same article.

Several *Stachys* spp. have been used as traditional herbal medicines for thousands of years. Therefore, accumulating studies have been performed in order to explore the chemical compounds and the pharmacological properties of these species to validate their claimed ethnomedicinal properties. However, the present review data shows that there are still species phytochemically and pharmacologically unexplored. This comprehensive survey could serve as useful tool for scientists searching uncharted and interesting species to study, as well as it could be an informative guide for researchers aimed to identify leads for developing novel drugs. Although many pharmacological studies have demonstrated the great properties of these plants, only the clinical effects of one species have been investigated. As a result, further studies should be performed to validate the clinical efficiency of several *Stachys* spp. and if there is any potential toxicity. To be mentioned that there are still yet much to be done on the detailed documentation (safety and efficacy data) of genus *Stachys* in order to be developed an official monograph as a traditional use or well-established use plants.

**Author Contributions:** Conceptualization and supervision: H.S.; writing—original draft preparation: E.-M.T. & C.B.; writing—review and editing: all authors. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Quantification of the Ability of Natural Products to Prevent Herpes Virus Infection**

**Kunihiko Fukuchi 1, Hiroshi Sakagami 2,\*, Yoshiaki Sugita 3, Koichi Takao 3, Daisuke Asai 4,**†**, Shigemi Terakubo 4, Hiromu Takemura 4, Hirokazu Ohno 5, Misaki Horiuchi 6, Madoka Suguro 6, Tomohiro Fujisawa 6, Kazuki Toeda 6, Hiroshi Oizumi 6, Toshikazu Yasui <sup>7</sup> and Takaaki Oizumi <sup>6</sup>**


Received: 11 August 2020; Accepted: 30 September 2020; Published: 6 October 2020

**Abstract: Background:** Herpes simplex virus (HSV) is usually dormant and becomes apparent when body conditions decline. We investigated the anti-HSV activity of various natural and synthetic compounds for future clinical application. **Methods:** Mock- and HSV-infected Vero cells were treated for three days with various concentrations of samples. For short exposure, 100-fold concentrated virus were preincubated for 3 min with samples, diluted to normal multiplicity of infection (MOI), before the addition to the cells. Anti-HSV activity was evaluated by the chemotherapy index. **Results:** Alkaline extracts of the leaves of *Sasa* sp. (SE) and pine cone (PCE) showed higher anti-HSV activity than 20 Japanese traditional herb medicines (Kampo formulas), four popular polyphenols, and 119 chromone-related compounds. Exposure of HSV to SE or PCE for 3 min almost completely eliminated the infectivity of HSV, whereas much longer exposure time was required for Kakkonto, the most active Kampo formulae. Anti-HSV activity of PCE and Kakkonto could be detected only when they were dissolved by alkaline solution (pH 8.0), but not by neutral buffer (pH 7.4). Anti-HSV activity of SE and povidone iodine was stable if they were diluted with neutral buffer. **Conclusions:** The present study suggests the applicability of SE and PCE for treatment of oral HSV and possibly other viruses.

**Keywords:** Kampo formulae; alkaline extract of *Sasa* sp.; pine cone extract; povidone-iodine; HSV; HIV; loss of infectivity; solubilization method

#### **1. Introduction**

In the oral cavity, there are many viruses including norovirus, rabies, human papillomavirus, Epstein–Barr virus, herpes simplex viruses (HSVs), hepatitis C virus, and human immunodeficiency virus (HIV). Viral infections have been diagnosed using an oral sample (e.g., saliva mucosal transudate or an oral swab) based on the correlation of HIV anti-IgG/sIgA detection with saliva and serum samples [1]. Oral herpes viruses, HSV-1 and HSV-2, are very common and infectious, and debilitate patients, affect oral health, and have important psychological implications. The therapies currently used for the treatment of HSV infection are pharmacological, topical, systemic, or instrumental, occasionally with laser devices [2]. Many natural products have been investigated for their anti-HSV activity in vitro or in vivo. These include low molecular weight polyphenols [3–6], water-extracts [7–10] including Japanese traditional herb medicine (Kampo formulae) [11,12], and alkaline extracts [13] including a lignin-carbohydrate complex [14–17].

We have already reported the anti-HSV activity of five plant extracts, 13 tannin-related compounds determined by plaque assay [18], and the anti-HSV activity of eight licorice root extracts, 10 licorice flavonoids (including isoliquiritin apioside), five polymethoxyflavonoids (including tricin), and five polyphenols (including epigallocatechin gallate, chlorogenic acid, *p*-coumaric acid, curcumin, and resveratrol) determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) method [19]. Quantitative structure-activity relationship (QSAR) analysis of these 19 polyphenols and 1705 chemical descriptors demonstrated that their anti-HSV activity correlated well with six chemical descriptors that represent polarizability (MATS5p, GATS5p), ionization potential (GATS5i), number of ring systems (NRS), atomic number (J\_Dz(Z)) and mass (J\_Dz(m) (r2 = 0.684, 0.627, 0.624, 0.621, 0.619, and 0.618, respectively, *p* < 0.0001) [19]. However, most of lower molecular weight polyphenols showed very low anti-HSV-activity.

In the present study, we report the in vitro anti-HSV activity of 20 Kampo formulas and alkaline extracts of the leaves of *Sasa* sp. (SE) and pine cone combined with dextrin (PCE), representative polyphenols, and a total of 119 chromones, esters, and amides [20–26], synthesized from chromone (to search for new type of anti-HSV agents), a back-bone structure of flavonoids, together with positive control acyclovir [27], representative polyphenols (resveratrol, *p*-coumaric acid, and curcumin used as negative controls) [19] and povidone iodine (PVP-I), a popular gargle [28].

Since gargling time with mouth wash is usually a minute order, we investigated whether short exposure of HSV (1.5 or 3 min) is enough to inactivate HSV. Since PCE and the Kampo formula contain many acidic substances such as a lignin-carbohydrate complex and its degradation products, it is expected that an alkaline solution may be useful to extract the active substances in higher yield compared with the neutral buffer, although some elevation of degradation would be inevitable. Therefore, we also compared the anti-HSV activity and its stability using either an alkaline solution (1.39% NaHCO3, pH 8.0) or a neutral buffer [phosphate-buffered saline (PBS), pH 7.4].

#### **2. Materials and Methods**

#### *2.1. Materials*

The following chemicals and reagents were obtained from the indicated companies: Eagles minimum essential medium (MEM) (Gibco BRL, Grand Island, NY, USA); fetal bovine serum (FBS), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), resveratrol, azidothymidine (AZT), 2',3'-dideoxycytidine (ddC) (Sigma-Aldrich Inc., St. Louis, MO, USA); dimethyl sulfoxide, dextran sulfate (DS) (5 kDa) (Wako Pure Chemical Ind., Ltd., Osaka, Japan); acyclovir, curcumin, trans *p*-coumaric acid (Tokyo Chemical Industry Co. Ltd., Tokyo, Japan); tricin (Carbosynth Ltd., Berkshire, UK); curdlan sulfate (79 kDa) (Ajinomoto Co., Inc., Tokyo, Japan); and PVP-I (Showa Seiyaku Co. Ltd., Tokyo, Japan). Twenty Kampo formula (Table 1) were provided by Tsumura & Co, Tokyo, Japan. Culture plastic dishes and plates (96-well) were purchased from Becton Dickinson Labware (Franklin Lakes, NJ, USA).


**Table 1.** Twenty Kampo formulas used in this study.

#### *2.2. Preparation of Sasa sp. (SE)*

SE was prepared by iron ion substitution, alkaline extraction, and neutralization/desalting (Figure 1A). Lyophilization and measurement of the dry weight of SE showed that it contained 58.2 ± 0.96 mg solid materials/mL [29]. The components of SE are shown in our previous review article [30].

**Figure 1.** Scheme for large scale preparation of *Sasa* sp. (SE) (**A**) and pine cone extract (PCE) (**B**).

#### *2.3. Preparation of Pine Cone Extract (PCE)*

Pine cone extract was prepared by modification of the original method of preparation of the lignin-carbohydrate complex [31,32]. In brief, pine cone of *Pinus parviflora* Sieb et Zucc. was washed by hot water extract to remove contaminants and hot-water extractable materials, and then extracted with 0.15 N NaOH to obtain the lignin-carbohydrate complex. The lignin-carbohydrate complex was recovered by ethanol precipitation, and separated from salts and fat-soluble degradation products such as phenylpropanoids. Nine volumes of dextrin were added and spry dried to yield PCE (Figure 1B).

#### *2.4. Preparation of Chromones, Esters, and Amides*

Twenty four 2-azolylchromone derivatives (E) were synthesized by the conjugated addition reaction of 3-iodochromone derivatives with various azoles [20]. Seventeen 3-benzylidenechromanone derivatives were synthesized by base-catalyzed condensation of the corresponding 4-chromanone with substituted benzaldehyde derivatives [21]. Fifteen chalcone derivatives were synthesized by base-catalyzed condensation of the corresponding acetophenones with various benzaldehyde derivatives [22]. Ten cinnamic acid phenethyl esters were synthesized by the condensation of cinnamic acid and its analogs such as caffeic acid, ferulic acid, and *p*-coumaric acid with the corresponding phenethyl alcohols [23]. Ten 3-flavene derivatives were synthesized by the reductive intramolecular cycloaddition reaction of 2-hydroxychalcone derivatives [22]. Eleven piperic acid amides were synthesized by the condensation of the acid chloride of piperic acid with various amines. Piperic acid was prepared by alkaline hydrolysis of piperine [24]. Eighteen 2-styrylchoromone derivatives were synthesized by base-catalyzed condensation of the corresponding 2-methylchromones with selected benzaldehyde derivatives [25]. Fourteen 3-styrylchoromone derivatives were synthesized by Knoevenagel condensation of the corresponding 3-formylchromones with various phenylacetic acid derivatives [26]. All compounds were dissolved in DMSO at 40 mM and stored at −20 ◦C before use.

#### *2.5. Assay for Anti-Herpes Simplex Virus (HSV) Activity*

We dissolved the samples using the following three methods. (i) Method 1: Sample was dissolved at 1 mg/mL with culture medium (MEM + 10% FBS) and then sterilized by passing through a Millipore filter (pore size: 0.45 μm); (ii) Method 2: Sample was dissolved at 60 mg/mL with 1.39% NaHCO3 (pH 8.0) and then diluted to 3 mg/mL with medium and filtered; (iii) Method 3: Sample was dissolved at 100 mg/mL with phosphate-buffered saline (PBS, pH 7.4) or 1.39% NaHCO3, vortexed, and shaken overnight at 4 ◦C. After centrifugation, the supernatant was collected and then filtered (Figure 2A).

**Figure 2.** Experimental protocol. (**A**) Long treatment. (**B**) short treatment.

For the long treatment schedule (upper column in Figure 2B), Vero cells, isolated from the kidney of African green monkey (*Cercopithecus aethiops*) were infected with HSV-1 (multiplicity of infection (MOI) = 0.01). HSV-1 and test samples were mixed and stood for 20 min, and the mixture was then added to the adherent Vero cells. After incubation for three days, the relative viable cell number was determined by the MTT reagent.

For the short treatment schedule (lower column in Figure 2B), 100-fold concentrated HSV (MOI = 1) was mixed with samples and stood for 1.5, 3, or 20 min. Then, virus concentration was reduced to 1/100

(MOI = 0.01), added to the cells and incubated for three days. Mock-infected cells were first treated for 1.5, 3, or 20 min with the same concentrations of test samples without HSV, then the sample was removed by suction, washed once with PBS, and incubated for three days in the fresh culture medium. The viability of both HSV-infected and mock-infected cells was determined by the MTT method as described above.

From the dose-response curve, 50% cytotoxic concentration (CC50) in mock-infected cells, and the 50% effective concentration (EC50) in HSV-infected cells were determined. EC50-I was defined as the concentration at which the viability was restored to the midpoint between that of HSV-infected cells and that of mock-infected cells. EC50-II was defined as the concentration at which the viability was restored to 50% of that of the mock-infected cells. The anti-HSV activity was evaluated by the selectivity index (SI-I and SI-II), which was calculated using the following equation: SI-I = CC50/EC50-I; SI-II = CC50/EC50-II (Figure 3).

**Figure 3.** Calculation of anti-herpes simplex virus (HSV) activity.

#### *2.6. Assay for Anti-Human Immunodeficiency Virus (HIV) Activity*

Human T-cell leukemia virus I (HTLV-I)-bearing CD4-positive human T-cell line MT-4, established by Dr. Miyoshi [33], was cultured in RPMI-1640 medium supplemented with 10% FBS and infected with HIV-1IIIB at a multiplicity of infection (MOI) of 0.01. HIV- and mock-infected MT-4 cells (3 <sup>×</sup> <sup>10</sup><sup>4</sup> cells/96-microwell) were incubated for five days with different concentrations of extracts and the relative viable cell number was determined by the MTT assay. The concentration that reduced the viable cell number of the uninfected cells by 50% (CC50) and the concentration that increased the viable cell number of the HIV-infected cells to 50% of the control (mock-infected, untreated) cells (EC50) was determined from the dose-response curve with mock-infected and HIV-infected cells, respectively. The anti-HIV activity was evaluated by the selectivity index (SI), which was calculated using the following equation: SI = CC50/EC50 [34,35]. Since the viable cell number of HIV-treated cell reached the baseline (zero), the EC50-I and SI-I were nearly identical to EC50-II and SI-II, respectively.

#### *2.7. Statistical Treatment*

Experimental values were expressed as the mean ± standard deviation (SD). The statistical significance between the groups was assessed with analysis of variance (ANOVA) followed by Dunette's multiple comparison test. A *p* value less than 0.05 was considered significant.

#### **3. Results**

#### *3.1. Establishment of Assay Condition for Anti-HSV Activity*

We have reported that the anti-HSV activity of most lower molecular weight polyphenols as assessed by SI value (CC50/EC50) was very low (SI < 1) when compared with the positive control (acyclovir and tricin) (SI > 27.3, 7.1) [19]. We conducted the accurate determination of anti-HSV activity of natural products with low anti-HSV activity using two different plates: 96-well (for CC50 determination by the MTT method) and 6-well plates (for EC50 determination by plaque assay) may be difficult, especially for short exposure experiments that require quick medium change. Therefore, it was necessary to first investigate whether the infectivity of HSV-1 measured by the MTT method with the 96-well plate correlated with the plaque count of the 6-well plate under the light microscopy. We found that this was the case. With the dilution fold of the virus solution increased, the viable cell number was increased, reaching the plateau level where no plaque formation was observed (Figure 4A). We found that the virus titer of MOI = 0.01 was much better than MOI = 0.1 for the quantitative determination of the anti-HSV of acyclovir (ACV), one of the positive controls used in this study (Figure 4B). When we used MOI = 0.1, accurate evaluation of the viable cell number was difficult due to the presence of too much virus. Therefore, we used the fixed MOI = 0.01 during the cell treatment period to maintain the viability of HSV-infected Vero cells at 20 ~ 40% (Figure 3). The selection of three days was the best incubation time for the determination of anti-HSV activity. We had to use 100-fold concentrated HSV solution (MOI = 1) to treat the virus with extremely higher concentrated samples. We reduced the MOI 100-fold during the days of cell culture (MOI = 0.01).

**Figure 4.** (**A**) Confirmation of correlation of optical density measurement with the MTT method and plaque counting. Vero cells (10,000 cells) were inoculated on a 96-microwell plate and incubated overnight at 37 ◦C. HSV-1 solution at the indicated dilution fold was then added. After incubation for three days, viable cell number (absorbance at 595 nm) was determined by the MTT method, and plaque formation was counted by light microscopy. (**B**) Effect of different MOI on the anti-HSV activity of ACV. Each value represented as mean ± S.D. was determined (n = 3). Significant difference between MOI = 0.01 and MOI = 0.1 (*p* < 0.05).

#### *3.2. Anti-HSV Activity of Natural Products*

#### 3.2.1. Hot-Water Extract (Kampo Formula) and Alkaline Extracts (SE, PCE)

It was important to first establish the method for dissolving the Kampo formulas. As a first step, we directly mixed 20 Kampo formulas with culture medium (MEM + 10% FBS), and then sterilized them by passing through a Millipore filter (Method 1 in Figure 2A). Both mock-infected and HSV-infected cells were incubated for three days without or with various concentrations of samples, and the viable cell number was determined by the MTT method. From the dose-response curve, we determined the 50% cytotoxic concentration (CC50) and 50% protective concentration (EC50-I, EC50-II) and maximum cell recovery (%) (MCR). The anti-HSV activity was assessed as the SI value (SI-I = CC50/EC50-I or SI-I = CC50/EC50-II) (Figure 3). It was unexpected that any Kampo formula, except for S19 (SI-I > 2.2; SI-II > 5.9) did not reduce the cytopathic effect of HSV infection (Supplementary Figure S1). We thought that the failure to detect anti-HSV activity may be due to the interaction of Kampo ingredients and medium components. Based on these data, we did not choose method I to solubilize the Kampo formula. We found that higher anti-HSV activity of Kampo formulas was recovered by dissolving them with 1.39% NaHCO3 (pH 8.0) than with PBS (pH 7.4) (Supplementary Table S1, Supplementary Figure S2). Therefore, we dissolved all Kampo formula with 1.39% NaHCO3 to make the initial concentration of 60 mg/mL, diluted with culture medium to make a 3 mg/mL solution, and sterilized by passing through a Millipore filter (Method 2 in Figure 2A).

Mock-infected and HSV-infected cells were incubated with increasing concentrations of samples and determined for viability (Figure 5). Among the 20 Kampo formulas (S1 ~ S20), Kakkonto (S5) showed the highest anti-HSV activity (SI-1 > 5.2; SI-II > 5.4; MCR = 78%), followed by Yokukansan (S9) (SI-I > 1.0; SI-II > 1.6; MCR = 66%), Yokuininto (S12) (SI-II = 1.3; MCR = 52%) > Jumihaidokuto (S11) (SI-II = 1.1; MCR = 52%). However, their anti-HSV activity was much lower than that of the alkaline extract of *Sasa* sp. (SE) (Supplementary Table S2) (SI-1 = 4.5; SI-II = 6.8; MCR = 90%), alkaline extract of pine cone of *Pinus parviflora* Sieb. et Zuc. (PCE) (SI-1 = 13.1; SI-II = 14.7; MCR = 82%) and acyclovir (ACV) (SI-1 > 23.1; SI-II > 27.3; MCR = 108%) (Table 2).

#### 3.2.2. Polyphenols and Chromone-Related Compounds

Among the four polyphenols, resveratrol (MCR = 20%), *p*-coumaric acid (MCR = 43%), and curcumin (MCR = 42%) showed little or no anti-HSV activity, whereas tricin (SI-II = 7.1; MCR = 68%) showed weak anti-HSV activity (Figure 5, Table 2).

Among the 119 chromone-related compounds, only 2-(1*H*-pyrazol-1-yl)-4*H*-1-benzopyran-4-one (2a), 2-(1*H*-imidazol-1-yl)-6-methoxy-4*H*-1-benzopyran-4-one (3c), (3*E*)-2,3-dihydro-3-[(4-hydroxyphenyl) methylene]-7-methoxy-4*H*-1-benzopyran-4-one (14), (2*E*,4*E*)-5-(3,4-methylenedioxyphenyl)-2,4 pentadienoic acid (4-hydroxy-3-methoxyphenyl)methyl ester (2), 2-[(1*E*)-2-(4-fluorophenyl)ethenyl]-6 methoxy-4*H*-1-benzopyran-4-one (8), and 2-[(1*E*)-2-(3,4-dimethoxy)ethenyl]-6-methoxy-4*H*-1-benzopyran -4-one (12) showed weak anti-HSV activity (Figure 5). Chromone derivatives with higher antitumor activity (assessed with tumor-specificity determined by the ratio of mean CC50 against four human oral squamous cell carcinoma cell lines (Ca9-22, HSC-2, HSC-3, HSC-4) to that for three normal oral cells such as human gingival fibroblasts, human periodontal ligament fibroblasts, and pulp cells (indicated by red color) showed no anti-HSV activity. Similarly, chromones with higher anti-HSV activity did not have higher antitumor activity (Supplementary Table S3).

**Figure 5.** Anti-HSV activity of Kampo formulas, SE, PCE, and acyclovir (ACV). Mock-infected (-) and HSV-infected (•) cells were treated for three days and viable cell number [%t of control (untreated, uninfected cells)] were determined. Each value is represented as mean ± S.D. (n = 3).




**Table 2.** *Cont.*

Data were derived from Supplementary Figure S2 (for all 20 Kampo formula) and from Figure 4. (2a), 2-(1*H*-pyrazol-1-yl)-4*H*-1-benzopyran-4-one; (3c), 2-(1*H*-imidazol-1-yl)-6-methoxy-4*H*-1-benzopyran-4-one; (14), (3*E*)-2,3-dihydro-3-[(4-hydroxyphenyl)methylene]-7-methoxy-4*H*-1-benzopyran-4-one; (2), (2*E*,4*E*)-5-(3,4 methylenedioxyphenyl)-2,4-pentadienoic acid (4-hydroxy-3-methoxyphenyl)methyl ester; (8), 2-[(1*E*)-2-(4 fluorophenyl)ethenyl]-6-methoxy-4*H*-1-benzopyran-4-one; (12), 2-[(1*E*)-2-(3,4-dimethoxy) ethenyl]-6-methoxy -4*H*-1-benzopyran-4-one. Supplementary Tables S2 and S3 show the anti-HSV activity of SE (assayed 52 times) and a total of 119 chromone derivatives, esters, and amides.

#### *3.3. Augmentation of Antiviral Potential of Alkaline Extracts by Reducing the Treatment Time*

#### 3.3.1. Rapid HSV Inactivation by SE and PCE

Since SE and PCE showed approximately 10-fold higher anti-HSV activity (SI-II = 6.8, 14.7) than the twenty Kampo formulas (mean SI-II < 1.1), 6-fold higher than four polyphenols (mean SI-II = 1.9), and 6-fold higher than five of the most potent chromones (mean SI-II = 1.7) (Table 2), we next investigated whether short exposure of HSV to these samples could instantly reduce the infectivity.

Exposure of HSV with SE (A) (1 or 3 mg/mL), PCE (C) (1 or 3 mg/mL) as well as povidone iodine (B) (2.33 or 7 mg/mL) rapidly eliminated its infectivity within 3 min, whereas Kampo preparation (S5) (D) took 20 min to express HSV inactivation (Figure 6).

The HSV inactivation effect of SE was reproducibly diminished by dilution with 1.39% NaHCO3 (pH 8.0), rather than (PBS, pH 7.4) in four independent experiments (compare left and right column in Figure 6A). Povidone iodine showed similar instability under alkaline conditions (compare left and right column in Figure 6B). On the other hand, the anti-HSV activity of powders such as PCE and Kakkonto (S5) was enhanced more than 21.4 (=3/0.14) and 1.25-fold (=3/2.4), respectively, when they were first dissolved with 1.39% NaHCO3 rather than PBS (Figure 6C,D).

**Figure 6.** Effect of short exposure of HSV to SE (means of four independent experiments) (**A**), povidone iodine (**B**), PCE (**C**) and Kakkonto (S5) (**D**). A 100-fold higher titer of HSV was exposed to these samples for 0, 1.5, 3, or 20 min, and then added to the cells after dilution of 100-fold. After incubation for three days, viable cells were determined. Each value represents mean ± S.D. (n = 3).

We next investigated the cytotoxicity (measured by CC50) and protective effect (measured by EC50) of short exposure (3 min) of SE (A, B), PCA (C, D) and PV-I (E, F) (Figure 7). From the dose-response

curve, we could calculate the SI values (Table 3). It is apparent that SE showed 2- to 3-fold higher anti-HSV activity when it was diluted with PBS (SI-I = 26.1, SI-II = 31.6) (B), rather than with NaHCO3 (SI-I = 9.4, SI-II = 11.4) (A). On the other hand, PCE showed higher anti-HSV activity when it was dissolved by NaHCO3 [SI-I > 222, SI-II > 322, MCR (maximum cell recovery) = 101.3%) (C), than by PBS (SI-I > 62.5, SI-II > 76.9, MCV = 26.8%) (D). However, we could not calculate the SI value of Kakkonto (S5) due to the lower protection effect. Povidone iodine (PVP-I) showed much lower anti-HSV activity, whenever diluted by NaHCO3 (SI-I = 1.3, SI-II = 2.3) (E) or PBS (SI-I = 2.0, SI-II = 3.1) (F) (Figure 7).

**Figure 7.** Dose-response curve of cytotoxicity and protective effect of short exposure (3 min) to SE (**A**, **B**), PCE (**C**,**D**), and PVP-I (**E**,**F**). Samples were dissolved and diluted either in 1.39% NaHCO3 or PBS. HSV and Vero cells were preincubated for 3 min, and chased into fresh medium, and the viable cell number was determined. The indicated concentrations in the abscissa is the concentration at the time of contact to samples for 3 min. Each value is represented as mean ± S.D. (n = 3).

#### 3.3.2. Rapid HIV Inactivation by SE

We investigated whether the short exposure with SE enhanced the anti-HIV activity. We found that exposure of HIV to SE for as little as 1 ~ 30 min quickly inactivated the virus (Figure 8A).


**Table 3.** Quantification of HSV inactivation by short exposure to SE and PCE.

**Figure 8.** Rapid inactivation of HIV by SE. (**A**) Effect of preincubation time. A 20-fold higher titer of HIV (MOI = 0.2) was exposed to 0.2 mg/mL SE for 0, 1, 3, 10, or 30 min, and then added to the cells after dilution of 20-fold to make the final MOI = 0.01. After incubation for five days, viable cells were determined. Each value represented as the mean ± S.D. was determined (n = 3). HIV infection significantly reduced the viable cell number (*p* < 0.05). (**B**) Effect of long (five days) and short (10 min) exposure of HIV to SE, and popular anti-HIV agents on the cell viability. Each value represented as the mean ± S.D. was determined (n = 3).

Four popular anti-HIV agents (AZT, ddC, DS, CRDS, used as the positive controls) showed potent anti-HIV activity (SI = 5082, 1913 > 29,485, 4666), verifying this system for measuring the anti-HIV activity (lower panel in Figure 8B). Compared with regular long exposure (five days) (SI = 95), short exposure to SE more effectively inactivated the virus (SI > 560, >369), yielding more than a 4 ~ 6-fold increase (upper panel in Figure 8B). We repeated the same experiment with more wider dose ranges, and found that 10 min exposure of HIV with SE showed approximately 20-fold increase of anti-HIV activity (25.1 and 16.6-fold in two different assays) when compared with regular longer exposure (Table 4).


**Table 4.** Enhancement of anti-HIV activity of SE by shortening the treatment time.

#### **4. Discussion**

The present study demonstrated that the anti-HSV activity of the Kampo formula solely depended on the solubility. Kampo formulas are all powder, and therefore they have to be dissolved well and sterilized before treatment. We found that when they were directly dissolved with medium, no anti-HSV activity (Method 1) (Supplementary Table S1) and anti-HIV activity [36] were detected. On the other hand, when they were dissolved in alkaline solution such as 1.39% NaHCO3 (pH 8), some anti-HSV activity was recovered (Method 2). However, when they were dissolved with neutral buffer such as PBS (pH 7.4), no anti-HSV activity was recovered. Using Method 3, contact with Kakkonto for 20 min reduced the infectivity of HSV, consistent with previous reports of the anti-HSV activity of Kakkonto [12]. We recently reported that most Kampo formulas including Kakkonto showed protection against cisplatin and amyloid-β-induced neurotoxicity [36].

We have previously separated various polysaccharide fractions of the pine cone of *Pinus parviflora* Sieb et Zucc by successive hot water and alkaline extractions. Hot water extract contains Fr. I (neutral), Fr. II (uronic acid-rich), and Fr. V (tightly bound to diethylaminoethyl cellulose (DEAE) cellulose chromatography). Alkaline extract contains Fr.VI (acid-precipitable), and Fr. VII, VIII, and IX (step-wise precipitated by increasing amounts of ethanol). All of these fractions contain glucose, mannose, galactose, and arabinose or fucose as the main component of polysaccharide. Chemical analysis (infrared spectroscopy (IR), nuclear magnetic resonance (NMR), thin layer chromatography (TLC)) identified Frs. V, VI, VII, VIII, and IX as a lignin-carbohydrate complex [31]. We found that only Frs. V, VI, VII, VIII, and IX showed potent anti-HIV activity whereas neutral and acidic polysaccharides (Frs. I and II) were inactive [32]. It is reasonable that PCE was richer in lignified material than the Kampo formula, and showed higher anti-HSV activity. Removal of dextrin from PCE power may further increase the specific activity of anti-HSV.

SE itself is an alkaline solution containing a lignin-carbohydrate complex and its degradation product such as *p*-coumaric acid, which has no anti-HSV activity. We found that dilution of SE with 1.39% NaHCO3 reproducibly reduced the anti-HSV activity, possibly due to the degradation of the lignin-carbohydrate complex under alkaline condition. We reported diverse biological activity of SE (anti-inflammatory, antiviral, antibacterial, anti-UV, and anti-halitosis activity, and synergism with acyclovir or vitamin C), some of which overlapped that of the lignin-carbohydrate complex [13,16,17,29,32,37–45] (Table 5).


**Table 5.** Diverse biological activity of SE and similarity with the lignin-carbohydrate complex.

There are three commercially available products of alkaline extract of *Sasa* sp. (products A, B and C). SE (Product A) contains Fe (II)-chlorophyllin, whereas products B and C contain Cu (II)-chlorophyllin and less lignin-carbohydrate complex. Product C is supplemented with ginseng and pine (*Pinus densiflora*) leaf extracts. We found that SE (Product A) exhibited higher anti-HIV, anti-UV, and hydroxyl radical-scavenging activities compared to those of products B and C [46]. This finding further strengthens that major biological principles in SE may be the lignin-carbohydrate complex. The lignin-carbohydrate complex can be extracted by alkaline solution, but, once isolated by the alkaline solution, may be unstable under alkaline conditions [47–49], and gradually decompose into phenylpropanoid monomers and oligomers with little or no antiviral activity [50,51]. Based on these unique biological activities, we manufactured various medicines, cosmetics, toiletries, supplements, and foods using SE (Figure 9). If we could remove lower molecular weight degradation products that have essentially no antiviral activity, specific activity of SE may be further elevated. Considering that saliva is neutral with a pH of 7.2 ~ 7.3 [52] or 7.0 ~ 7.2 [53], SE may be stable in the oral cavity.

It was unexpected that lower molecular weight polyphenols such as resveratrol, curcumin, and *p*-coumaric acid had little or no anti-HSV activity. This may be due to their potent cytotoxicity against Vero cells. We have previously reported that tricin, but not the other four polymethoxyflavonoids (3,3',4',5,6,7,8-heptamethoxyflavone, nobiletin, tangeretin, and sudachitin), showed potent anti-HSV activity, suggesting the importance of the 3D-structure of these polymethoxyflavonoids for expressing anti-HISV activity [19]. It remains to be investigated whether short-term exposure of HSV to a higher concentration of these polyphenols may inactivate HSV or not.

We recently found that many chromone derivatives, esters, and amides showed much higher cytotoxicity against human oral squamous cell carcinoma cell lines when compared with human normal mesenchymal oral cells (gingival fibroblast, periodontal ligament fibroblast, pulp cells). Their tumor-specificity exceeded that of the lower molecular weight polyphenols. Furthermore, they showed much less normal keratinocyte toxicity than conventional anticancer drugs. Our preliminary study demonstrated that some of them alleviated the HSV-induced cytopathic effects. However, there was no correlation between their tumor-specificity and anti-HSV activity (Supplementary Table S3).

Recently, povidone iodine has been broadcasted to improve the symptoms of corona virus-infected patients in TV ASAHI super-channel in Japan on August 5, 2020. However, many authorities of medical sciences have shown cautionary stance, since it may kill the good bacteria that protect the mouth and reduce thyroid function. The present study showed that it also rapidly reduced the infectivity of HSV. Further study of the safety of this gargle as an antiviral agent should be performed.

We have previously demonstrated that 125I-labeled lignin-carbohydrate complex bound tightly to the influenza virus with sucrose gradient centrifugation [54]. The lignin-carbohydrate complex [55] and tannic acid [56] significantly inhibited the adsorption of 3H-labeled HSV to Vero cells [18]. Anti-HSV activity of these substances was much greater when they were added during virus adsorption to the cells rather than before and after adsorption [18,52]. These data suggest that the target of these substances may be virus or cell surface components. Since the lignin-carbohydrate complex significantly enhanced the expression of dectin-2 [53], possible interactions with the cell surface receptor should be investigated. Further study is necessary to identify the antiviral mechanisms of these substances.

**Figure 9.** Medicines, cosmetics, toiletries, supplements, and foods manufactured from SE. Right column: Yatsugatake mountain peak (**top**), factory of the Daiwa Biological Research Institute Co. Ltd., Chino, Nagano, Japan (**middle**), and the botanical garden in the "*Sasa* Rikyu (imperial villa)" (**bottom**).

#### **5. Conclusions**

The present study demonstrated for the first time that:


*Medicines* **2020**, *7*, 64

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2305-6320/7/10/64/s1, Figure S1: Method I. Direct mixing of Kampo preparations with culture medium resulted in low recovery of anti-HSV activity, Figure S2: Method 2: Weak anti-HSV activity was detected in S5, S9, S11 and S12, Table S1: Higher anti-HSV activity of Kampo formulas was recovered by dissolving with 1.39% NaHCO3 than with PBS, Table S2. Anti-HSV activity of SE from 52 experiments, Table S3. Anti-HSV activity of chromones, esters, and amides (119 compounds).

**Author Contributions:** Conceptualization, formal analysis, investigation and funding acquisition, K.F., H.S., Y.S., D.A., S.T. and K.T (Koichi Takao).; Methodology and resources, H.O. (Hiroshi Oizumi), M.H., M.S., T.F. and K.T. (Koichi Takao); Data curation, F.K., D.A., S.T., and H.S.; Writing—original draft preparation, H.S.; Writing—review and editing, F.K.; Supervision, H.S. and K.T. (Kazuki Toeda).; Project administration, T.Y., K.T. (Kazuki Toeda), H.T., H.O. (Hirokazu Ohno), and T.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by KAKENHI from the Japan Society for the Promotion of Science (JSPS) (No. 16K11519, No. 20K09885) (H.S); the Showa University Graduate School Research Fund (K.F); thee Meikai University Miyata Research Fund B (H.S); the Josai University Research Fund (Y.S., K.T.(Koichi Takao)); the Joint Research Fund of Meikai University; Daiwa Biological Co. Ltd.; and Maruzen Pharmaceuticals Co., Ltd. (H.S.).

**Acknowledgments:** The authors thank the Tsumura and Co. for providing the 20 Kampo formulas.

**Conflicts of Interest:** M.H., M.S., T.F., K.T. (Kazuki Toeda), H.O. and T.O. come from Daiwa Biological Research Institute Co., Ltd., Kanagawa, Japan. H.O. comes from Maruzen Pharmaceuticals Co., Ltd., Hiroshima, Japan. Corresponding authors (H.S.) received the financial support of research funds from Daiwa Biological Research Institute and Maruzen Pharmaceuticals. The authors confirm that such financial supports have not influenced the outcome of the experimental data. The other authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Review*
