Screening the dermatological potential of Plectranthus species components: antioxidant and inhibitory capacities over elastase, collagenase and tyrosinase

Abstract A series of Plectranthus spp. plant extracts (aqueous, acetonic, methanolic and ethyl acetic) obtained from eight different species, and previously isolated compounds (ranging from polyphenols, diterpenes and triterpenes), were assayed for in vitro inhibition of the skin-related enzymes tyrosinase, collagenase and elastase, and for studying their antioxidant properties. The ethyl acetic extracts of P. grandidentatus and P. ecklonii registered the highest antioxidant activity, whereas acetonic, methanolic and ethyl acetic extracts of P. ecklonii, P. grandidentatus, P. madagascariensis and P. saccatus concerning the enzymatic inhibition assays revealed high anti-tyrosinase and anti-collagenase activities. From the isolated compounds tested, abietane diterpenes and triterpenes were highly active against tyrosinase and elastase activity. Overall, the experimental results showed the powerful antioxidant and inhibitory action on skin-related enzymes tyrosinase, collagenase and elastase of Plectranthus spp. extracts and/or isolated compounds, supporting their further research as bioactive metabolites against skin sagging and hyperpigmentation in cosmetic and pharmaceutical formulations.


Introduction
Skin care is a growing concern for the consumer, who is increasingly aware of its maintenance for the preservation of favourable aesthetic appearance, and on its direct impact on the prevention of many skin disorders 1 . In the last years, more attention has been devoted to skin as a mirror for general health condition. Dermatological "links" have been reported to the main cardiovascular pathologies [2][3][4] , metabolic disease 5 , renal disease 6 , obesity 7,8 and, of course, primary ageing 9,10 . Neurodegenerative diseases such as dementia, Alzheimer's and Parkinson's diseases have also been related to dermatological disorders 11,12 . All these examples illustrate how skin provides new directions in the search of new indicators for diseases prevention, detection, following and treatment.
The origin of many of these skin disorders has been linked to oxidative stress 13,14 , which is commonly promoted by chemicals, microorganisms and ultraviolet (UV) solar radiation [15][16][17] . Exposure to UV radiation can be particularly adverse for people with low melanin production, such as Caucasians, due to the fact of generating reactive oxygen species (ROS), such as superoxide anions, hydrogen peroxide (H 2 O 2 ) and hydroxyl radicals. ROS can react with cellular lipids, deoxyribonucleic acid (DNA) and proteins to evoke lipid peroxidation, enzyme denaturation, mutagenesis and eventually cell death 16,18 . One of the determinants of extrinsic ageing processes are these oxidative reactions, which are catalysed in the presence of redox-active metal ions like Fe 2þ and Cu 2þ , commonly found in the biological environment. All these factors, particularly UV, can modify the extracellular matrix (ECM) breaking down structural proteins such as collagen and elastin (Figure 1), which are the major components of the dermal tissue, essential for skin's structural stability and biomechanics 17,19 . ROS have been shown to upregulate the expression of several proteinases, including matrix metalloproteinases (MMP) and serine proteases, such as collagenase and elastase 14 . They were also reported to be involved in chronic inflammatory skin diseases, neurodegenerative disorders as well as in extrinsic ageing 9,17,20 .
The search for new and more effective molecules from different sources has clearly been focussed on these potential mechanisms, with the goal of preventing oxidation, thus reducing the production of ROS and protecting the tissues 14,21 . Plants from the genus Plectranthus, belonging to the Lamiaceae family, are widely distributed across the warm and tropical areas of Africa, Asia and Oceania, and were likely brought to Mediterranean areas in the 16th century during the period of Portuguese discoveries 22,23 . Rich in essential oils and in mono-and sesquiterpenes [24][25][26] , extracts from these plants are widely used in traditional medicine with different ethnopharmacological uses, including antimicrobial, analgesic, antipyretic, anti-inflammatory and antitumoral. Moreover, they have been applied in many skin disorders including wound healing [25][26][27][28] . Diterpenoids (abietane, labdane, kaurane, and clerodane skeleton), together with triterpenoids, phytosterols, phenolic acids, flavonoids and other polyphenolic compounds are the specific compounds of interest within this genus, according to their displayed effects compiled in the literature [28][29][30] . Polyphenols, such as rosmarinic acid which is ubiquous to all Plectranthus spp., and chlorogenic acid, obtained primarily from P. saccatus, and usually present in the aqueous extracts [31][32][33][34] , are powerful antioxidants 17,22 . Phenol groups in polyphenolic compounds are able to accept an electron to form relatively stable phenoxyl radicals, thereby disrupting radical chain oxidation reactions in cellular components 35 . Abietane diterpenes from P. ecklonii, P. madagascariensis and P. grandidentatus are able to chelate the 2,2diphenyl-1-picrylhydrazyl (DPPH) radical comparably to positive control quercetin, a natural flavonoid 23,33 .
Another consequence of ROS overproduction on Caucasian skin is hyperpigmentation, frequently leading to lentigine lesions, freckles and melanoma 21,36 . All of these result from the stimulation of tyrosinase, a binuclear copper rate-limiting oxidase acting in the biosynthesis of melanin from L-tyrosine 21,[36][37][38] . Overall, the chelating ability of Plectranthus spp., notably of P. ecklonii, P. madagascariensis and P. grandidentatus secondary metabolites, seems to be useful in a synergy for both antioxidant and anti-pigmentation skin treatment.
The degradation of the dermal ECM is also associated with these phenomena, such as the reduction or inhibition of collagenase and/or elastase activities, which might contribute to their prevention 17,39 . Recent studies have reported that many phenolic compounds isolated from plants with antioxidant properties, such as catechin and epigallocatechin gallate (EGCG), are also collagenase and/or elastase inhibitors. Inhibition usually involves metal chelation, making the catalytic Zn 2þ of collagenase unavailable 19,40 . The hydrogen bonding between the hydroxyl groups of polyphenols and hydrogen bond donor or acceptor groups from elastase domains, hydrophobic interactions between the benzene rings of polyphenols and hydrophobic functional groups of the enzymes, and induced conformational changes 17,37 . Triterpenoids common to many plants, including some Plectranthus spp. such as P. neochilus, P. ornatus and P. ecklonii, are strong elastase and collagenase inhibitors 20,36,[40][41][42] ; they can bind reversibly to the catalytic sites of the enzymes, since an activity recovery has been observed upon dilution of the enzyme-inhibitor mixture 43 .
This study represents a primary evaluation of Plectranthus spp. extracts and isolated compounds as potential antioxidant agents capable of inhibiting the skin-related enzymes tyrosinase, collagenase and elastase, for the synergistic treatment of skin disorders, namely ageing, sagging and hyperpigmentation of the skin. Organic and aqueous extracts of several Plectranthus spp. plants, as well as their isolated compounds ( Figure 2) were obtained, characterised and screened by our team, following the procedures described in previous phytochemical studies 23,[30][31][32][33][34][44][45][46] . Their antioxidant and specific anti-enzymatic activities were tested in support of their potential interest for application in preventive dermatology and/or synergistic therapeutics. Furthermore, this is the first report on the in vitro activity of Plectranthus spp. extracts and isolated compounds as tyrosinase, collagenase and elastase inhibitors.
The organic extracts were obtained from 10 0.010 g of air-dried aerial parts and powdered plants in 200 ml of methanol, ethyl acetate or acetone after sonication at room temperature for 1 h. The organic extracts were filtered and the solvent was removed in a rotary evaporator at 40-50 C. The crude extracts (21 samples) were stored at 20 mg/mL in DMSO.

Antioxidant activity assay
The free radical scavenging activity of the organic extracts (100 g/mL) was evaluated by the DPPH assay according to literature procedures 33,48 . All extracts were dissolved in ethanol 70% (w/v), mixed with DPPH solution (100 mM in ethanol) and incubated at room temperature for 30 min, in the dark. The absorbance of the solutions was read at 517 nm (Perkin-Elmer Lambda 2 UV-vis spectrophotometer) against a blank containing the same concentration of the organic extracts in ethanol. Quercetin (IC 50 ¼ 10.3 1.5 g/mL) was used as positive control while negative control corresponded to DPPH in ethanol. Radical scavenging activity (RSA, %) was determined from Equation (1).

Enzyme inhibitory assays
Anti-tyrosinase activity assay The isolated compounds ( Figure 2) and extracts (both organic and aqueous) from the described Plectranthus plants were tested in the anti-tyrosinase activity assay with modifications 15 . The assay was performed using 180 L of the substrate L-tyrosine (0.5 mM) in PBS 50 mM (pH 6.8) and 10 L of the tested samples (50 g/mL) incubated for 5 min at 37 C before starting the reaction by adding 10 L of tyrosinase (5000 U). After incubation at 37 C for 5 min, production of dopachrome was detected from absorbance measurements at 450 nm every 2 min, for 10 min, in a microplate reader (Thermo-Fisher Scientific). Kojic acid (0.8 mM) was used as positive control, with reported IC 50 of 43.7 M 49 , and sample solvent (DMSO 0.5% (v/v) in PBS buffer) as negative control. All assays were performed in triplicate. Results were expressed as percentage inhibition (%) determined from Equations (2) and (3). The absorbance variation (Abs) registered by Equation (2) for enzyme velocity reaction of negative control (Abs/time) must be in the linear range.

(3)
Anti-collagenase activity assay The anti-collagenase enzymatic assay was optimised based on several methods reported in the literature 16,17,40,50 . The synthetic substrate N-[3-furyl-acryloyl]-Leu-Gly-Pro-Ala (FALGPA) 0.1 mM was dissolved in tricine buffer 50 mM (pH 7.5) supplemented with 400 mM sodium chloride (NaCl) and 10 mM calcium chloride (CaCl 2 ) (assay buffer). Collagenase was prepared in the assay buffer at 1 U, knowing that 1 U hydrolyses 1 mol of FALGPA per minute, at 25 C, in the presence of calcium ions. EGCG at 40 M was used as positive control with reported IC 50 of 0.9 mM 51 , and the sample solvent (DMSO 0.3% v/v in tricine buffer) as the negative control. The assay mixture containing 80 L of tested samples (100 g/mL) and 100 L of collagenase was incubated at 37 C for 10 min before starting the reaction by adding 20 L of FALGPA. The assay was performed in triplicate. Absorbance of FALGPA was read at 405 nm for 10 min, continuously, in a microplate reader (Thermo-Fisher Scientific). Results were expressed as percentage inhibition (%), according to Equations (2) and (3).

Anti-elastase activity assay
The anti-elastase enzymatic assay was based on spectrophotometric methods described in the literature, with some modifications 16,17,19,52 . The substrate N-succinyl-Ala-Ala-Ala-p-nitroanilide (SANA) 1 mM was dissolved in Tris-HCl buffer 50 mM (pH 8.0), knowing that 1 U enzyme (elastase) converts 1 mol of SANA per minute in this buffer at 25 C. Ursolic acid (100 g/mL) was used as positive control with reported IC 50 of 10 M 19 , and the sample solvent (DMSO 1% (v/v) in Tris-HCl buffer) as negative control. The reaction was initiated by adding 20 L of SANA and 150 L of Tris-HCl buffer followed by addition of elastase (6 U) and the samples (100 g/mL). The mixture was incubated at 25 C for 10 min. Formation of p-nitroaniline from SANA hydrolysis was detected from absorbance measurements at 405 nm, performed immediately after starting the reaction and then every 30 s for 3 min, in a microplate reader (Thermo-Fisher Scientific). All assays were performed in triplicate. Results were expressed as percentage inhibition (%) according to Equations (2) and (3).

Statistical analysis
Data comparisons were conducted with one-way analysis of variance (ANOVA) followed by post-hoc Tukey honest significant difference test, for pairwise comparisons. Analysis and graphical presentation were performed with the GraphPad Prism Software Version 5 (GraphPad Software, Inc., San Diego, CA, USA). Values of p < 0.05 were statistically significant. Results were presented as mean standard deviation (SD).

Antioxidant activity
The antioxidant activity of several Plectranthus spp. organic extracts was evaluated concerning their ability for scavenging the DPPH radical, using quercetin as positive control. The results, expressed in percentage of RSA, are shown in Figure 3.

Inhibition of skin-related enzymes
In vitro tyrosinase inhibition The present study performed a screening of several Plectrathus spp. extracts (both aqueous and organic) and previously isolated secondary metabolites regarding their in vitro inhibition of skinrelated enzymes, namely tyrosinase, collagenase and elastase.
In the in vitro anti-tyrosinase assay, the enzymatic activity was evaluated by using L-tyrosine as substrate and detecting the produced chromophore (dopachrome) at 450 nm 49

as described in
Materials and Methods -Anti-tyrosinase Activity Assay. The results obtained for the Plectranthus spp. organic extracts shown in Figure 4(A), expressed as percentage of inhibition of tyrosinase activity, ranged from 25 to 68% compared to 92.9 7.4% obtained for kojic acid, used as positive control.
Aiming at a more extensive comprehension of the agents causing tyrosinase inhibition, the aqueous extracts obtained from Plectranthus spp. were additionally tested, and the results obtained, presented in Figure 4(B), revealed that in contrast with the organic extracts, the aqueous extracts were less effective as tyrosinase inhibitors.
To better understand the observed results, an additional assay was performed to evaluate which natural products present in the organic and aqueous extracts could be responsible for the exhibited anti-tyrosinase activity. Thus, previously isolated compounds from both organic and aqueous extracts of Plectranthus spp. were tested for their ability to inhibit tyrosinase in vitro, and the results obtained are shown in Figure 4(C).

In vitro collagenase inhibition
Evaluation of the in vitro anti-collagenase activity of Plectranthus spp. extracts and isolated compounds was performed using FALGPA as substrate and measuring the collagenase (ChC) activity by the decrease in absorbance at 405 nm due to FALGPA hydrolysis as described in Materials and Methods -Anti-collagenase Activity Assay. Figure 5(A) represents the results obtained for the organic extracts, expressed as percentage inhibition, revealing a mild to high ChC inhibitory activity, ranging from 28% to 76%, compared with EGCG used as positive control.
The aqueous extracts from Plectranthus spp., characterised by high amounts of phenolic compounds, which have been described as collagenase inhibitors 14 , were also studied for ChC inhibition, and results are displayed in Figure 5(B).
Since both organic and aqueous extracts from Plectranthus spp. in general exerted an inhibitory effect on collagenase, isolated compounds were further tested in order to determine their contribution for the enzymatic inhibition observed, and results obtained are exhibited in Figure 5(C).

In vitro elastase inhibition
The anti-elastase activity of Plectranthus spp. extracts and isolated compounds was evaluated in vitro using SANA as substrate and detecting the formation of p-nitroaniline at 405 nm resulting from SANA hydrolysis, as described in the Materials and Methods -Antielastase Activity Assay, using ursolic acid as positive control. Results of the enzymatic assay concerning the organic extracts from Plectranthus spp. plants are expressed in Figure 6(A), revealing in general very weak elastase inhibitory activity.
Since polyphenolic compounds have been suggested to possess anti-elastase activity 20 , the aqueous extracts expected to have this type of compounds were tested and the results are shown in Figure 6(B).
Although aqueous extracts from Plectranthus spp. plants were also not effective as elastase inhibitors (p > 0.05), according to Figure 5(B), the isolated compounds were still further assessed for their ability to inhibit elastase and results are displayed in Figure 6(C).

Antioxidant activity
The antioxidant properties of Plectranthus spp. organic extracts were evaluated based on their scavenging activity for the DPPH radical. According to Figure 3, the methanolic extracts held the highest RSA (20-76%) among the organic extracts, except for the ethyl acetic extract from P. grandidentatus, with a RSA of 62.3 0.4%. The observed results are possibly related with the high content of polyphenols usually present in the methanolic extracts, which are known for their antioxidant activity 53 . Nevertheless, recent findings concerning the antioxidant activity of abietane diterpenes, namely compounds (1) and (4) 23 , which have been isolated from Plectranthus spp., suggest that these compounds may also contribute to the observed bioactivity. The quinone moiety present in abietane diterpenes is probably responsible for their biological activity, since quinones represent important features at stabilising free radicals in many biological systems suffering from oxidative stress, protein inactivation and intermediate melanin synthesis pathway in human skin 54,55 .
The antioxidant results obtained for the methanolic extracts of P. ecklonii (75.9 1.0%) and P. madagascariensis (48.4 1.2%) as well as for the ethyl acetic extract of P. grandidentatus (62.3 0.4%) can also be explained by their main diterpenoid compounds with an abietane backbone associated with antioxidant activity 23 . Additionally, P. ecklonii ethyl acetic (55.5 1.7%) and acetonic (56.9 1.5%) extracts showed relatively higher RSA when compared to the remaining Plectranthus spp. extracts, which held high RSA values only for the methanolic extracts. Actually, P. ecklonii and P. grandidentatus had the highest antioxidant activity among the studied Plectranthus species, with RSA similar to that of quercetin (89.0 2.5%) used as positive control, which is a potent antioxidant flavonoid compound 23,33 .
Recently there has been an increased concern about synthetic antioxidants, such as the widely used butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT), which are suspected to induce liver damage and carcinogenesis in animals 39 . Therefore, less cytotoxic and eventually more powerful antioxidants from natural sources, such as Plectranthus plants, represent promising alternatives.

Tyrosinase inhibition
The Plectranthus spp. extracts and isolated compounds have been evaluated as tyrosinase inhibitors. The organic extracts showed promising results (Figure 4(A)) with inhibitory activity as high as 68% compared to 92.9 7.4% obtained for kojic acid used as positive control. According to the results obtained (Figure 4(A)), the methanolic (25-66%) and acetonic (27-68%) extracts exhibited the highest tyrosinase inhibition activity (p < 0.05). High tyrosinase inhibition was observed for the P. grandidentatus acetonic (67.9 3.6%), P. ecklonii methanolic (65.9 3.4%) and P. saccatus acetonic (56.4 5.7%) extracts. Moreover, P. grandidentatus and P. ecklonii showed the highest anti-tyrosinase activity in vitro, and these were also the plants that showed increased antioxidant activity. The observed results can be attributed to the abietane diterpenoids mainly present in the organic extracts of these two plants 23,33 .
In contrast with the organic extracts, the aqueous extracts (Figure 4(B)) were less effective as tyrosinase inhibitors (p > 0.05), except for the aqueous extract of P. porcatus (p < 0.0001), showing tyrosinase inhibition of 65.0 8.7% in comparison with 92.9 7.4% obtained for kojic acid used as positive control. Actually, the P. porcatus aqueous extracts obtained with microwave extraction have been previously characterised by high-performance liquid chromatography (HPLC) concerning quantification of polyphenols 22 . The obtained HPLC profile revealed that P. porcatus aqueous extracts have some amounts of rosmarinic acid (17) and caffeic acid, however P. ecklonii and P. saccatus had the highest polyphenol content 22 . Therefore, it can only be suggested that the anti-tyrosinase activity of the aqueous extract of P. porcatus is probably due to a synergistic effect of the present compounds.
Further evaluation of the effects on tyrosinase activity of isolated compounds from both organic and aqueous extracts ( Figure  4(C)) confirm some of the high inhibition values previously found for the extracts. Notably, the abietane diterpenes (1), (3) and (4) present mainly in P. grandidentatus, P. madagascariensis, and P. ecklonii organic extracts, are able to inhibit tyrosinase activity in more than 46% and up to 75%. Moreover, it is possible to better understand the activity of the P. porcatus aqueous extract according to the result obtained for compound (9) which inhibited tyrosinase activity by 43.9 1.3%. Although this compound is mainly isolated from the acetone extract of P. porcatus, which showed low tyrosinase inhibition (28.8 3.4%, Figure 4(A)), its presence in the aqueous extract cannot be excluded since the microwave aqueous extraction method has a higher efficiency in the recovery of bioactive compounds 22 . Additionally, (16) and (17) were able to inhibit tyrosinase by 47.9 1.4% and 40.4 0.7%, respectively. However, these compounds are not present in high amounts in P. porcatus aqueous extracts, therefore it can only be suggested that the extract activity was probably due to the presence of compound (9). Overall, the results suggest that Plectranthus spp. polyphenols and abietane diterpenes are capable of inhibiting tyrosinase being almost as efficient as kojic acid used as positive control.
Although neither the mechanism of inhibition nor the type of inhibition were studied, the most potent tyrosinase inhibitors such as hydroquinone, kojic acid, azelaic acid and other electron-rich phenols, have been reported for their capability of inhibiting melanin overproduction 21,37 . However, the use of these compounds in the fight against skin ageing and hyperpigmentation is limited due to their adverse side effects, low formulation stability, and poor skin penetration, hence the search for new agents from natural products.
Hydroquinones and phenols, part of the chemical structure of abietane diterpenes (1-4), (16) and (17), have been recognised for their chelating ability 21 , which can probably explain their high anti-tyrosinase activity. Moreover, due to the presence of polyphenolic compounds such as chlorogenic acid (16) and quercetin, other plants have been used for the treatment of skin depigmentation 14 . Actually, quercetin, a strong antioxidant flavonoid compound, has been reported to be a strong inhibitor of tyrosinase, with an IC 50 of 0.10 mM 37 . This strongly suggests that the observed synergistic effect of Plectranthus spp. natural products with both antioxidant and anti-tyrosinase inhibitor properties can be useful for anti-pigmentation skin treatment.

Collagenase inhibition
The organic extracts from Plectranthus spp. exhibited mild to high ChC inhibitory activity, in the range 28-76% ( Figure 5(A)). The highest anti-collagenase activity was observed for P. neochilus methanolic extract (76.4 2.1%) in comparison with the positive control EGCG (93.1 5.3%). Additionally, the organic extracts of P. madagascariensis, P. grandidentatus, and P. ecklonii were effective at inhibiting ChC in more than 60%.
The obtained results for the P. neochilus organic extracts strongly suggest that the anti-ChC activity observed is mainly due to pentacyclic triterpenes typically present, particularly (14) and (15), previously reported as ChC inhibitors 43 . On the other hand, the high inhibition of P. ecklonii organic extracts is most likely due to the presence of other pentacyclic triterpenes, such as (7) and (8) and/or abietane diterpenoids (1)(2)(3)(4). Actually, compounds (7), (8), (14) and (15) have been widely studied for their ability to inhibit both collagenase and elastase, possibly by reversibly binding to the catalytic sites of these enzymes 19,36,43 .
Phenolic compounds have also been described as collagenase inhibitors 14 , thus the aqueous extracts from Plectranthus spp. containing the larger amounts of these compounds were also evaluated for ChC inhibition (Figure 5(B)). In contrast with the organic extracts, the P. neochilus aqueous extract was the lowest ChC inhibitor (24.9 9.4%), although the remaining aqueous extracts revealed more promising results, the highest inhibition (75.6 6.5%) being obtained for the P. ecklonii aqueous extract.
Previous studies concerning polyphenols quantification in Plectranthus spp. plants have established that P. ecklonii aqueous extracts obtained from microwave extraction had one of the highest contents on compound (17) 34 . This study helps to understand the high collagenase activity observed for the P. ecklonii aqueous extract since many polyphenolic compounds, such as catechin and EGCG, have been reported to inhibit collagenase 40 , probably by acting as metal chelators, making the Zn 2þ ion unavailable for catalytic activity 19,40 .
In general, both organic and aqueous extracts from Plectranthus spp. exerted an inhibitory effect on collagenase, thus isolated compounds were further tested ( Figure 5(C) in order to determine their contribution for the enzymatic inhibition observed. According to Figure 5(C), the diterpene abietanes (3) and (4) mainly isolated from organic extracts along with (17), the major polyphenol present in the aqueous extracts, showed the highest ability for collagenase inhibition. Notably, compound (4) was found to inhibit ChC by 84.6 5.9%, being almost as efficient as positive control EGCG (93.1 5.3%). To our knowledge, this is the first report on the in vitro inhibition of ChC by diterpenes with an abietane backbone.
On the other hand, the royleanones (1) and (2) inhibited ChC by only 33.5 3.3% and 24.0 3.0%, respectively. These compounds are also abietane diterpenes, but with a royleanone motif that may be lowering the inhibitory capacity when compared to (4), whose structure includes different donor atoms resulting in higher metal chelation ability 23 . Compound (17) inhibited collagenase by 44.8 4.5% suggesting the presence of other polyphenols, such as compound (16) or caffeic acid, in the aqueous extracts of Plectranthus spp. to possibly justify the reported activity in Figure  5(B). Overall, Plectranthus spp. plants are highly promising natural sources for developing potential cosmetic agents against skin ageing induced by increased collagenase activity.

Elastase inhibition
The anti-elastase activity of Plectranthus spp. organic extracts ( Figure 6(A)) revealed impaired elastase inhibition, in contrast to the previous enzymatic assays regarding tyrosinase and collagenase inhibitory activity. Under the experimental conditions, the maximum elastase inhibition obtained was 42.8 4.2% for P. grandidentatus ethyl acetic extract, while the positive control, ursolic acid, inhibited elastase by 69.9 3.7%. The organic extracts from P. madagascariensis and the acetonic extract from P. grandidentatus showed no elastase inhibition (p > 0.05).
Mild elastase inhibition (around 30%) was observed for the organic extracts from P. neochilus and P. ecklonii, probably due to the presence of different types of triterpenes, such as compounds (7), (8), (14) and (15). Moreover, previous studies suggested that polyphenolic compounds may have anti-elastase activity due to interaction of the hydroxyl groups with the elastase domain 20 . Therefore, the Plectranthus spp. aqueous extracts expected to have this type of compounds, were further evaluated for anti-elastase activity (Figure 6(B)).
The results on Figure 6(B) reveal that the aqueous extracts from Plectranthus spp. plants were not effective as elastase inhibitors (p > 0.05), being the P. grandidentatus aqueous extract the only one able to inhibit the elastase enzyme, by 23.8 10.0% (p < 0.0001). This result can be attributed to different and decreased synergy of the polyphenols present in P. grandidentatus aqueous extract, in comparison with the remaining aqueous extracts 34 . Nevertheless, the isolated compounds from both organic and aqueous extracts were further assessed for their ability to inhibit elastase activity (Figure 6(C)).
In marked contrast to the Plectranthus extracts, the isolated compounds were strong elastase inhibitors. In accordance with the positive control, ursolic acid, compounds (7) and (8) in 4:1 mixture showed a high anti-elastase activity of 63.4 2.7% (p < 0.0001). These preliminary results strengthen the literature reports regarding the anti-elastase activity of the pentacyclic triterpenes 20,36,[40][41][42] . On the other hand, compounds with pentacyclic triterpene structure similar to that of compound (8), such as (14) and (15), which were anticipated to have high anti-elastase activity according to previous studies 40 , showed no significant activity. This may be due to interaction on elastase possibly involving different subdomains 20 .
Besides the triterpenes, royleanone-like diterpenes were also very effective in elastase inhibition. These compounds, including (4) (52. The results obtained with the isolated compounds from Plectranthus spp. plants concerning the in vitro enzyme inhibitory assays of the three skin-related enzymes (tyrosinase, collagenase and elastase), which are summarised in Table 2, suggest that Plectranthus spp. natural products are promising bioactive agents for future formulations against hyperpigmentation, wrinkle, and sagging of the skin.
The experimental results showed that the methanolic extracts from P. ecklonii and P. madagascariensis, as well as the P. grandidentatus ethyl acetic extract, generally had the highest values for radical scavenging activity and enzymatic inhibition. Interestingly, the isolated compounds mainly present in these plant extracts (abietane diterpenes, triterpenes, and polyphenols) also revealed high inhibitory effects on tyrosinase, collagenase and elastase enzymes. The highest inhibitory activity of P. porcatus aqueous extract in the anti-tyrosinase assay, which contains only minor amounts of polyphenolic compounds, is probably due to the presence of the diterpenoid (9). Despite the lower efficiency observed in the elastase inhibition assay, the abietane diterpene (4) showed high inhibitory activity only surmounted by the triterpene mixture of (7)(8).
In summary, all the plants had antioxidant properties, with methanolic extract having the highest scavenging activity compared to other organic extracts. The organic extracts showed promising results. P. grandidentatus acetonic extract, P. eckolonii methanolic extract and P. saccatus acetonic extract showed high tyrosinase inhibition. The aqueous extracts were not good inhibitors, except for P. porcatus extract which was better than the other plants extracts. The isolated compounds (1), (3) and (4), from both organic and aqueous extracts, were able to inhibit tyrosinase. The organic extract showed high collagenase inhibition, the highest anti-collagenase activity was observed for P. neochilus methanolic extract. For the aqueous extracts, P. ecklonii showed the highest inhibition. Compounds (3), (4) and (17) also showed good inhibition.
The organic extracts were not good inhibitors of the elastase enzyme, the percentage inhibition was too low compare to the positive control. P. grandidentatus ethyl acetic extract have some activity but it was low. The aqueous extracts were also not good inhibitors of elastase, again P. grandidentatus was better than all the other plants extracts but it was not good.
Thus, due to their antioxidant properties and in vitro ability to inhibit the skin-related enzymes (tyrosinase, collagenase, and elastase), the Plectranthus spp. extracts and isolated compounds represent promising bioactive agents with strong potential in cosmetic and/or pharmaceutical formulations to be developed against wrinkling, photo-ageing, hyperpigmentation and sagging of the skin. With this objective in mind, an article has recently been published that studies two different basic formulations containing P. ecklonii extracts, one in an organic solvent and the other using water 56 .
Recent discoveries on non-neuronal cholinergic system as a regulator of skin (patho)physiology have drawn attention to the importance of acetylcholinesterase (AChE) in the human skin 57,58 . Apart from the extensively known role of AChE in the termination of nerve impulse transmission at the cholinergic synapses by rapid hydrolysis of the neurotransmitter acethylcholine, the rationale underlying the use of AChE inhibitors in the symptomatic treatment of Alzheimer's disease 59,60 . Thus, further studies concerning anti-AChE activity in vitro will be conducted based on previous reports of high AChE inhibition by some Plectrathus spp. aqueous extracts (P. ecklonii, P. grandidentatus, P. ornatus, P. porcatus and P. saccatus) attributed to the presence of rosmarinic (17) and caffeic acids as major compounds 14,34,44,61,62 .

Disclosure statement
The authors declare no competing financial interests.