Preparation of the strains

Positive controls of Klebsiella pneumoniae (ATCC 13883), Enterococcus faeccalis (ATCC 29212), Staphylococcus aureus (ATCC 25923), Escherichia coli (ATCC 25922) and Candida albicans were used. The microorganisms were obtained from the Mexican National Laboratory for Food Safety Research and activated in culture medium of Trypticase Soy Broth (TSB) (BD Bioxon, Mexico) incubated at 37 °C for 24 h; subsequently, they were streaked on Trypticase Soy Agar (TSA) (BD Bioxon, Mexico) and incubated at 37 °C for 24 h (Microbank, Pro Lab Diagnostics, USA). Finally, the colonies obtained were used to prepare a suspension of 1.5 x 10⁸ CFU/mL corresponding to McFarland scale standard 0.5 (^{CLSI, 2017}).

Drugs and reagents

Reagents naringin, apigenin, chlorogenic acid, diosgenin, methanol, hexane, 2-aminoethyl diphenyl borate, anisaldehyde and dimethyl superoxide were obtained from Sigma-Aldrich (St. Louis, MO, USA); ethyl acetate, glacial acetic, formic, and sulfuric acids from J.T. Baker (Xalostoc, State of Mexico, Mexico); and polyethylene glycol 4000 from Alfa Aesar (Tewksbury, MA, USA).

Plant collection

Leaves of T. diffusa Willd ex Schult. var. aphrodisiaca Ward Urb. and T. diffusa Willd. ex Schult. var. diffusa were collected in the stage of flowering and fruiting in Imala, Culiacán, Sinaloa, Mexico (24° 51› 35» N; 107° 13’ 1” W; 92 masl) from 2016-2017. The varieties were identified at the Universidad Autónoma de Sinaloa (UAS) in Culiacán, Sinaloa and Centro de Investigaciones Biológicas del Noroeste (CIBNOR) La Paz, B.C.S. Mexico. Samples with access numbers HCIB30319 for Tdd and HCIB300069 for Tda were deposited in CIBNOR herbarium.

Extract preparation

The leaves, previously washed with tap water and dried in the shade for seven days, were pulverized and macerated with methanol or hexane, in proportion 1:10, under constant agitation of 175 rpm (Shaker DOS-10L, SKY LINE, ELMI, CA, USA) for 24 h. Then, the extracts were filtered and concentrated under vacuum in a rotary evaporator (BUCHI, CH) at 335 mbar, 45 °C and 35 rpm, and at 360 mbar, 40 °C and 35 rpm for the methanolic and hexanic extract, respectively. The concentrated extracts were stored at 4 °C until use.

High-performance thin-layer chromatography

The HPTLC analysis was performed based on the methodology proposed by ^{Reich (2012)}. Damiana methanolic and hexanic extracts (5 μL of a 1:10 solution in methanol or hexane, respectively) and standard solutions naringin, chlorogenic acid, apigenin and diosgenin (4 μL of a 1:1 solution in methanol or hexane, respectively) were loaded as 10-mm band length and 15-mm distance between each band in a 10 x 10-cm glass plates HPTLC 60F₂₅₄ (CAMAG, Merck, DE) using a semi-automatic applicator Linomat V (Camag, Muttenz, CH) which was controlled with VisionCATS 2.4 software. The plate was developed in a double channel chamber (CAMAG, Merck, DE) with the respective mobile phase up to 80 mm. The ^{FEUM (2013)} methodology was followed by separating the phenolic compounds from the methanolic extract, which consisted of developing the plate with the mobile phase: ethyl acetate, acetic acid, formic acid and water (100:11:11:27) with a previous saturation of 20 min. The developed plate was let to dry at room temperature and heated at 100 °C for three min. The bands were derivatized spraying with 2-aminoethyl diphenyl borate 1 % in methanol (NP) and with polyethylene glycol 4000 5 % in ethanol (PEG). For the hexanic extract, the mobile phase used was hexane, acetic acid (7.2: 2.9) and the plate was developed under the same conditions as the methanolic extract, but derivatization was performed with anisaldehyde in concentrated sulfuric acid (170 mL of cold methanol, 20 mL of glacial acetic acid, 10 mL sulfuric acid, and 1 mL of anisaldehyde) (^{Sophia et al., 2011}). The bands obtained on the plate were observed on UV light at 254 and 366 nm (CAMAG, Merck, DE) and detected with a TLC scanner (CAMAG, Merck, DE) at 254 and 366 nm for the methanolic extract, and 500 nm for the hexanic extract (^{Sophia et al., 2011}; ^{Seasotiya et al., 2014}).

Antimicrobial susceptibility assays

Chromatographic profile of methanolic extract

Crude extracts were used for the analyses. In Figure 1, profiles of the two varieties are shown, before (a) and after (b) derivatizing the metabolites with the NP-PEG developer. After derivatization, four yellow bands were observed in both varieties with greater intensity and definition in Tda variety; a yellow to brown band was also observed at daylight. At 254 nm, eight and seven yellow bands were observed in Tda and Tdd, respectively. At 366 nm, the methanolic extract chromatograms of T. diffusa leaves of both varieties revealed yellow to orange band colorations, these results confirmed the presence of flavonoids, which were in accordance to the ^{FEUM (2013)}.

Figure 1 Chromatographic profile of Tda and Tdd methanol extract. Line 1: standards naringin, chloro genic acid and apigenin (1 mg/mL), line 2: Tda and line 3: Tdd. a) Plates before derivatizing, b) Plates after derivatizing. 

The HPTLC chromatograms generated for the standards and the methanolic extract of Tda and Tdd are shown in Figure 2. The 366 nm chromatogram in Tda showed ten peaks (2A) while that of Tdd showed eight (2B). These peaks may represent groups of metabolites, since crude extracts were used.

Figure 2 a) Methanolic chromatogram of Tda by HPTLC b) Methanolic chromatogram of Tdd. c) Chromatogram of standards, peak 1: naringin, peak 2: chlorogenic acid and peak 3: apigenin. 

Chromatographic profile of the hexanic extract

In Figure 3, profiles of the hexanic extract of the two varieties are shown, before (a) and after (b) deriving the metabolites with anisaldehyde in sulfuric acid, revealed bands from blue to violet colorations (Figure 3b), which corresponded to terpenoid natural metabolites (^{Sophia et al., 2011}). The chromatograms obtained by HPTLC of the hexanic extracts revealed nine peaks for the two varieties (Figure 4) and differences in retention by the stationary phase, shown in Table 2.

Figure 3 Chromatographic profile of Tda and Tdd hexanic extract. Line 1: standard diosgenin (1 mg/mL), line 2: Tda and line 3: Tdd. a) Plates before derivatizing, b) Plates after derivatizing. 

Figure 4 a) Hexanic chromatogram of Tda by HPTLC, b) Hexanic chromatogram of Tdd, c) Chromatogram of standard diosgenin. 

It is well known that damiana has sesquiterpenes (caryophyllene), triterpenes (squalene and sisosterol), and polyterpenes (ficaprenol) (^{Zhao et al., 2007}), whose compounds correspond to the chemical groups seen in this study.

Microbial susceptibility

Table 3 shows the results obtained from the antibacterial and antifungal activity evaluation by disc diffusion method of the crude methanolic and hexanic extracts of Tda and Tdd against S. aureus, E. faecalis, E. coli, K. pneumoniae and C. albicans. When the results were compared with control (ciprofloxacin for bacteria and nystatin for yeast), the methanolic and hexanic extracts showed similar or inferior capacity to the positive control. The 3 mg of the methanolic extract of Tdd showed an inhibition halo of 14.49 mm in the test against S. aureus, showing the best activity but no significant differences within the concentrations of 2.5, 2.0 and 1.5 mg of the same extract. The hexanic extract of Tdd was the only one that showed activity against E. coli, without significant differences between 1-3 mg and no activity against E. faecalis while the Tda and Tdd methanolic extract at 2.5 and 3 mg had the best results on this bacterium. As for K. pneumonia, all the extracts showed inhibitory activity, but they did not show statistically significant differences when compared with the control group. Only the hexanic extracts of the two varieties (Tda and Tdd) show antimicrobial activity at 3 mg against C. albicans.

All the statistical analyses had a square R of 98.00-99.80, with a 95 % level of significance.

Although some extracts showed no inhibitory activity in the antimicrobial activity test by the disc diffusion method, the Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC) assays were performed for all the extracts. For C. albicans, where only the highest concentrations of hexanic extracts showed activity, the MIC and MBC were found at 0.05 mg for the methanolic and hexanic extracts of the two varieties. As for E. coli and K. pneumoniae, they were recorded from 0.05 to 0.1 mg. The MIC was found from 2-3 mg for the methanolic extract with respect to S. aureus and E. faecalis. On the other hand, the hexanic extract of the two varieties for S. aureus and E. faecalis was found at 0.5 mg and 1.5 mg, respectively. Also, the MBC for Enterococcus faecalis was not observed for the hexanic extracts even though the highest concentration of 5 mg was used (Data not shown) (Table 4).

The in-vitro antimicrobial evaluation of the Tda and Tdd methanolic extracts showed that both were effective against E. faecalis, K. pneumoniae and S. aureus, but the Tdd methanolic extract was statistically more effective (Tabla 3). Although, the Tda and Tdd methanolic extracts showed no visible activity against C. albicans using the disc diffusion method, the extracts of both varieties showed antimicrobial activity at 0.05 mg when the microdilution method was performed. It could have been related with the way the metabolites were diffused on the agar by using the diffusion method, which might not have been as homogenous as the microdilution method, since the metabolites are place in a homogenous aqueous medium.

On the other hand, the hexanic extracts of both varieties showed inhibition against K. pneumonia and C. albicans at a concentration of 3 mg, but no significant differences among them were observed (Table 4). Escherichia coli and S. aureus showed halos of inhibition from 10.6412.09 mm only in the Tdd hexanic extract using the diffusion method; however, both hexanic extracts of Tda and Tdd showed more effecacy with the microdilution method. Moreover, lower concentrations of hexanic extracts than methanolic extract were needed to have a higher efficacy on all the microorganisms tested, except for E. faecalis. The antimicrobial activity of methanolic extracts has been demonstrated against Gram-positive and Gram-negative bacteria where flavonoids were responsible for the activity (^{Taleb-contini et al., 2003}). The metabolites present in the methanolic extract are mainly flavonoids (Figure 1), for which their antimicrobial activity has been demonstrated both in Gram-positive and Gram-negative bacteria; furthermore, a structureactivity relationship has been observed on flavonoids, of which the most active are those that have dihidroxy groups at C3’ and C4’ positions, the position C6’ unsubstituted and C7’ with a hidroxy (^{Taleb-contini et al., 2003}). In addition, apigenin, another flavonoid present in damiana (^{Cushnie & Lamb, 2005}; ^{Nayaka et al., 2014}; ^{Cha et al., 2016}) has also been proven to have antimicrobial activity; thus, our study has demonstrated that the flavonoids found in both Tda and Tdd varieties were responsible for the antimicrobial activity against the urinary tract pathogens analyzed.

The methanolic extracts of our study also showed antimicrobial activity against both Gram-positive and Gram-negative bacteria showing maximum activity against E. coli and K. pneumonia according to the MIC obtained. In relation to the statistical analysis, the methanolic extracts of Tdd had significant differences compared to those of Tda. These differences might have been related with the type of metabolites present in each extract, as it was observed in the chromatographic analysis; the extracts had differences in peaks 3 and 4 of Tda, which were not found in Tdd, and peak 7 of Tdd that was not shared with Tda (Figure 2, Table 1). In addition, at 366 nm without derivatization, the chromatogram showed two bands (blue to green color) that were unique for Tdd. Thus, beside the flavonoids present they may have other metabolites that are causing the maximum activity. However, more studies are needed to make this assumption.

Finally, the antimicrobial activity of the metabolite terpenes and terpenoids have also been evaluated, demonstrating that terpenes have better activity when they are in a crude extract than individually isolated (^{Nazzaro et al., 2013}). Thus, the crude oil of the hexanic extracts in this study might have shown this activity on all the microorganisms evaluated, except for E. faecalis. The MIC and MBC obtained did not show significant differences against the microorganisms tested; however the chromatographic analysis of the Tda and Tdd hexanic extracts showed statistical differences. Therefore, the metabolites responsible for the inhibition activity of both varieties may not be those that are different between the varieties but those that are shared between them. Nevertheless, more studies are needed to identify the metabolites responsible for the antimicrobial activity in each variety and establish the relationship of each metabolite structure with the bioactive metabolites.