Chemical Composition and Determination of the Antibacterial Activity of Essential Oils in Liquid and Vapor Phases Extracted from Two Different Southeast Asian Herbs—Houttuynia cordata (Saururaceae) and Persicaria odorata (Polygonaceae)

Essential oils obtained via the hydrodistillation of two Asian herbs (Houttuynia cordata and Persicaria odorata) were analyzed by gas chromatography coupled to mass spectrometry (GC–MS) and gas chromatography with flame ionization detector (GC–FID). Additionally, both the liquid and vapor phase of essential oil were tested on antimicrobial activity using the broth microdilution volatilization method. Antimicrobial activity was tested on Gram-negative and Gram-positive bacteria—Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, Enterococcus faecalis, Streptococcus pyogenes, Klebsiella pneumoniae, Seratia marcescense and Bacillus subtilis. Hydrodistillation produced a yield of 0.34% (Houttuynia cordata) and 0.40% (Persicaria odorata). 41 compounds were identified in both essential oils. Essential oils contained monoterpenes and their oxidized forms, sesquiterpenes and their oxidized forms, oxidized diterpenes, derivates of phenylpropene and other groups, such as, for example, aldehydes, alcohols or fatty acids. Both essential oils were antimicrobial active in both vapor and liquid phases at least in case of one bacterium. They expressed various antimicrobial activity in the range of 128–1024 μg∙mL−1, 512–1024 μg∙mL−1 in broth and 1024 μg∙mL−1, 512–1024 μg∙mL−1 in agar, respectively. Research showed new interesting information about P. odorata and H. cordata essential oils and demonstrated that both essential oils could be possibly used in the field of natural medicine or natural food preservation.


Introduction
In recent years, many researchers have been focused on finding new antimicrobial agents that could be applied to multi-resistant microorganisms. Medicinal herbs and their products, such as essential oils (EOs), are the main source of natural remedies. They have been used since the time immemorial as the most affordable means of treating diseases. As it has been proven several times, EOs have natural medicine and determine their EOs composition and especially their antimicrobial efficiency in both vapor and liquid phase. In this study were two EOs obtained by hydrodistillation and thereafter analyzed by standard techniques GC-MS and GC-FID. Antimicrobial activity was determined by the modern, recently developed method of Houdkova et al. [35] called the broth microdilution volatilization method. It is a simple and rapid simultaneous determination of the antibacterial potential of plant volatile compounds in the liquid and the vapor phase at different concentrations [35].

Extraction Yield and Chemical Composition
Hydrodistillation in the Clevenger-type apparatus of Houttuynia cordata produced a pale-yellow liquid with a fishy scent. The essential oil content of distilled aerial parts of dried plant was 0.34%. The extraction yield is higher in comparison with those previously published by R. S. Verma et al. [24] who only achieved a yield of 0.06-0.14%. A total of 41 compounds were identified that made up 90.6% of the essential oil composition ( Table 1). The essential oil contained a higher amount of terpenoid compounds (75.5%), followed by non-terpenoid compounds (15.1%), such as derivates of phenylpropene, aldehydes, ketones, esters and fatty acids. The major group of substances was monoterpenes with a content of 59.4%, followed by the group of other compounds with a content of 14.8%, oxidized monoterpenes with a content of 7.2% and sesquiterpenes with a content of 6.6%. Other groups were oxidized sesquiterpenes and derivates of phenylpropene. Major compounds of the essential oil were myrcene (51.6%), 2-undecanone (6.7%), tridecan-2-one (6.1%), cis-β-ocimene (5.7%), geranyl acetate (3.1%), bornyl acetate (2.9%) and cis-caryophyllene (2.6%). The other compounds were present at less than 2%. These results are similar with the results of previous reports [19,22,24]. Only a few fluctuations from other reports were found and are probably attributed to the origin of the plant samples or different extraction method. For the characteristic fishy scent and flavoring of H. cordata essential oils is responsible compound houttuynin (decanoyl acetaldehyde). This compound was not identified in our essential oil due to its instability. It is usual that decanoyl acetaldehyde is during the process of distillation easily oxidized into 2-undecanone [24]. This compound had the second highest concentration in our essential oil. Therefore, the amount of 2-undecanone is the primary indicator for the quality of Houttuynia cordata essential oil [23,24]. Hydrodistillation in Clevenger-type apparatus of Persicaria odorata produced a deep yellow liquid with a strong spicy coriander-like aroma. Due to its aroma, it is also called Vietnamese coriander [25]. The essential oil content of distilled aerial parts of dried plant was 0.41%. The extraction yield is lower in comparison with those previously published by A. A. Almarie et al. [33] who achieved a yield of 0.64%. A total of 41 compounds were identified that made up 90.4% of the essential oil composition. In comparison with other reports, we identified more compounds [38,39]. N. X. Dung et al. [38] used steam distillation for the isolation of essential oils and they identified 28 compounds and the most abundant were β-caryophyllene, dodecanal and caryophyllene oxide. M. V. Hunter et al. [39] only identified 17 compounds using steam distillation as an extraction technique for the isolation of essential oil from P. odorata, where the most abundant compounds were α-humulene, decanal and dodecanal. Our essential oil contained a higher amount of non-terpenoid compounds (72.7%) followed by terpenoid compounds (17.7%). Carbonyls and alcohols, especially C10 and C12, made up 68.8% of essential oil composition, followed by the group of sesquiterpenes with a content of 11.5% and oxidized sesquiterpenes with a content of 5.7%. Other groups (monoterpenes, oxidized monoterpenes and oxidized diterpenes) made up less than 1% of essential oil constitution. The major compounds of the essential oil were n-dodecanal (37.1%), n-decanal (18.1%), 1-decanol (5.4%), 1-dodecanol (4.8%), α-humulene (4.5%), cis-caryophyllene (3.9%) and n-undecane (2.5%). The other compounds were present at less than 2%. These results in relative percent content are similar with the results of previous reports. It can clearly be seen that the essential oil from Persicaria odorata is rich in C10 and C12 carbonyls. Dodecanal and decanal are the main compounds of Persicaria odorata essential oil in all previous published reports and ours [25,33,38,39].
When comparing both EOs, it was found that they have only 12 common compounds out of 41 but they vary in the percent content. The essential oil from Hottuynia cordata contained much more monoterpenes and monoterpenoids, where the most abundant compound was myrcene (51%) that was not found in the essential oil from the P. odorata. On the other hand, the essential oil from Persicaria odorata contained more sesquiterpenes, sesquipterpenoids and especially aldehydes, where n-dodecanal (37.1%) was the dominant compound compared to H. cordata essential oil, where it was only 0.02%. In general, the composition of both essential oils is different. The results are adequate because both herbs are neither from the same genus nor family, so their similarity in composition was not expected. They were selected for this study according to their similar use and same geographical occurrence.

Antimicrobial Activity
The antimicrobial activity of H. cordata and P. odorata essential oils is reported in Table 2. Both EOs showed antimicrobial efficiency but in different concentrations. H. cordata and P. odorata essential oil expressed various antimicrobial activity in the range of 128-1024 µg·mL −1 , 512-1024 µg·mL −1 in broth and 1024 µg·mL −1 , 512-1024 µg·mL −1 in agar, respectively. In the liquid phase, the lowest MIC was showed for H. cordata (128 µg·mL −1 ) against E. faecalis and for P. odorata (512 µg·mL −1 ) against S. pyogenes, E. faecalis and B. subtilis. In the vapor phase, the lowest MIC was observed for H. cordata (1024 µg·mL −1 ) against E. faecalis and E. coli and for P. odorata (512 µg·mL −1 ) against E. coli. There are observable differences between the efficiency of the vapor and liquid phases of observed essential oils (EOs). In most cases, the higher MIC reached the liquid phase, except in the case of EOs from P. odorata on E. coli, where the vapor phase was twice as effective as the liquid phase.
In general, Gram-negative bacteria are more resistant to EOs than Gram-positive bacteria [40]. This is supported by our results because, as it is shown in Table 2, Gram-positive bacteria were more sensitive to tested EOs in comparison with Gram-negative bacteria. It is possible that active compounds in EOs can more easily break important bonds (peptidoglycan) in the cell wall structure of Gram-positive bacteria. The structure of the Gram-positive bacteria cell wall allows hydrophobic molecules to easily penetrate the cells and act on both the cell wall and within the cytoplasm. After the cell wall is broken, the reactive constituents of the essential oil can penetrate the interior of the cell and further damage its DNA. The other fact is that phenolic compounds, which are also present in the EOs, generally show antimicrobial activity against Gram-positive bacteria. On the other hand, the cell wall of Gram-negative bacteria is far more complex, and it is, among other reasons, why they are more resistant to biologically active compounds (EOs) [4]. The most abundant compound in H. cordata essential oil is myrcene. Myrcene has an antimicrobial activity and moreover enhances the activity of antibiotics [41]. EOs with a high content of myrcene have a positive effect on urinary and genital infections [42,43]. These infections may be caused among others by E. coli and E. faecalis; therefore, we could assume that EOs from H. cordata will affect them, which has been confirmed in this study. The most abundant compound in P. odorata essential oil was α-humulene, which is known for its anti-inflammatory effect. It is well known that EOs with α-humulene are natural antimicrobial agents [44][45][46]. Pichette et al. [46] have tested the antimicrobial activity of α-humulene against E. coli and S. aureus using the microdillution method. α-humulene exhibited an MIC of 2.6 µg·mL −1 against S. aureus and an MIC of more than 20 µg·mL −1 against E. coli. Jang et al. [45] have tested the antimicrobial activity of α-humulene against B. fragilis and obtained MIC of 0.5 µg·mL −1 . The high content of α-humulene could be the reason why P. odorata EOs inhibited the growth of six from eight tested bacteria. On the other hand, there is one possible disadvantage when using these oils orally or internally, which is possible irritation or allergy caused by cis-caryophyllene, which both EOs contains [47]. It would be necessary to further examine the negative effect of each compound in the essential oil on the human body before using it for treating illness.
As far as authors know, there are no previous reports about Persicaria odorata and Houttuynia cordata essential oils and its antimicrobial activity in the vapor phase, so it is not possible to further compare those results with other publications. However, there are some reports about testing the liquid phase of EOs from Houttuynia cordata. Verma et al. [24] tested the antimicrobial activity of the essential oil from Houttuynia cordata against four bacteria (Staphylococcus aures, Streptococcus mutans, Mycobacterium smegmatis and Enteroccocus faecalis). Their essential oil exhibited MIC in the range of 0.52-1.04 µL·mL −1 . Ji et al. [20] performed a disc diffusion test to determine antimicrobial activity of H. cordata essential oil against Bacillus subtilis, Escherichia coli and Staphylococcus aureus; unfortunately, the disc diffusion test is only a screening method, which is not possible to compare with MIC. Lu et al. [22] tested the antimicrobial activity of H. cordata EOs against Staphylococcus aureus and Sarcina ureae using the broth and agar dilution method. Their reached minimal inhibitory concentration was in the range of 0.5-1.0 µL·mL −1 .

Plant Material
Approximately 120 g of fresh Chinese herbs (H. cordata and P. odorata) was purchased in a local Vietnamese market (TTTM Sapa, Prague, Czech Republic). Each sample was air dried in a dark room at the laboratory's temperature. Prior to the distillation, both herbs, including leaves and stems, were crushed into smaller pieces.

Essential Oil Isolation
Essential oils were obtained by hydrodistillation using Clevenger-type apparatus. The EOs was prepared as follows: 26.7 g (H. cordata) or 18.4 g (P. odorata) of dried herb was weighted into a 2000 mL distillation flask, 1000 mL of water was added and the EOs was distilled for 4 h. The essential oil was then separated from hydrosol and stored in sealed dark-glass vials at 4 • C until the analysis.

Antimicrobial Activity Assay
The antimicrobial activity of the liquid and vapor phase of Eos was determined by the broth microdilution volatilization method [35]. The experiments were carried out in 96-well microtiter plates with one well volume of 400 µL. The test is designed for the testing of 6 essential oils in total. For this study, we tested only 2 essential oils, and different samples were in other wells. The plates were covered with wooden plates and clamped to prevent vapor phase leakage. The edge wells were left blank to avoid the edge effect. First, essential oil samples were prepared as follows: approximately 2 µL of EOs was added to corresponding amount of dimethyl sulfoxide (DMSO) at a concentration of 1%, then further diluted in the corresponding broth to initial concentration. Then, the antibiotic was prepared at an initial concentration of 4 µg·mL −1 . In the first part of the experiment, 30 µL of agar was pipetted onto the plate lid and inoculated with 5 µL of bacterial suspension for vapor phase testing. In the second part (liquid phase assay), 100 µL of buffered Mueller-Hinton broth was pipetted into the wells. Each well had a final volume of 100 µL. Seven two-fold diluted concentrations of samples starting at a concentration of 1024 µg·mL −1 were prepared for each essential oil in one row. A positive and negative control of bacterial growth was prepared in the first two columns. In the last column, 6 two-fold diluted concentrations of antibiotic starting at 4 µg·mL −1 were prepared. Finally, all wells except the negative control were inoculated with 5 µL of bacterial suspension. Plates were closed, fixed and incubated at 37 • C for 24 h. After incubation, minimal inhibitory concentrations of EOs were evaluated by the visual assessment of bacterial growth after the coloring of a metabolically active bacterial colony with thiazolyl blue tetrazolium bromide dye (MTT; Sigma Aldrich, Prague, Czech Republic). A total of 25 µL of 600 µg·mL −1 dye was applied to the lid and each well of the plate and equilibrated for 10 min. The color changed from yellow (dead cells) to purple (live cells). Thereafter, the MICs were recorded. All experiments were performed in triplicate in three independent experiments. The results were expressed as the median of minimal inhibition concentration of the antimicrobial agent values.

GC-MS Analysis
The GC-MS analysis of samples was carried out by using a Gas Chromatograph GC 2010 coupled to a Mass Selective Detector GCMS-QP2010 Plus (both Shimadzu, Kyoto, Japan) and Combi Pal Autosampler (CTC Analytics, AG, Zwingen, Swizerland) on a capillary column SLB-5ms Supelco (30 m × 0.25 mm, 2.5 µm film thickness; Bellefonte, PA, USA). The carrier gas was Helium 5.0 (Linde, Prague, Czech Republic) with a constant flow of 30 cm·s −1 . The oven temperature program was set at an initial temperature of 40 • C for 3 min, then heated up to 250 • C at 2 • C·min −1 and held at 250 • C for 10 min. The injector and detector temperatures were set at 200 • C. The mass spectrometry detector was operated under electron ionization mode at ionization energy of 70 eV when ions with m/z 33-500 were scanned. A total of 1 µL of diluted essential oil (200 times, n-hexane) was injected with a split ratio 1:50. The experimental results of retention indices were calculated relative to C8-C33 n-alkanes in concentrations of 100-200 µg·mL −1 , dissolved in n-hexane (Restek, Bellefonte, PA, USA). The calculation was performed according to the van den Dool and Kratz method, and the results were further compared to published data [36,37]. Compounds were identified by comparing their mass spectra with mass spectra of several standards (Table 1) and commercial mass spectral databases NIST'14 Mass Spectral Library and FFNSC 2 GC/MS Library Release 2.0 (Flavor and Fragrance Natural and Synthetic Compounds Library) and further checked out by manual mass spectra evaluation.

GC-FID Analysis
The GC-FID analysis of samples was carried out by using a Gas Chromatograph GC 2010 with a flame ionization detector (Shimadzu, Kyoto, Japan) and Autosampler Combi Pal (CTC Analytics, AG, Zwingen, Swizerland) on a capillary column SLB-5ms Supelco (30 m × 0.25 mm, 2.5 µm film thickness; Bellefonte, PA, USA). The GC-FID conditions were the same as in case of GC-MS analysis. The injector temperature was set at 200 • C and the detector temperature was set at 260 • C. A total of 1 µL of diluted essential oil (200 times, n-hexane) was injected with a split ratio 1:50. As in the case of GC-MS, experimental retention indices were calculated and compared to published data.

Conclusions
This study shows new interesting knowledge about EOs distilled from two Asian herbs-Persicaria odorata and Houttuynia cordata. The chemical composition of EOs corresponds to previous studies with a minor deviation that may be caused by agronomic factors, sample storage or sample preparation and other factors. Both EOs showed antimicrobial activity in different concentrations to different bacteria. Due to great antibacterial activity, along with the composition of Eos, we see a great potential for future usages of these oils, such as natural antimicrobials or food preservatives. As far as we know, we were the first to describe the antimicrobial properties of those EOs in both vapor and liquid phases on eight selected bacteria. Furthermore, it is necessary to study the possible cytotoxicity of these oils. The disadvantage is that both oils contain cis-caryophyllene that causes allergic reactions and skin irritation. It would be necessary to find the balance in concentrations of beneficial antimicrobial active compounds and potentially toxic compounds.

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