Moradi et al., 2013; Benedek and Kopp, 2007). This report was novel as this was the first time yarrow was determined to have acne treating abilities such as reducing inflammation, treating infections caused by P. acnes and S. epidermidis and removing dark pigmentation (Shah et al., 2015).

 The next steps in this study were to use bioactivity guided fractionation, high performance liquid chromato-graphy (HPLC), mass spectrometry and nuclear magnetic resonance (NMR) to identify the compounds contributing to the anti-acne effects in the petroleum ether extract. A novel compound from Achillea ‘Moonshine’ petroleum ether extract responsible for the anti-acne activity was isolated and characterized. 

Collection and extraction

Full bloomed Achillea ‘Moonshine’ were collected from Holly Days Nursery Inc. near Philadelphia (United States) in June 2013. The leaves and stems were washed three times with distilled water and grounded into fine powder using a simple grinder. 500 g of plant material (leaves and stem) was extracted with petroleum ether (3 × 1 L). The solvent was evaporated and the extracts were stored at 4°C till further use.

Bacterial strains and media

The test organism, P. acnes, was obtained from American Type Cell Collection (ATCC). P. acnes was incubated in brain-heart infusion media for 48 h at 37°C under anaerobic conditions in an anaerobic jar, and the density was adjusted to approximately 0.5 McFarland Standard.

Fractionation of the petroleum ether extract

Column chromatography

The petroleum ether extract was subjected to column chromate-graphy to separate the extract into its component fractions. Silica (#200) was used as the stationary phase while petroleum ether: diethyl ether (1:1) was used as the mobile phase. In the setting up of the column (length 45c m, diameter 3 cm), the lower part of the glass column was stocked with glass wool with the aid of a glass rod. The slurry was prepared by mixing 45 g of activated silica mixed with 10% AgNO₃ and 200 ml of hexane. This slurry was poured down carefully into the column. The tap of the glass column was left open to allow free flow of solvent into a conical flask below. The column was allowed to stabilize, after which, the clear solvent on top of the silica gel was allowed to drain down until the level of the silica gel. The sample was prepared in a glass mortar by adsorbing 300 mg of the petroleum ether extract with 500 mg of silica gel. The dry powder was gently layered on top of the column. The mobile phase was added and the flow rate was adjusted to about 40 drops/min. The eluted fractions were collected in test-tubes. 40 fractions were collected at the end of the run. Fractions 1-12 were clear in appearance. Fractions 13 to 35 were yellowish green in color while the later fractions were blue to brown in color.

Thin layer chromatography and pooling of the fractions

The content of each test tube was spotted on a TLC plate to separate the different fractions based on their relative mobility in solvent systems and the TLC profile was observed under the UV light. These fractions were eluted with toluene: methanol (9.6:0.4) as the mobile phase. The developed plates were dried under stream of fast moving air. The plates were then observed under long and short UV and were sprayed with 1% sulphuric acid in methanol. The fractions which showed similar patterns were pooled together. These fractions were then stored in 4°C till further use.

TLC bio-autography of the fractions for antimicrobial activity

From each of the pooled fraction as described in the above procedure, 10 µl of sample was loaded onto TLC plates in a narrow band and eluted with toluene : methanol (9.6:0.4). The developed plates were dried under stream of fast moving air to remove the traces of solvent. The prepared chromatograms were dipped in the bacterial suspension of P. acnes (ATCC, approximately 10⁶ CFU/ml) for 5-10 s. The plates were then incubated for 48 h at 37°C anaerobically in an anaerobic jar. Following incubation, the plates were sprayed with a 2 mg/ml solution of p-iodonitrotetrazolium violet (INT). These plates were further incubated anaerobically for 8 h. White bands on the developed plates indicate that the reduction of INT to the colored formazan did not take place due to the presence of compounds that inhibited the growth of tested organisms (Raman et al., 1995).

Isolation of compounds from active fractions by preparative HPLC and identification of active peak

The fraction from the column chromatography that inhibited the growth of P. acnes was analyzed with HPLC. Preparative HPLC was performed to isolate individual compounds from the active fractions. The HPLC was performed on Agilent 1100 series apparatus with an auto sampler, quaternary gradient pump and a UV/Vis diode array detector. Zorbax 300SB-C18 column (4.6 mm × 25 cm) was used. Mobile phase A consisted of HPLC water with 0.1% formic acid while mobile phase B was acetonitrile with 0.1% formic acid. The gradient eluting phase was A : B (10:100 v/v) for 50 min. The mobile phase was pumped at 0.4 ml/min with the column temperature of 30°C. The injection volume was 75 µl. All the major peaks were isolated and the solvent was evaporated. The residue was re-dissolved in methanol and TLC bio-autography was performed to identify the active isolated peak.

Identification of the active compounds separated by HPLC

The active compound was analyzed by IR, mass spectrometry and NMR.

Mass spectrometry

The mass spectrometry data was obtained by liquid chromato-graphy-mass spectrometry (LC-MS) using Thermo Scientific™ Orbitrap™. The mobile phase was A : B (10:100 v/v) for 50 min, where A was water with 0.1% formic acid and B was acetonitrile with 0.1% formic acid.

Nuclear magnetic resonance

All NMR experiments were conducted on a Bruker Advance III 400 MHz NMR Spectrometer (Bruker, Karlsruhe, Germany) operating at a proton NMR frequency of 400.13 MHz. Spectra were recorded in CDCl₃ at 25°C using standard Bruker 1D and 2D sequences. Experiments for structure elucidation include ¹H, ¹³C, 2D-COSY, HMBC and HMQC. The COSY (Correlation Spectroscopy) experiment indicates the hydrogen-hydrogen connectivity. The HMQC (Heteronuclear Multiple Quantum Coherence) and HMBC (Heteronuclear Multiple Bond Coherence) experiments indicate the carbon and hydrogen connectivity.

Isolation of compounds from active fractions by preparative HPLC and identification of active peak

Fraction YL-7 was further fractionated by normal phase HPLC to isolate individual peaks and identify the active peaks. Figure 1 shows the HPLC chromatogram of YL-7. Three peaks for YL-7 (P1, P2 and P3) were isolated and each peak was tested again for its antimicrobial activity against P. acnes by TLC bioautography (Figure 2). Out of the three peaks, peak 3 was found to have anti-microbial assay (Table 2).

Compound identification

YL-7 peak-3

YL-7 peak-3 was identified as [N-(21-hydroxy-21-(piperidin-1-yl) henicosa-17, 19-diyl-1-yl) acetamide]. White Powder: UV λ _max 253 nm, FT-IR (hexane) V_max 3428.53cm^-1 (-OH, -NH), 2088.85 cm^-1 (-C≡C), 1643.92 cm^-1 (-C=O), 2922.53 cm^-1 (-CH₃ stretch), 2852.49 cm^-1 (-CH2 stretch), 1461.87 cm^-1 (-CH2 bend). HRESI m/z 445.12 (calculated for C₂₈H₄₇NO₃ 445.356).

The presence of 11 CH₂ groups was corroborated by the presence of their high field signals at δ_H =1.28 in ¹H-NMR and ¹H-¹H COSY. This was also confirmed by ¹³C-NMR (δ_C=29.7) (Table 3). The HMQC helped to determine the chemical shift of the carbons on the molecule through direct correlation with their hydrogens. Diagnostic HMQC correlations and ¹H-¹H COSY also confirmed the presence of the piperidine ring (Figure 3).

The IR spectra indicated the presence of –OH and –CONH groups. The ¹³C- NMR and ¹H-NMR spectral data (Table 3) allowed the assignments of most of the hydrogens and carbons signals of the proposed structure. The presence of 11 CH₂ groups was corroborated by the presence of their high field signals at δ_H =1.28 in ¹H-NMR (Figure 4) and ¹H-¹H COSY (Figure 5). This was also confirmed by ¹³C-NMR (δ_C=29.7). The HMQC helped to determine the chemical shift of the carbons on the molecule through direct correlation with their hydrogens (Figure 6). 

From interpretation of all the spectral data, a novel piperidide with molecular formula C₂₈H₄₇NO₃ was isolated and identified from Achillea ‘Moonshine’ petroleum ether extract. This piperidide can be used as a novel drug for acne management and treatment. The compound can be further tested on animal models to validate their activity in vivo. Efforts can be made to synthesize the compound on a laboratory scale and optimize its yield. Further clinical trials can be conducted to evaluate the compound’s safety and efficacy.

This research was funded by the Startup grant given to Dr. Bela Peethambaran by University of the Sciences, Philadelphia.

The authors have not declared any conflict of interests.