Helleborus purpurascens—Amino Acid and Peptide Analysis Linked to the Chemical and Antiproliferative Properties of the Extracted Compounds

There is a strong drive worldwide to discover and exploit the therapeutic potential of a large variety of plants. In this work, an alcoholic extract of Helleborus purpurascens (family Ranunculaceae) was investigated for the identification of amino acids and peptides with putative antiproliferative effects. In our work, a separation strategy was developed using solvents of different polarity in order to obtain active compounds. Biochemical components were characterized through spectroscopic (mass spectroscopy) and chromatographic techniques (RP-HPLC and GC-MS). The biological activity of the obtained fractions was investigated in terms of their antiproliferative effects on HeLa cells. Through this study, we report an efficient separation of bioactive compounds (amino acids and peptides) from a plant extract dependent on solvent polarity, affording fractions with unaffected antiproliferative activities. Moreover, the two biologically tested fractions exerted a major antiproliferative effect, thereby suggesting potential anticancer therapeutic activity.

Helleboreus spp. (family Ranunculaceae) are spontaneous perennial flowering herb plants. In the Romanian spontaneous flora, there are two species of hellebore: Helleborus purpurascens and Helleborus odorus, both of them flowering in early spring [2,3]. Helleborus purpurascens is extremely toxic, yet known since long ago for its folkloric therapeutic effects. It has been used in traditional The developed RP-HPLC method allowing for separation and identification of the amino acids in the hellebore extract (S 1 , S 4 , S 5 and S 6 ) is based on the retention times of the used standards.

GC-MS Analysis
The content of hellebore extracts in amino acids and peptides and its dependence on solvent type were confirmed by GC-MS analysis. The obtained chromatograms are shown in Figure 1. The mass spectra of the components from the GC-MS chromatograms were compared with those from the NIST/NBS (National Institute of Standards and Technology/National Bureau of Standards) spectral database, and the identified amino acids and peptides are presented in Table 2.
The proposed isolation strategy unraveled the influence of solvent polarity on the efficiency of amino acid and peptide separation from the hellebore extract. Thus, the hexane fraction (S 2 ) contains the largest number of amino acids, with the fewest compounds separated in the ethanol fraction (S 1 ).

Hellebore Thionins Characterized by GC-MS
The results of the GC-MS for samples S1T (petroleum ether) and S2T (acetone) are shown in Figure 2.

Hellebore Thionins Characterized by GC-MS
The results of the GC-MS for samples S 1T (petroleum ether) and S 2T (acetone) are shown in Figure 2.

Hellebore Thionins Characterized by GC-MS
The results of the GC-MS for samples S1T (petroleum ether) and S2T (acetone) are shown in Figure 2.   The compounds identified by mass spectral interpretation through GC-MS analysis are listed in Table 3.
The efficiency of the chosen separation strategy for thionins was evaluated by GC-MS analysis. The results show the influence of solvent polarity on the number of isolated compounds. In the low polarity fraction S1T (petroleum ether); eleven amino acids and one dipeptide were isolated and identified in the GC-MS chromatogram (Table 3).

Nano-ESI-Chip-MS Analysis
All five hellebore fractions S1-S5 were submitted to high-throughput positive nanoESI chip MS and screening under identical solution and instrumental parameters. The obtained mass spectra are The compounds identified by mass spectral interpretation through GC-MS analysis are listed in Table 3.
The efficiency of the chosen separation strategy for thionins was evaluated by GC-MS analysis. The results show the influence of solvent polarity on the number of isolated compounds. In the low polarity fraction S 1T (petroleum ether); eleven amino acids and one dipeptide were isolated and identified in the GC-MS chromatogram (Table 3).

Nano-ESI-Chip-MS Analysis
All five hellebore fractions S 1 -S 5 were submitted to high-throughput positive nanoESI chip MS and screening under identical solution and instrumental parameters. The obtained mass spectra are illustrated in Figures 3-9, while the ion fragments and the corresponding amino acids and peptides (assignments to amino acids and peptide structures) are listed in Table 4.      Figures 3-9, while the ion fragments and the corresponding amino acids and peptides (assignments to amino acids and peptide structures) are listed in Table 4.     Figures 3-9, while the ion fragments and the corresponding amino acids and peptides (assignments to amino acids and peptide structures) are listed in Table 4.                   To elucidate the chemical composition of the hellebore fractions, we have developed a fully automated new mass spectrometry method based on nanoESI high-capacity ion trap (HCT), in a similar manner as described in the literature [28]. The mass spectrometry results confirmed the chemical structures detected by chromatographic techniques. Although, the developed chromatographic analysis method offered information about the great majority of the chemical species present in the hellebore extracts, mass spectrometry provided many more data on the complexity of the protein structures found in the studied hellebore fractions. Comparison of the screened mass spectra of the hellebore fractions S 1 -S 5 provided direct evidence of the variable composition of each fraction due, primarily, to the solvent employed. The investigation of thionin composition from the hellebore fractions S 1T and S 2T highlighted the same compounds identified by previous analytical methods. The information gained from this study corroborates with the existing data from the literature [29,30]. Our results revealed that the proposed methods proved to be useful tools for the separation and identification of individual compounds from complex natural mixtures.

Biological Activity Investigation
The usefulness of timelapse videomicroscopy in the investigation of cell behavior and biological activity of different pharmaceuticals was previously argued upon and proven [31,32]. Here, studies were conducted to assess, in a cancer cell line (HeLa), the effects of two fractions obtained from the alcoholic extract of Helleborus purpurascens. As shown in Figure 10, the cells treated with various concentrations of S 5 ( Figure 10A) and S 2T ( Figure 10D) failed to multiply. The number of mitoses significantly decreased for treated cells for all concentrations used of both S 5 ( Figure 10B) and S 2T ( Figure 10E). The number of dead cells did not spectacularly increase, although it varied significantly in comparison to untreated cells for both S 5 ( Figure 10C) and S 2T ( Figure 10F).

12
To elucidate the chemical composition of the hellebore fractions, we have developed a fully automated new mass spectrometry method based on nanoESI high-capacity ion trap (HCT), in a similar manner as described in the literature [28]. The mass spectrometry results confirmed the chemical structures detected by chromatographic techniques. Although, the developed chromatographic analysis method offered information about the great majority of the chemical species present in the hellebore extracts, mass spectrometry provided many more data on the complexity of the protein structures found in the studied hellebore fractions. Comparison of the screened mass spectra of the hellebore fractions S1-S5 provided direct evidence of the variable composition of each fraction due, primarily, to the solvent employed. The investigation of thionin composition from the hellebore fractions S1T and S2T highlighted the same compounds identified by previous analytical methods. The information gained from this study corroborates with the existing data from the literature [29,30]. Our results revealed that the proposed methods proved to be useful tools for the separation and identification of individual compounds from complex natural mixtures.

Biological Activity Investigation
The usefulness of timelapse videomicroscopy in the investigation of cell behavior and biological activity of different pharmaceuticals was previously argued upon and proven [31,32]. Here, studies were conducted to assess, in a cancer cell line (HeLa), the effects of two fractions obtained from the alcoholic extract of Helleborus purpurascens. As shown in Figure 10, the cells treated with various concentrations of S5 ( Figure 10A) and S2T ( Figure 10D) failed to multiply. The number of mitoses significantly decreased for treated cells for all concentrations used of both S5 ( Figure 10B) and S2T ( Figure 10E). The number of dead cells did not spectacularly increase, although it varied significantly in comparison to untreated cells for both S5 ( Figure 10C) and S2T ( Figure 10F). Both tested fractions (S5 and S2T) inhibited cell proliferation at significantly low concentrations, as shown by HeLa cell growth in the absence or presence of compounds ( Figures 10A,D). Cells that Both tested fractions (S 5 and S 2T ) inhibited cell proliferation at significantly low concentrations, as shown by HeLa cell growth in the absence or presence of compounds ( Figure 10A,D). Cells that were not treated proliferated and doubled their number after a 24 h monitoring. The number of observed mitoses was significantly higher in the absence of the tested compounds ( Figure 10B,E). It is noteworthy that the number of dead cells during the experimental period of 24 h, although significantly increased after treatment, still remained very low ( Figure 10C,F), indicating a reduced cytotoxic effect of both fractions S 5 and S 2T , isolated from the Helleborus purpurascens alcoholic extract. With regard to the dynamics of the identified effects, fraction S 5 arrested cell division almost completely after 6 h, when the highest concentration was used. For the medium and low concentration, the same effect was observed at about 12 h and beyond. A very interesting result was observed in terms of kinetics of the few mitoses still occurring in the treated samples. If for untreated HeLa cells a starting mitosis succeeded finalizing its cytokinesis stage in no more than half an hour, treated cells starting mitosis needed a significantly longer period until cytokinesis occurred, meaning more than 3 h ( Figure 11).
Accordingly, an untreated cell, spread on the culture surface ( Figure 11A), started a mitotic process ( Figure 11B) and accomplished cytokinesis after 20 min (Figure 11C), whereas the two daughter cells had already spread, 1 h after mitosis has begun ( Figure 11D). A treated cell ( Figure 11E) that entered mitosis ( Figure 11F) needed 200 min to accomplish cytokinesis ( Figure 11G), and 90 min more to spread after the mitotic event ( Figure 11H), meaning that it needed 4 h and 50 min to finalize a significantly, extremely slow division. This observation supports the conclusion that compounds in both S 5 and S 2T fractions exerted a major antiproliferative effect, recommending them as potential anticancer therapeutic agents. There are very strong evidences on the fact that in those fractions was not identified other molecules, except amino acids and tionins. However, further investigations are needed to assert if the identified biological activity is due exclusively to amino acids, thionins or to other small molecules, as yet not identified by the analytical techniques used.

13
It is noteworthy that the number of dead cells during the experimental period of 24 h, although significantly increased after treatment, still remained very low ( Figures 10C,F), indicating a reduced cytotoxic effect of both fractions S5 and S2T, isolated from the Helleborus purpurascens alcoholic extract. With regard to the dynamics of the identified effects, fraction S5 arrested cell division almost completely after 6 h, when the highest concentration was used. For the medium and low concentration, the same effect was observed at about 12 h and beyond. A very interesting result was observed in terms of kinetics of the few mitoses still occurring in the treated samples. If for untreated HeLa cells a starting mitosis succeeded finalizing its cytokinesis stage in no more than half an hour, treated cells starting mitosis needed a significantly longer period until cytokinesis occurred, meaning more than 3 h (Figure 11).
Accordingly, an untreated cell, spread on the culture surface ( Figure 11A), started a mitotic process ( Figure 11B) and accomplished cytokinesis after 20 min (Figure 11C), whereas the two daughter cells had already spread, 1 h after mitosis has begun ( Figure 11D). A treated cell ( Figure 11E) that entered mitosis ( Figure 11F) needed 200 min to accomplish cytokinesis ( Figure 11G), and 90 min more to spread after the mitotic event ( Figure 11H), meaning that it needed 4 h and 50 min to finalize a significantly, extremely slow division. This observation supports the conclusion that compounds in both S5 and S2T fractions exerted a major antiproliferative effect, recommending them as potential anticancer therapeutic agents. There are very strong evidences on the fact that in those fractions was not identified other molecules, except amino acids and tionins. However, further investigations are needed to assert if the identified biological activity is due exclusively to amino acids, thionins or to other small molecules, as yet not identified by the analytical techniques used.
Plant roots and rhizomes were collected in February 2011, in Arieseni, Alba (Romania) and identified by Dr. Dana Bobit (vice-president of the Romanian Ethnopharmacology Society, Dacia Plant SRL Brasov, Romania). Voucher samples: 00321 and 00322 were deposited at the Botanical Garden Cluj Napoca, Romania.

Chromatographic Techniques
The content of amino acids and peptides from the hellebore samples was analyzed by high performance liquid chromatography (HPLC 3000, Ultimate, Dreieich, Germany) using a photodiode array detector and an EZ: faast 4u AAA-MS Column (250ˆ3 mm ID). Qualitative analysis of free amino acids and peptides was performed on a GC-MS 7890A-5975C system (Agilent, Waldbronn, Germany) using the EZ: faast GC-MS free amino acids kit and ZB-AAA GC column (Phenomenex, Torrance, CA, USA). The employed analysis conditions were the standard conditions provided with the kit.

Spectroscopy Techniques
Mass spectrometry was conducted on a High Capacity Ion Trap Ultra mass spectrometer (HCT Ultra, PTM discovery) from Bruker Daltonics (Bremen, Germany). All mass spectra were acquired in the mass range m/z 100-3000, with a scan speed of 2.1 scan per second.

Isolation Strategy for Amino Acids and Peptides
The nature of the target bioactive compounds is influenced by the type of solvent employed. In general, a mixture of solvents was preferred to achieve the selective extraction of interested bioactive compounds from natural products. A preliminary sample was prepared using several solvents with different polarities (ethanol, dichloromethane, carbon tetrachloride, hexane, and n-butanol) in order to extract the corresponding phytochemicals from the hellebore extract. It was observed that in fractions with low and medium polarity (dichloromethane, carbon tetrachloride and hexane) lipophilic compounds were usually found, including: fatty acids, terpenes, steroids, peptides, depsipeptides, etc. High polarity fractions contained saponins, amino acids, alkaloids, sugars, etc. [24][25][26]33].

Derivatization of Standard Amino Acids
A pre-column derivatization method using PITC (phenylisothiocyanate) or Edman's Reagent was carried out. The obtained phenylisothiocyanate derivatives (PITC-amino acids) were analyzed by reverse-phase high-performance liquid-chromatography [34][35][36][37][38][39]. This method was chosen because the derivatization reagent reacts easily with all amino acids in an alkaline milieu and produces stable products. A sample (1.5 µmol) of standard amino acids is mixed with 1 mL derivatization reagent: pyridine: H 2 O (40:60) and 15 mg PITC. The obtained mixture is heated at 40˝C for 1 h. Then, 1 mL H 2 O is added. The excess of PITC is removed by washing four times with 2 mL of benzene. The aqueous phase is evaporated and dried in vacuum. The residues are dissolved in 1.5 mL methanol and analyzed on HPLC.

Thionins
For the separation of thionins, the solvent from a 50 mL hellebore hydroalcoholic extract was removed under vacuum and the resulting residue was extracted successively with petroleum ether (20 mL) and acetone (20 mL). Both isolated fractions-S 1T (petroleum ether) and S 2T (acetone)-were investigated by analytical methods to reveal the presence of the isolated compounds.

HPLC-DAD Separation Conditions
The separation was performed by isocratic elution at Flow Rate: 0.6 mL/min., Col.

Mass Spectrometry Analysis
Tandem mass spectrometry was carried out by collision-induced dissociation (CID) using He as the collision gas. For MS/MS sequencing, the precursor ions were selected within an isolation width of 2 u. Fully automated chip-nano ESI performed in a NanoMate 400 robot incorporating ESI chip technology (Advion BioSciences, Ithaca, NY, USA) coupled on a High Capacity Ion Trap Ultra mass spectrometer (HCT Ultra, PTM discovery, Bruker Daltonics). The robot was controlled and manipulated by the ChipSoft software (Advion BioSciences) operating under Windows. The position of the electrospray chip was adjusted to the sampling cone potential so as to give rise to an optimal transfer of the ionic species into the mass spectrometer. In order to avoid any contamination in all experiments, a glass-coated microtiter plate was used. Five µL aliquots of the working sample solutions were loaded onto the 96-well plate. The robot was programmed to take up the whole volume of sample, followed by 2 µL of air into the pipette tip and then deliver the sample on the inlet side of the microchip. Each nozzle has an internal diameter of 2.5 µm and under the given condition delivered a flow rate of about 200 nL/min. NanoESI process was initiated by applying voltages within 1.5 to 1.8 kV and a head pressure of 0.5-0.7 p.s.i. After spray initialization, infusion parameters were optimized: ESI voltage in pipette tip, voltage and desolvation gas flow. The values of Nano-ESI source parameter, ESI capillary, cone potential and desolvation gas (Nitrogen) were optimized to achieve an efficient ionization and produce the optimum transfer of ions in MS. Measurement parameters were: Capillary voltage 1 kV; Counter electrode voltage (cone voltage) 60 V; Acquisition time 2 min; Scan speed 2.1 scan/s; m/z 100-3000 mass range. The NanoMate HCT MS system was tuned to operate in the positive ion mode. This technique was chosen, as protein and peptide ionization shows high ionization efficiency in the positive ion mode. The source block, maintained at the constant temperature of 80˝C, provided an optimal desolvation of the generated droplets without a need of a desolvation gas. In all experiments, the desolvation gas was maintained at 30-50 liters per hour. For prevention of any cross contamination or carry-over, the pipette tip was ejected and replaced with a new one, after every sample infusion and MS analysis. All mass spectra were processed by the Data Analysis 3.4 software from Bruker Daltonik (Bremen Germany). Mass spectra were calibrated using sodium iodide as a calibrating agent. The accuracy on the determination of the average mass was 20 ppm, with a resolution of about 4000. Samples were dissolved in methanol at a concentration of about 5 pmol/µL. For an acquisition time of 2 min, the required volume of sample was about 2 pmols, a value which reflects a very high sensitivity analysis.

Study of the Biological Activity
HeLa cells were grown in DMEM-Ham F12 medium (Sigma-Aldrich, Munich, Germany), supplemented with 10% fetal bovine serum (Gibco, Basel, Switzerland) and 1% antibiotic-antimycotic solution (Lonza, Cologne, Germany). They were seeded at a density of 10 4 cells/well in Hi-Q 4 35 mm dishes (Ibidi GmbH, Martinsried, Germany) After cell attachment (2 h), the medium was replaced with a fresh volume supplemented with various concentrations of S 5 and S 2T lyophilized fractions (1, 2 and 4 µg/mL respectively, dissolved in culture medium). For the control group, the medium was aspirated and replaced with fresh standard medium. The dishes were then placed inside the Biostation IM equipment (Nikon Corp., Kawasaki, Japan), a mini-incubator with an incorporated optical system and a CCD camera for timelapse imaging. The cells were monitored for 24 h and images were collected every 10 min. To assess cell viability and proliferation, viable cells and mitoses were counted. For each experimental condition, images were collected from 5 different microscopic fields and the experiments were repeated three times. The wells which contained only the cells, without the added compounds, were considered as control.

Conclusions
This study was designed to investigate the chemical composition of extracts of a valuable medicinal plant, the hellebore, with therapeutic effects known since long ago in traditional medicine. The amino acids and peptides contained in a hellebore hydroalcoholic extract were identified and isolated. An appropriate isolation strategy was developed that can be applied for the isolation of such compounds from complex mixtures. The presented results indicate that the new isolation strategy and the feasibility of the developed characterization method show great potential as efficient separation methods for amino acids and peptides from natural products. The methods used in our study to isolate various fractions of the hellebore extract did not affect the antiproliferative activity, as shown by treating HeLa cells and monitoring their behavior by time lapse videomicroscopy.