Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography

Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography

G Model ARTICLE IN PRESS CHROMA-460422; No. of Pages 11 Journal of Chromatography A, xxx (xxxx) xxx Contents lists available at ScienceDirect Jou...

3MB Sizes 0 Downloads 39 Views

G Model

ARTICLE IN PRESS

CHROMA-460422; No. of Pages 11

Journal of Chromatography A, xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography Simone dos Santos Grecco a,d,∗ , Emmanuel Letsyo b , André Gustavo Tempone c , João Henrique Ghilardi Lago d , Gerold Jerz e,∗∗ a

Anhanguera University of São Paulo, Biotechnology and Innovation in Health and Pharmacy Postgraduate Programs, 05145-200, São Paulo, Brazil Department of Food Science and Technology, Ho Technical University, P.O Box HP 217, Ho, Ghana c Center of Parasitology and Mycology, Adolfo Lutz Institute, São Paulo, 01246-902, Brazil d Center of Natural Sciences and Humanities, Federal University of ABC, 09210-580, Santo André, Brazil e Institute of Food Chemistry, Technische Universität Braunschweig, Schleinitzstrasse 20, 38106 Braunschweig, Germany b

a r t i c l e

i n f o

Article history: Received 14 March 2019 Received in revised form 29 June 2019 Accepted 3 August 2019 Available online xxx Keywords: Nectandra leucantha Lauraceae Neolignans Off-Line ESI-MS/MS Semi-preparative high-performance countercurrent chromatography Spiral-coil countercurrent chromatography

a b s t r a c t Nectandra leucantha (Lauraceae) is a tree indigenous to the tropical Atlantic forests of Brazil, one of the most biodiverse flora hotspots worldwide. This plant species contains high concentrations of neolignan and dehydrodieugenol derivatives that express significant in-vitro activities against various parasite strains. These activities are however responsible for severe tropical human infections, such as Leishmaniasis (Leishmania spp.) and Chagas disease (Trypanosoma cruzi), which have been classified by the World Health Organization (WHO) as Neglected Tropical Diseases (NTDs). In order to optimize the isolation process for these target metabolites, n-hexane extract of the leaves was separated by means of semi-preparative high performance countercurrent chromatography (HPCCC) and scale-up spiral-coil countercurrent chromatography (sp−CCC) systems. Several biphasic solvent mixtures were evaluated for their partitioning effects on neolignans, resulting in the selection of an optimized system n-hexane – ethylacetate – methanol – water (7:3:7:3, v/v/v/v). The chromatographic experiments on the HPCCC and sp-CCC were run in the head-to-tail mode with 500 mg and 16 g injections, respectively. For specific and multiple metabolite detection, the recovered CCC-fractions were off-line injected, in the sequence of recovery, to an electrospray mass-spectrometry (ESI-MS/MS) device. A projection of the single ion traces of the target compounds, in the positive ionization mode at a scan range of m/z 100–1500, located chromatographic areas where the co-elution effects occurred and pure target metabolites were present. Five major target neolignans were specifically detected, which enabled the accurate pooling of CCC-fractions for an optimum recovery of the metabolites. The direct comparison of the performance characteristics of the two CCC-devices, with very different mechanical designs was achieved by the conversion of the time axis into a partition ratio (KD ) separation scale. As a result, the compound specific KD -elution values of the target neolignan were determined in high precision, while the comparison of the calculated separation factor (␣) and resolution factor (RS ) values revealed a superior separation performance for the HPCCC system. Also, the reproducibility of detected metabolites in the two CCC experiments was confirmed by small variations (KD ±0.1). Neolignan target compounds with anti-parasite activities were successfully isolated in the 100 mg to 4 g range in a single lab-scale countercurrent chromatographic process step. © 2019 Published by Elsevier B.V.

1. Introduction ∗ Corresponding author at: Anhanguera University of São Paulo, Biotechnology and Innovation in Health and Pharmacy Postgraduate Programs, 05145-200, São Paulo, Brazil. ∗∗ Corresponding author. E-mail addresses: [email protected] (S. dos Santos Grecco), [email protected] (G. Jerz).

Global scientific efforts have been pursuing alternative and more sustainable process protocols, especially through the concepts of Green Chemistry and IUPAC Clean Technology [1,2]. The implementation of these concepts in the field of drug discovery from natural resources could be achieved through rationalization,

https://doi.org/10.1016/j.chroma.2019.460422 0021-9673/© 2019 Published by Elsevier B.V.

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11 2

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

and optimization of extraction, fractionation and isolation procedures. In this context, we highlight the importance of countercurrent chromatography (CCC) as a technique with a better ratio of solvent consumption and product recovery compared to classical solid phase chromatography [3]. Besides, the former provides options for recycling the stationary and mobile phases, as well as the use of green solvents that could drastically reduce the harmfull effects to environment [4,5]. Since 1970, CCC has been intensively applied in lab-scale studies to recover natural products on preparative scale, however in the past decade, the technique has been redeveloped and adjusted to levels of industrial-scale application [6,7]. Overall advantage of CCC separations is based on the principle that all analytes or bioactive target compounds in solution are distributed between mobile and stationary phases, and that the effects of chemisorptive compound loss is negligible relative to the classical solid phase chromatography. Furthermore, the separation mechanism of CCC involves the rigorous phase cycling, consisting of mixing and separating effects of the liquid phase layers that is induced by strong and fast alternating centrifugal force fields [8]. The separation occurs in an open column along polytetrafluorethylene tubings winding in multi-layers on a holder, coil-column system. In commercial CCCdevices, up to three coil-systems can be set up in a sequence, which usually rotate synchronously in a planetary rotation field to induce the Archemedian screw force that immobilizes the stationary phase on the separation column [8]. Commercially available machines are classified, according to their technical design, as J-type CCC centrifuges (e.g. HPCCC system) which are operated at high rotational speeds between 800 to 1600 rpm. Conversely, in the non-commercial spiral-coil CCC prototype, the velocity is much lower between 80 to 280 rpm and the mode of separation is not comparable to the high-speed systems with planetary double rotations (i.e. HSCCC and HPCCC). Moreover, the rotation of the spiral-device is in a single axis and the tubing set-up is in a spiral geometry. In addition, the device consists of convoluted-shaped Teflon tubings with a large internal diameter supporting the stationary phase retention. Four single spiral tubes, located in the separation unit, are connected in sequence, ten units (i.e. 550 mL each) are connected in series totalling 5500 mL of the coil-column volume. In both CCC systems applied, separations were achieved depending on the metabolite and solvent system of the specific partition ratio values (KD ). The accurate determination of elution/retention volumes were achieved by off-line injections of the CCC-fractions, in the sequence of recovery, to the ESI-MS spectrometer containing all relevant mass-spectrometric data in a single data file. The selection of single ion-traces was used to monitor the exact elution profiles of separated target metabolites, as well as the projected chromatographic areas of the co-eluted or potentially pure metabolites in their respective fractions [9]. The slow rotating and larger lab-scale spiral-coil CCC system has been previously applied to the recovery of bioactive natural products in combination with HSCCC and HPCCC fractionations [10–12]. In our present preparative CCC isolation study, we state the proof-ofconcept that all-liquid countercurrent chromatography is able to recover larger amounts of neolignan target compounds (eugenol dimers, cf. Figs. 1 and 2) from the crude leaf extract of Nectandra leucantha Nees & Mart (Lauraceae) in one single chromatographic step. Recent studies on this tree from the tropical Atlantic coastal forest zones in Brazil, notably between Rio Grande do Sul and Rio de Janeiro [13], revealed significant in-vitro activities against insect vector transmitted parasites, such as Leishmania donovani, Leishmania chagasi, and Trypanosoma cruzi (Chagas disease) [14–16]. According to the World Health Organization (WHO) [17,18] these infections are defined as Neglected Tropical Diseases (NTDs) which affect hundred thousands to million people worldwide [19,20]. Actual available chemotherapy in these cases are extremely dif-

ficult to handle. The major drawbacks are: frequent severe side effects of the medications and loss of efficacy as result of rising drug resistance over the past decade, inappropiate research efforts by industry, and lately, missing approvals for novel medications [21]. Therefore, the use of all-liquid countercurrent chromatography could strongly support the continuous screening of biodiverse floral sources for drug development in the field of NTDs control. 2. Experimental 2.1. Reagents In the solvent extraction of Nectandra leucantha leaves, p.A. quality n-hexane (Casa Americana, São Paulo, Brazil) was used. All chromatographic and partitioning procedures were performed using HPLC grade solvents: n-hexane, acetonitrile (ACN), cyclohexane, iso-propanol (i-Prop), ethyl acetate (EtOAc), and methanol (MeOH) (LC–MS-grade, Fisher Scientific, Loughborough, UK). Ethanol without additives, in European Pharmacy Quality (EuAB) (Carl Roth, Karlsruhe, Germany) and ultra-pure water, Nanopure® (Barnstead, USA) were used in the CCC-solvent preparation, as well as LC-ESI-MS analysis. 2.2. Preparation of neolignan enriched extracts from the leaves of Nectandra leucantha Nectandra leucantha leaves were collected at the Parque Ecológico do Perequê (March 2014) located in Cubatão city at São Paulo state, Brazil. The botanical collection and identification were carried out in collaboration with MSc. Euder G.A. Martins. A voucher specimen was deposited in the Herbarium of the Institute of Biosciences, University of São Paulo under the code EM357 (SPF00215113). About 2.55 kg of previously shade-dried and milled leaves of N. leucantha were subjected to extraction with n-hexane by exhaustive maceration. After vacuum evaporation of the solvent, 84.5 g of crude extract (NLH) was obtained. In order to eliminate interfering lipophilic materials (e.g. parafins, long chain waxes, triterpenoids), a substantial amount (51.9 g) of the n-hexane extract was suspended in n-hexane (500 mL), and liquid-liquid extracted with acetonitrile (3 × 250 mL) in a separatory funnel. The recovered and vacuum-dried acetonitrile phase yielded 31.6 g of neolignan enriched fraction (NEF) containing the principal neolignan derivatives 1–5 constituted of dimers of eugenol (cf. Figs. 1 and 2). 2.3. TLC analysis TLC-analysis of neolignans, coupled with the partitioning of the metabolite in the biphasic solvent system were performed on silica gel 60 plates F254 (Merck KGaA, Darmstadt, Germany). The target compound was eluted with equal amounts of n-hexane and EtOAc and later visualized with UV light (␭ 254 and 365 nm) by spraying with sulfuric acid – glacial acid – anisaldehyde – universal reagent [22] while applying flash-heat on a hot plate at 105 ◦ C. 2.4. Solvent system selection for countercurrent chromatography In order to evaluate the distribution properties of the bioactive neolignans between the biphasic solvent systems (SoSy), 5 mg of NLH or NEF was dissolved in 6 mL of the SoSy after which the shake flask technique [8] was applied. The distribution of the compounds was monitored by silica gel TLC, as previously described. For the lipophilic crude sample (NLH), the tested solvent systems comprised: n-hexane – MeOH (1:1 - SoSy1); n-hexane – ACN (1:1 - SoSy2) and cyclohexane – ACN (1:1 - SoSy3). While the NEF sample was tested with the following different biphasic solvent systems: n-hexane – MeOH (1:1 - SoSy1); n-hexane – EtOH – water

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

3

Fig. 1. Semi-preparative HPCCC experiment (injection: 500 mg) in the elution-mode (flow 5.0 mL/min) with ESI-MS/MS off-line injection of fractions in the sequence of recovery to monitor target neolignans from N. leucantha visualizing co-elution effects by selected single ion traces in the positive ionization mode. Used biphasic CCC-solvent system: HEMWat 7 : 3 : 7 : 3 (v/v/v/v). ESI-MS injection amount of fractions 9 ␮L with changing intervals: every tube fraction (t12 – t33), every third (t34 – t51), and every tube (t52 – t74).

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11 4

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

Fig. 1. (Continued)

(6:5:1 - SoSy4); n-hexane – i-propanol – water (3:2:2 - SoSy5 and 3:3:2 - SoSy6), and n-hexane – EtOAc – MeOH – water (HEMWatsystems 7:3:7:3 - SoSy7, and 6:1:6:1 - SoSy8). The most suitable solvent system was found to be SoSy7, based on the compound partition effects evaluated by thin layer chromatography, coupled with the observed settling times in the shake-flask assay [8].

2.5. HPCCC, spiral coil CCC apparatus and peripheral devices 2.5.1. Semi-preparative HPCCC (125 mL coil-volume) In the semi-preparative experiment, neolignans from N. leucantha were separated on a high-performance multilayer coil planet J-type HPCCC centrifuge (Spectrum, Dynamic Extractions, Gwent, UK) with two self-balanced coil bobbins and ß-value range of 0.52 - 0.86 (equation: ß = coil radius r /revolution radius R) [8]. Two semi-preparative tube column systems (62.5 mL, 1.6 mm bore size i.d.) made of polytetrafluoroethylene (PTFE) were connected in series to yield a coil column volume VC of 125 mL. The separation was performed in the head-to-tail (reversed-phase) mode using the evaluated solvent system, with the more dense aqueous phase layer as mobile phase. The head-end of the coil configuration was located at the periphery of the coils. The separation was performed at the machine’s maximum velocity of 1600 rpm (g-force level 243). The temperature in the HPCCC was kept constant at 30 ◦ C using a liquid cooling thermostat (RC6, Lauda Dr. Wobser GmbH & Co. KG, Lauda-Königshofen, Germany). After equilibration of the different solvents in a separatory funnel, the two phase layers were separated shortly before use. The freshly prepared solvents were pumped with a preparative LC

pump (solvent delivery system, K-501, Knauer Wissenschaftliche Geräte GmbH, Berlin, Germany). The flow rate for pumping the mobile phase during the HPCCC experiment was set to 5.0 mL/ min. The neolignan enriched fraction (NEF, 500 mg) was dissolved in aliquot volumes of both phase layers of the SoSy7 (HEMWat 7:3:7:3, v/v/v/v) with a total volume of 5.0 mL. The measured stationary phase retention (SF ) at the hydrodynamic equilibrium (i.e. with the emergence of mobile phase) was 82.0% (not corrected with dead volume of connecting periphery tubing). The sample was dissolved in 2.5 mL aliquot volumes of the phase layers, filtered over a Chromafil Xtra GF-100/25 fiberglass membrane disc filter (1 ␮m pore size, 25 mm i.d., Macherey & Nagel, Düren, Germany), and injected through a 5.0 mL sample loop to the separation column by a manual low-pressure sample injection valve (Rheodyne, Cotati, CA, USA). Detection was done at ␭ 210 nm with a UV-detector K-2501 (Knauer Wissenschaftliche Geräte GmbH Berlin, Germany). The fraction collector SuperFrac type B racks (Pharmacia, Uppsala, Sweden) collects fractions during the elution process in every min. Aliquots of the collected fractions were filled, in sequence of collection, to HPLC vials (cf. 2.9.3. and 2.9.4.) to perform the off-line metabolite elution profile by sequential ESI-MS/MS injections (cf. Fig. 1).

2.5.2. Preparative lab-scale spiral-coil CCC (5500 mL coil-volume) The most suitable solvent system, HEMWat (SoSy7), was used in the separation of the bioactive neolignans in both the HPCCC and scale-up spiral-coil countercurrent chromatography (sp−CCC) systems. The latter is a prototype, which was originally designed by Dr. Yoichiro Ito (National Heart, Lung, and Blood Institute, National

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

5

Fig. 2. Spiral−CCC experiment (injection: 16 g) in the elution- (flow 15.0 mL/min), and extrusion- mode (N2 -gas, 7 bar) with ESI-MS/MS off-line injection of fractions in the sequence of recovery to monitor target neolignans from N. leucantha visualizing co-elution effects by selected single ion traces in the positive ionization mode. Used biphasic CCC-solvent system: HEMWat 7 : 3 : 7 : 3 (v/v/v/v). ESI-MS injection amount of fractions 7 ␮L with changing intervals: every second tube fraction (t1 – t21), every fourth (t22 – t193), and every tube (t197 – t333).

Institutes of Health, Bethesda, MD) and Dr. Edward Chou (formerly of PharmaTech Research Corp., Baltimore, MD, USA). The system consists of a total coil column volume of 5500 mL (cf. Suppl. Fig. 1) and has been applied in previous natural product recovery studies [10–12,23]. The spiral-coil column is equipped with 10 separation tube-units (volume of each: 550 mL) all connected in series. The whole system performs a single axis rotation, with one tube-unit consisting of four convoluted shaped polytetrafluorethylene tubings (8.5 mm i.d.) wound in a spiral-coil set-up. All tubings start from the center and wind to the periphery, and back to the center again (labelled in colors green, blue, red, black to be connected with the next section). The diameter of the spiral-coil rotor is 50 cm. Supplementary Fig. 1 shows the system being operated at 15.0 mL/min mobile phase flow rate at rotational speed of 280 rpm.

For direct observation during the rotation, the front cover of the device was removed and observed by a high-speed camera. The high flow rate was not inducing a spin or rotational effect of the phase layers in the convoluted shaped tubings. The strong centrifugal g-force field eliminated a thorough mixing of phase layers. At the points where the column-tubes were fixed to the spiral-coil CCC-corpus (cf. Suppl. Fig. 1), the round circle shape was altered, and a higher volume of mobile phase was observed at these sections with some suspected phase mixing effects. Overall, the mass transfer of metabolites in the sp−CCC seems to occur only at the small surface area of the interfacial borders of the upper and lower phase layers. A vigorous mixing which induces solvent droplets, as observed for high velocity planetary J-type CCC systems, was not noticed in the analysis.

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

6

The separation was performed in the L-I-H mode (i.e. lower and more dense aqueous layer was the mobile phase, eluted from the center to the periphery) in the head-to-tail mode similar to the HPCCC separation. As seen in Suppl. Fig. 1, the chirality of the spiral tubings end with orientation to the right side, and the spinning direction was clockwise (CW). The mobile phase flow rate for the separation with n-hexane–EtOAc–MeOH–Water (7:3:7:3, v/v/v/v) was set to 15.0 mL/min. The automatic fraction collection was performed with larger glass tubes in type C-racks (SuperFrac, Pharmacia, Sweden) collecting 45.0 mL/ tube fraction in 3.0 min. In order to shorten the chromatographic elution time, higher flow rates of up to 25.0 mL/min would be feasible for the sp−CCC unit, but could potentially reduce the chromatographic resolution, as the stationary phase retention (SF ) could decrease unprofitable. For the sample preparation, 16 g of the dried neolignan enriched fraction (NEF) was dissolved in 130 mL of the selected solvent system and injected to the spiral-coil chromatograph directly oncolumn with a T-valve connector. The system was adjusted to a rotational speed of 280 rpm (clockwise), allowing the injection to occur immediately after the pumping of the mobile phase (i.e. separation without classical column equilibration). Consequently, the chromatographic parameter value of mobile phase take-up VM was not determined. After pumping 2700 mL of the mobile phase, the ‘break-through’ occurred with the elution of the most polar neolignan components. The sp−CCC chromatogram was annotated by displaying recovered tube fraction numbers, retention/elution volumes VR , and a KD -value based scale (cf. Fig. 2).

correctedV M = V M –VExt = 22.5 mL–7 mL = 15.5 mL

(3)

CorrectedV S = VC –correctedVM = 109.5 mL

(4)

After obtaining the corrected VS , the correct stationary phase retention factor SF was calculated as 87.6% (corrected value). This can be seen as the real percentage of stationary phase existing in the CCC-column being used for the experiment. It is important to note that a high SF value directly correlates to a higher resolution and efficiency of the CCC-separation. correctedS F = correctedVS /VC

Prior to the drying of the recovered CCC-fractions, aliquot and representative volumes were taken for the off-line ESI-MS profiling experiments (cf. Section 2.9.). The HPCCC fractions from the elution process containing n-hexane, ethyl acetate, methanol and water were gently evaporated with a SpeedVac Plus concentrator equipped with a rotor for fraction collector tubes (SC210A, and refrigerated vapor trap RVT 400, Thermo Savant, Holbrook, NY, USA). The recovered fractions (i.e elution and extrusion fractions) from the sp−CCC were pooled based on the results of the ESI-MS/MS injection profiling (cf. Section 2.9.3, 2.9.4.), and dried in 500 mL flasks by a vacuum rotary-evaporator, and final lyophilized (Martin Christ Gefriertrocknungsanlagen, Osterode, Germany). 2.7. Calculation of formula parameters from the HPCCC and spiral-coil experiments 2.7.1. HPCCC parameters As the study was conducted with two CCC-devices with completely different mechanical design, their separation performances were not compared with respect to their elution time, as usually displayed in published CCC-experiments. The elution time was converted over elution/retention volumes VR into their respective partition ratio values KD (cf. Eqs. 1–6). The VR values for the neolignan metabolites were determined with high accuracy by the off-line ESI-MS injection of consecutive fractions as described below (cf. 2.9.). (1)

(5)

The metabolite and solvent system specific partition ratio KD values were calculated by the Eq. (6) (cf. Table 1). K D = (VR –correctedVM )/correctedVS

(6)

The separation factor ␣ and resolution factors RS depend on the distances between peaks and the peak widths that will be compared. The calculation of both factors depend on the determined KD values. ˛ = K D2 /KD1 (withK D2 > KD1 )

2.6. Drying of HPCCC and spiral-coil CCC fractions

RetentionvolumeV R = elutiontime[min] × flowrate[mL/min]

VC : coil column volume/capacity VM : volume of mobile phase take up to the coil at equilibrium of HPCCC SF : stationary phase retention KD : partition ratio The SF -value was corrected by the extra column volume VExt (7 mL) of the connecting periphery tubing in the HPCCC set-up, using Eqs. (3–5)

(7)

To calculate the resolution factor (RS ) of the two peaks together with their respective KD values, the measured peak widths at the baseline were used, as seen in the next formula: RS = 2(K D2 -KD1 )/(W 2 +W1 )

(8)

Wn : peak width at baseline 2.8. Spiral-coil CCC parameters In the case of the sp-CCC separation, the procedure to reach the state of the hydrodynamic phase equilibrium was not implemented, as usually applied in HPCCC experiments. This would have required an enormous additional consumption of mobile phase solvent. After charging the system with stationary phase (5500 mL), the dissolved neolignan sample was directly on-column injected. The low volume of external tubing (< 20 mL) was neglected, while the pumped volume (2700 mL) at which the break-through of mobile phase occurred was used as starting point of the VR calculation. At this stage, the most polar neolignans (cf. Section 2.9.4, 1 and 2) eluted with a starting KD -value of 0.49. K D = VR (retentionvolume)/V C (totalcoil-columnvolume)

(9)

As the extrusion-mode was performed with pressurized nitrogen gas (7 bar) (i.e. for solvent saving reasons), the filling heights of the fractions in the tubes were inconsistent. As a result, a valid calculation of KD -values in the recovery range of neolignan 5 was not realized. All specific chromatographic key-parameter values of the HPCCC and sp-CCC experiments are given in Tables 1 and 2 .

The value for SF of the solvent system SoSy7 was determined by Eq. (2) using VC (125 mL), and VM (22.5 mL) resulting in a SF value of 82.0%.

2.9. Off-line ESI-MS/MS injection analysis for HPCCC and spiral coil−CCC molecular weight profiling

V S = (V C -V M )

2.9.1. General procedures For a target-guided detection of neolignan metabolites in the preparative HPCCC and sp-CCC experiments, aliquot volumes of

VS : retained experimental stationary phase volume

(2)

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model

ARTICLE IN PRESS

CHROMA-460422; No. of Pages 11

S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

7

Table 1 Compound specific partition ratio KD -values, KD -ranges, retention volumes VR , determined on semi-preparative HPCCC and spiral CCC using selected single ion traces from off-line ESI-MS injection profiling. The calculated KD off-sets between semi-preparative and spiral CCC-experiments are given. No.

Selected ion traces of neolignans m/z :[2M + Na]+ , [M + Na]+ , [M+H]+

Semi-preparative HPCCC, Peak range, Tube fractions, Elution volume VR [mL]

KD range / KD width

KD Mean value

Spiral coil CCC, Peak range, Tube fractions, Elution volume VR [mL]

KD range / KD width

KD Mean value

KD [semi-prep KD width – spiral KD width]

1

m/z 707, 365, 325

t13 – t14/ 65 – 70

0.45 – 0.50/ 0.05

0.475

0.51 – 0.60 0.11

0.567

−0.09

2

m/z 721

t15 – t16 / 75 – 80

0.54 – 0.59 / 0.05

0.565

0.51 – 0.79 0.28

0.65

−0.085

3a

m/z 735, 379, 339

t15 - t17/ 75 - 85

0.54 – 0.63 / 0.09

0.59

t3 – t13 / 2835 – 3285 t3 – t37 / 2835 4365 t3 - t41 / 2835 4545

0.54 – 0.83 0.29

0.685

−0.095

3b

t19 - t23/ 95 - 115

0.73 – 0.91 / 0.18

0.82

t19 – t21 / 95 – 105

0.73 – 0.82 / 0.09

0.775

4b

m/z 735, 379, 339 isomer lower concentrated m/z 675, 349, 327 isomer lower concentrated m/z 675, 349, 327

0.95 – 1.27 / 0.31

1.11

0.54 – 1.91 1.37

1.225

−0.115

5

m/z 703, 363, 341

t24 – t31 / 120 – 155 t51 – t58 / 255- 290

2.19 – 2.51 / 0.32

2.35

X > 2.13





4a

t41 – t173 / 4545 - 10,485 Extrusion with N2 gas – no valid volumes, t201 – t323 / X > 11,700

Table 2 Comparison of values of separation factor ␣ and resolution factor RS of some selected metabolite pairs from semi-preparative HPCCC and spiral-coil CCC. Separated Compound pairs: identities as m/z values

Semi-prep. HPCCC ␣-value

Spiral-CCC ␣-value

␣ [semi-prep ␣ – spiral ␣]

Semi-prep. HPCCC resolution factor RS

Spiral-CCC resolution factor RS

RS [semi-prep. RS – spiral RS ]

1/2 1 / 3a 1 / 3b 1 / 4a 1 / 4b 1/5 2 / 3a 2 / 3b 2 / 4a 2 / 4b 2/5 3a / 3b 3a / 4a 3a / 4b 3a / 5 3b / 4a 3b / 4b 3b / 5 4a / 4b 4a / 5 4b / 5

1.19 1.24 1.73 1.63 2.34 * 4.95 1.04 1.45 1.37 1.96 4.16 1.39 1.31 1.88 3.98 1.06 1.35 2.87 1.43 3.03 2.12

1.15 1.21 ** ** 2.16 *** 1.05 ** ** 1.88 *** ** ** 1.78 *** ** ** *** ** *** ***

+0.04 +0.03 ** ** +0.18 *** −0.01 ** ** +0.08 *** ** ** +0.10 *** ** ** *** ** *** ***

1.80 1.64 3.00 1.25 3.53 *** 0.36 2.21 3.00 3.03 *** 1.70 2.06 2.60 *** ** ** *** ** *** ***

0.425 0.59 ** ** 0.93 *** 0.12 ** ** 0.70 *** ** ** 0.65 *** ** ** *** ** *** ***

+0.725 +1.05 ** ** +2.60 *** +0.24 ** ** +2.33 *** ** ** +1.95 *** ** ** *** ** *** ***

* Principal compounds and best achieved ␣-, and RS -values (above 1.75) are written in bold numbers. ** Neolignan isomer pairs 3a – 3 b and 4a – 4b were not resolved in the off-line ESI-MS spiral-coil profile, calculation was done with main isomer from HPCCC. *** Neolignan 5 recovered in the extrusion mode of the spiral−CCC with uncertain KD -value. Determination of ␣- and RS -calculation was not possible. **** Sufficient ␣-values and RS -values (values > 1.5) in bold numbers. Formulas for ␣ = KD2 / KD1 (whereas KD2 > KD1 ). Formula for Rs = 2 (KD2 - KD1 ) / (W2 + W1 ).

recovered fractions were off-line injected to an ESI-MS/MS-ion-trap mass-spectrometer (HCT-Ultra ETD II, Bruker Daltonics, Bremen, Germany) in the sequence of chromatographic elution. By selecting the full base-peak chromatogram trace (BPC) and single ion target traces, the preparative chromatographic results were projected as molecular weights, which guided the accurate fractionation process. The acquired MS-data, including MS/MS fragmentation of seven precursor ions were recorded in one data file. For sample preparation, aliquot volumes from CCC-fractions (i.e. elution and extrusion modes) were measured and filled to the HPLC vials. The samples were stored frozen two days (−30 ◦ C) prior to the offline ESI-MS/MS profiling. The off-line injections to the ESI-MS/MS were conducted by an independent and programmable autosam-

pler system (AS-2000A, Merck-Hitachi, Tokyo, Japan). The injected fractions were delivered to the ESI-MS device by a HPLC-pump (binary pump, G1312 A, 1100 Series, Agilent, Waldbronn, Germany) at a flow rate of 0.5 mL/ min, and a make-up solvent system that is composed of ACN/ H2 O (9:1, v/v). The semi-polar character of the chosen solvent mixture guaranteed the fast delivery of metabolites from the injection loop to the MS-system with satisfactory injection signal widths. The time interval between the re-occurring injections was set to 2 min. Every observed injection peak corresponded to one injected vial from the CCC experiments. The selected ESIMS traces (Figs. 1 and 2) displayed the mass spectrometry profile information of the compounds located in their respective CCC fraction(s).

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11 8

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

The selected CCC-fractions were injected as ion abundance reference guides in a pre-experiment to evaluate the adequacy of the injected volumes before starting the complete profiling experiment. This routine procedure avoided injections with far too high sample that could potentially transfer loads to the mass-spectrometer. The highest ion intensities in the generated base-peak chromatograms (BPC) at the scanning range of m/z 100–1500 was observed at 1.0 × 109 , presenting the maximum tolerable level. The neolignan compounds showed a good ion response in the positive ESI-ionization mode, thus leading to only trace material deposited on the ESI-MS spray shield after the termination of the profile experiment. Also, the intention of using slightly higher injection volumes was to monitor potential minor metabolites and their respective MS/MS fragmented data.

2.9.2. ESI-MS/MS parameter settings The use of a positive ionization mode for the detection of neolignan metabolites in HPCCC and sp-CCC fractions resulted in a good ion responses of the generated ion species [M+H]+ , [M + Na]+ and [2M + Na]+ . The scan-range was set from m/z 100 to 1500 while a rapid ‘ultra’-mode with a mass scanning rate of 26.000 m/z per second was chosen. Drying gas was nitrogen (flow rate 10.0 L/min, 320 ◦ C), and nebulizer pressure was set to 60 psi. Ion optic parameter were set as follows: ionization voltage HV capillary - 3500 V; HV end plate off set −500 V; trap drive 55.6; octopole RF amplitude 187.1 Vpp; lens 2–60.0 V; Cap Ex +115.0 V; max. accumulation time 200 ms; averages 5 spectra; trap drive level 100%; target mass range m/z 500; compound stability 80%; smart ICC target 100,000; ICC charge control ‘on’ and smart parameter setting ‘active’. To monitor co-elution of metabolites, seven most abundant precursor ions were selected to obtain specific ESI-MS/MS fragmented data. The MS/MS fragmented amplitude value was set to 1 V.

2.9.5. Off-line ESI-MS/MS data of targeted neolignans All ESI-MS/MS experiments were performed in the positive ionization mode. The generated quasi-molecular ion signals; [2M + Na]+ , [M + Na]+ , [M+H]+ and [M−18+H]+ were used to characterize the five target neolignans in the recovered CCC fractions. In general, the observed [2M + Na]+ signals displayed the highest ion intensity values (cf. Figs. 1 and 2). Neolignan target compounds and their isomers were better resolved by the HPCCC experiment. In fact, the derived off-line injections from this experiment delivered more specific ESI-MS/MS data that were subsequently displayed: 1-[7-hydroxy-8-propenyl]-3-[3 -methoxy-1 -(8 -propenyl)phenoxy]-4-hydroxy-5-methoxy-benzene (1) [M+Na]+ at m/z365, [M−18+H]+ m/z 325, MS/MS (325) 177, 161, [2M + Na]+ m/z 707, MS/MS (707) 365; unknown neolignan (2) [M +Na]+ at m/z 721, MS/MS (721) 379, 365; 1-[7-hydroxy-8-propenyl]-3-[3 methoxy-1 -(8 -propenyl)-phenoxy]-4,5-di-methoxy-benzene (3a) [2M + Na]+ at m/z 735, MS/MS (735) 379, [M + Na]+ m/z 379, [M-18+H]+ at m/z 339, MS/MS (339) 298, 191, 176, 163, 151, 133. The minor unknown isomer (3b) with [2M + Na]+ at m/z 735, MS/MS (735) 379, [M + Na]+ m/z 379, [M+H]+ m/z 357, MS/MS (357) 339, 270, 193, 179, 165. The low concentrated isomeric structure dehydrodieugenol 4a [2M + Na]+ at m/z 675, MS/MS (675) 349, [M + Na]+ m/z 349, MS/MS (349), 333, 317, 308, 184. The neolignan 1-(8-propenyl)-3-[1 -(8 -propenyl)-3 -methoxyphenoxy]-4-hydroxy-5-methoxy-benzene–dehydrodieugenol B (4b) was seen with [2M + Na]+ at m/z 675, MS/MS (675) 349, [M+H]+ m/z 327, MS/MS (327) 295, 285, 256, 179, 153, 147, 119; 1-(8-propenyl)-3-[1 -(8 -propenyl)-3 -methoxy-phenoxy]-4,5di-methoxy-benzene (dehydro-dieugenol B methylether) (5) [2M + Na]+ m/z 703, [M + Na]+ 363, [M+H]+ m/z 341, MS/MS (341) 299, 271, 193, 175, 163, 133, 103.

3. Results and discussion 2.9.3. Semi-preparative HPCCC -off-line ESI-MS/MS profile In the sample preparation, 200 ␮L of the HPCCC fractions were taken and diluted with 1.0 mL EtOH in the HPLC vials. After the HPCCC run, the CCC-fractions were split into three different consecutive sections. The first section comprised of every recovered tube fraction from t12 to t33, while the second section consisted of every third tube (t34 – t51), and every tube (t52 – t72) was considered for the final section (cf. Fig. 1). About 9 ␮L of the selected fractions from the three sections were injected to the ESI-MS (cf. Fig. 1). Based on the retention volumes VR , the KD -based chromatographic scale was calculated and used for the recovery of neolignan.

2.9.4. Spiral-coil CCC -off-line ESI-MS/MS profile About 500 ␮L of the sp−CCC fractions, collected during the elution mode, were diluted with 1.0 mL ACN and transferred to the HPLC vials. The chosen chromatographic area for the ESI-MS injection profile started with the break-through of the mobile phase. The elution fractions were divided into two different consecutive chromatographic areas using every second fraction-tube (t1 – t21) as the first section and selecting every fourth tube (t22 – t193) represented the selection in the second section. Since the extrusion-mode was done with pressurized N2 -gas (pressure 7 bar), the volumes of the recovered fractions in the various tubes appeared to be inconsistent. Therefore, the calculation of elution volumes and KD -values was not possible, nevertheless, in order to monitor these compounds, the ESI-MS profiling considered every fourth tube from t197 to t333. Also, the ESI-MS profile was obtained by injecting 7 ␮L of the selected sp−CCC fractions and the results were projected (cf. Fig. 2).

Plant species of Nectandra are known for their biosynthetic production of miscellaneous natural product classes, such as aroma compounds (mono-, sesqui-, diterpenoids), phytosterols, alkaloids, phenyl-propanoids and lignoids [24]. In this preparative countercurrent chromatographic study, the fractionation and recovery of different C-C and C–O linked eugenoldimers were achieved. Previously, these specific phenyl-propanoid neolignans have been proved by their strong in-vitro activity against Neglected Tropical Diseases (NTDs), such as Leishmaniasis and Chagas disease [14–16]. In order to demonstrate an easy scale-up possibility in the laboratory work-flow, the fortified neolignan extract was separated by high performance countercurrent chromatography (HPCCC) and spiral-coil countercurrent chromatography (sp−CCC). Previous studies have separated C-C-linked phenyl-propanoid neolignans from the Chinese traditional medicinal plant, Magnolia offcinalis, by using a commercial high-velocity 4.6 L HPCCC system with a standard HEMWat system [25,26]. These studies have led to the recovery of lab-scale quantities of anti-tumor compounds (e.g. magnolol, honokiol). With respect to structural similarities, neolignans are not substituted by methoxy-groups, but are closely related to the target compounds in N. leucantha. The phase distribution properties of neolignans in the semilipophilic neolignan enriched fraction (NEF) sample was evaluated by different biphasic solvent systems (cf. 2.4.). Since the HEMWatsystem (SoSy7, n-hexane - EtOAc - MeOH - Water (7 : 3 : 7 : 3) has demonstrated to be the most suitable candidate, a method transfer from HPCCC to sp−CCC was achieved without further adjusting the solvent ratios. The elution sequences of metabolites appeared to be similar (cf. Figs. 1 and 2 and Table 1). However, in case of

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

the sp−CCC experiment, chromatographic resolution was much lower. 3.1. HPCCC and spiral−CCC separation and off-line ESI-MS/MS detection of neolignans from N. Leucantha The measured off-line ESI-MS experiments detected the main neolignan structures 1 – 5 by selected single ion traces and guided the approach in their respective CCC-fractions (cf. Fig. 1). In addition, the key chromatographic parameters, such as the partition ratio KD -, separation factor ␣, and resolution factor RS were determined specifically for the neolignan metabolites (cf. Tables 1 and 2). The methodology has proved to be a powerful tool for tracing major target compounds in fractions with high accuracy [9,27–29]. In our case, structurally known compounds were confirmed by their molecular weights and MS/MS fragmented pattern. In principle, the ESI-MS profiling approach could be implemented in all CCC/CPCseparations 3.1.1. Structural identification of neolignans 3.1.1.1. 1-[7-hydroxy-8-propenyl]-3-[3 -methoxy-1 -(8 -propenyl)phenoxy]-4-hydroxy-5-methoxy-benzene (1). The most polar metabolite was neolignan 1 with the quasimolecular ion signals [M-18+H]+ at m/z 325, [M + Na]+ m/z 365, and [2M + Na]+ at m/z 707. The MS/MS fragment ions of m/z 325 resulted in m/z 177, 161, and the neutral loss m/z 148, revealing the linkage of two different propenyl-benzene sub-units. 3.1.1.2. 1-[7-hydroxy-8-propenyl]-3-[3 -methoxy-1 -(8 -propenyl)phenoxy]-4,5-di-methoxy-benzene (3a). In the HPCCC run, the major neolignan 3a was clearly resolved from the counterpart positional isomer 3b, which was detected at much lower concentrated (cf. Fig. 1). Both structures displayed quasimolecular ions [2M + Na]+ at m/z 735, [M + Na]+ at m/z 379, and [M+H]+ at m/z 357. The MS/MS fragmented pattern of 3a was characterized by a strong in-source dehydroxylation from the hydroxyl-propenyl-side chain leading to [M-H2 O+H]+ at m/z 339. The observed MS/MS fragmented ions at m/z 298, 191, 176 in combination with the neutral loss m/z 148 corroborated the structure of 3a. 3.1.1.3. Dehydrodieugenol (4a). Also a minor but symmetrical CC-linked dimer dehydrodieugenol (4a) was soley detected in the HPCCC profile. The ion-intensity of [M+H]+ signal at m/z 327 was low and the ion-trap signal for the MS/MS observation of the compound was unsuccessful. This compound has previously been characterized from Nectandra polita [24,30]. 3.1.1.4. 1-(8-propenyl)-3-[1 -(8 -propenyl)-3 -methoxyphenoxy]-4hydroxy-5-methoxy-benzene (Dehydrodieugenol B) (4b). The high concentrated C–O-linked dimer dehydrodieugenol B 4b with the ion-signals [2M + Na]+ m/z 675, [M + Na]+ m/z 349, and [M+H]+ was structurally corroborated by the MS/MS-fragment ions at m/z 179, 163, 147 and the neutral loss m/z 148. The molecular weight isobar 4a/4b was clearly separated in the HPCCC run, but not completely resolved in the sp−CCC experiment. 3.1.1.5. 1-(8-propenyl)-3-[1 -(8 -propenyl)-3 -methoxyphenoxy]4,5-di-methoxy-benzene (dehydrodieugenol B methylether) (5). The most lipophilic neolignan 5 was detected by the ion-signals [2M + Na]+ m/z 703, [M + Na]+ m/z 363 and [M+H]+ at m/z 341. The indicative MS/MS-fragment ions at m/z 193, 163, 147, and the neutral loss m/z 148 indicated the cleavage of the lower unit. This compound crystallised directly from the CCC-eluate and was structurally confirmed by X-ray analysis [31].

9

3.2. Comparison of separation efficiency of semi-preparative HPCCC and spiral-coil CCC evaluated by selectivity factor ˛ and resolution factor RS In the recovery of neolignans from N. leucantha, the scale-up and transfer of semi-preparative HPCCC separation conditions (125 mL coil-volume, 500 mg injection, defined HEMWat system, head-totail operation mode) to a larger lab-scale spiral-coil prototype CCC system (5500 mL coil-volume) was proven. The scale-up factor was forty-four calculated by the ratio of their respective machine coil-column volumes. Nevertheless, due to limited sample availability, 16 g of the fortified NEF extract was injected, equivalent to a scale-up factor of thirty-two calculated from the used dry-extract amounts. Overall, the HPCCC was much more effective in achieving the chromatographic resolution, as observed in the neolignan separation, compared to the sp−CCC experiment (cf. Figs. 1 and 2). Metabolites with identical molecular weights (potentially positional isomers), such as the diastereomers (isobars) 3a/3b and 4a/4b were not resolved during the large scale sp−CCC experiment. In HPCCC, compound 1, containing the most polar compound, carrying two hydroxy-functions was resolved from the unknown neolignan 2. However, all attempts to separate 1 and 2 by traditional silica gel solid-phase chromatography proved unsuccessful. As a strategic approach for the recovery of 2, a sp−CCC separation could be run, as the first step, on larger scale for the recovery of the metabolites 1–3 in a batch. In the second step, HEMWat solvent system is optimized to re-run the recovered mixture for improved fractionation of 1 and 2 by the HPCCC system. Calculation of the separation factors ␣ for neolignans 1/2 resulted in similar values for both experiments (cf. Table 2), but the resolution factor RS 1.80 of the HPCCC indicated a very good performance with base-line separation. Overall, the HPCCC experiment had shown good separation factors ␣ for the metabolite pairs (1/4b, 1/5, 2/4b, 3a/4b, 3a/5, 4b/5, cf. Table 2) with values above 1.75, indicating base-line separations. Also, the resolution factor values RS supported the successful fractionation process.The high sample load injected to the sp−CCC system resulted in peak-broadening effects which was clearly determined by the molecular weight information and selected ESIMS ion traces. The calculated KD -widths of the eluting peaks are given in Tables 1 and 2. After recovery of neolignan 4, the system was flushed by a constant flow of N2 -gas to finalize the separation by extrusion, while saving organic stationary phase solvent in this final process step. However, the liquid flow to the various sections of the spiral column tubes was inconsistent over time, resulting in fraction tubes not being filled with identical solvent volumes. Therefore, a valid calculation of KD of the most lipophilic neolignan 5 and their respective separation and resolution factors in relation to the metabolites 1 - 4 was not successful.

3.3. Evaluation of experimental effciency Important key factors for the evaluation of the efficiency of the experiments are the ratio of used solvents versus recovered pure products, and the additional time required for the process steps. The total solvent consumptions of the HEMWat solvent system (7:3:7:3) in the two CCC experiments were calculated as the sum of used mobile and stationary phase volumes. In the HPCCC experiment, only the elution-mode was performed. Whereas the solvent consumed for this experiment was approx. 415 mL (mob. phase: 290 mL, stat. phase: 125 mL), the sp−CCC used approximately 16.2 L (mob. phase: 10.7 L, stat. phase: 5.5 L) without calculating the solvent amounts for the general rinsing steps.

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11 10

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

Calculations revealed that the HPCCC was slightly more economical, as 0.65 mg pure neolignan compounds (cf. Suppl. Fig. 2) was separated by using 1 mL of the biphasic solvent system. In comparison, the sp−CCC run consumed 1 mL of the biphasic solvent phases to recover 0.55 mg of the purified neolignans (1–5). Overall, both runs reached similar consumption levels taking into account that the sp−CCC extrusion was performed with nitrogen gas. The volumetric scale-up from HPCCC to sp−CCC would require a higher injection amount (22 g) of N. leucantha NEF extract. This would optimize the mass to volume ratio, but with a potential negative impact on the chromatographic resolution. Comparing both experiments, recovery rates for neolignans were quite similar (cf. Suppl. Fig. 2) as observed for compound 4 (HPCCC: 20%–100 mg, sp−CCC: 26% - 4.20 g), and for 5 (HPCCC: 17.8% - 89 mg, sp-CCC: 16% - 2.56 g). However, the experimental time requirements for the CCCruns had been extremely different. In the case of the HPCCC, the complete operation afforded approx. 2 h (stationary phase charging process with 10 mL/min - 15 min, fractionation by elutionextrusion - 100 min). The spiral-coil procedure, on the other hand, afforded 18 h of effective working time (spiral coil tube-system charging process by 50 mL/min flow rate - 2 h). After injection of the neolignan crude extract, the system was operated approx. 9.5 h in the elution-mode. The rotation was stopped and kept static overnight, and the extrusion with N2 -gas was started the next day (approx. 6 h). Nevertheless, a 16 g neolignan crude extract separation on the sp−CCC resulted in a much higher sample through-put ratio of 500 mg/ hr, instead of 135 mg/ hr for HPCCC (cf. Suppl. Fig. 2). With regard to the sustainability of the technique, in almost all cases, CCC provides the full option to recycle used solvents, either on lab-scale or in industrial operation units. In a study by Garrard et al., varieties of HEMWat systems were evaluated for complete re-distillation purposes. The unified upper and lower phases, which were recovered from the evaporated fractions, were analyzed by GC-FID/WLD, and missing solvent volumes supplemented before re-use in a new CCC-experiment [4].

4. Conclusions The mechanical designs, as well as the underlying separation modes of the applied devices (i.e. semi-preparative HPCCC and sp−CCC) for the recovery of bioactive neolignans from N. leucantha were extremely different. However, the separation results were comparable in many aspects. The method transfer was easily achieved from the smaller to the larger laboratory scale without any solvent system variation. Therefore, the scaleablity of the CCC-methods has been proved by the reproducible of the chromatograpic results (CCC-based KD -values of metabolite) and the recovery of the purified neolignan targets from N. leucantha in the range of 100 mg to 4 g (cf. Suppl. Fig. 2.). CCC has supplied sufficent amounts of pure neolignans, which is now available for future in-vitro, in-vivo and other clinical studies. One of the most important analytical technique applied in this study was the mass-spectrometry based profiling approach, which allows accurate tracing of target compounds in the selected CCC-fractions. This enabled the conversion of the chromatography into molecular weight based preparative recovery profiles, as well as the KD chromatographic scales for HPCCC and sp-CCC experiments. Furthermore, the low velocity of the sp−CCC system appeared to be a reliable technique for the recovery of natural products in a normal laboratory environment without implementation of additional safety precautions.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors are grateful for the financial support of ‘Coordination for the Improvement of Higher Education Personnel (CAPES)’ through a PDSE grant for SSG (99999.003062/2014–07). SSG is thankfull for grant 2018/09083-0 from São Paulo Research Foundation (FAPESP), and CCC 2018 travel bursary, financed by PhytoLab GmbH & Co. KG (Vestenbergsgreuth, Germany) for attending the 10th International Conference on Countercurrent Chromatography in Braunschweig, Germany. JHGL is thankful for financial support and fellowships provided by the National Council for Scientific and Technological Development (CNPq) and grant 2018/07885-1, São Paulo Research Foundation (FAPESP). This study is an activity within the Research Network Natural Products against Neglected Diseases (ResNetNPND: http://www. resnetnpnd.org/). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.chroma.2019. 460422. References [1] J.H. Clark, Green chemistry: challenges and opportunities, Green Chem. 1 (1999) 1–8. [2] J.H. Clark, R. Sheldon, C. Raston, M. Poliakoff, W. Leitner, 15 years of Green Chemistry, Green Chem. 16 (2014) 18–23. [3] I. Sutherland, S. Ignatova, P. Hewitson, L. Janaway, P. Wood, N. Edwards, G. Harris, H. Guzlek, D. Keay, K. Freebairn, D. Johns, N. Douillet, C. Thickitt, E. Vilminot, B. Mathews, Scalable technology for the extraction of pharmaceutics (STEP): the transition from academic know how to industrial reality, J. Chromatogr. A 1218 (2011) 6114–6121. [4] I. Garrard, L. Janaway, D. Fisher, Minimising solvent usage in high speed, high loading, and high resolution isocratic dynamic extraction, J. Liq. Chromatogr. Relat. Technol. 30 (2007) 151–163. [5] K. Faure, E. Bouju, P. Suchet, A. Berthod, Use of limonene in countercurrent chromatography: a green alkane substitute, Anal. Chem. 85 (2013) 4644–4650. [6] I. Sutherland, P. Hewitson, S. Ignatova, Scale-up of counter-current chromatography: Demonstration of predictable isocratic and quasi-continuous operating modes from the test tube to pilot/ process scale, J. Chromatogr. A 1142 (2007) 115–122. [7] A. Kotland, S. Chollet, C. Diard, J.-M. Autret, J. Meucci, J.-H. Renault, Luc Marchal, Industrial case study on alkaloids purification by pH-zone refining centrifugal partition chromatography, J. Chromatogr. A 1474 (2016) 59–70. [8] Y. Ito, Golden rules and pitfalls in selecting optimum conditions for high-speed counter-current chromatography, J. Chromatogr. A 1065 (2005) 145–168. [9] G. Jerz, Y.A. Elnakady, A. Braun, K. Jäckel, F. Sasse, A.A. Al Ghamdi, M.O.M. Omar, P. Winterhalter, Preparative mass-spectrometry profiling of bioactive metabolites in Saudi-Arabian propolis fractionated by high-speed countercurrent chromatography and off-line atmospheric pressure chemical ionization mass-spectrometry injection, J. Chromatogr. A 1347 (2014) 17–29. [10] T. Esatbeyoglu, V. Wray, P. Winterhalter, Isolation of dimeric, trimeric, tetrameric and pentameric procyanidins from unroasted cocoa beans (Theobroma cacao L.) using countercurrent chromatography, Food Chem. 179 (2015) 278–289. [11] M.I. Fernández-Marín, R.F. Guerrero, M.C. García-Parrilla, B. Puertas, T. Richard, M.A. Rodriguez-Werner, P. Winterhalter, J.-P. Monti, E. Cantos-Villar, Isorhapontigenin: a novel bioactive stilbene from wine grapes, Food Chem. 135 (2012) 1353–1359. [12] J.B. Althaus, G. Jerz, P. Winterhalter, M. Kaiser, R. Brun, T.J. Schmidt, Antiprotozoal activity of Buxus sempervirens and activity-guided isolation of O-tigloyl-cyclovirobuxeine-B as main constituent active against Plasmodium falciparum, Molecules 19 (5) (2014) 6184–6201. [13] Global Biodiversity Information Facility - GBIF Secretariat: GBIF Backbone Taxonomy. https://doi.org/10.15468/39omei. Accessed via https://www.gbif. org/species/7830782 on 09 March 2019.

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422

G Model CHROMA-460422; No. of Pages 11

ARTICLE IN PRESS S. dos Santos Grecco, E. Letsyo, A.G. Tempone et al. / J. Chromatogr. A xxx (xxxx) xxx

[14] T.A. Da Costa-Silva, S.S. Grecco, F.S. De Sousa, J.H.G. Lago, E.G.A. Martins, C.A. Terrazas, S. Varikuti, K.L. Owens, S.M. Beverley, A.R. Satoskar, A.G. Tempone, Immunomodulatory and antileishmanial activity of phenylpropanoid dimers isolated from Nectandra leucantha, J. Nat. Prod. 78 (2015) 653–657. [15] S.S. Grecco, T.A. Costa-Silva, G. Jerz, F.S. De Sousa, G.A. Alves Conserva, J.T. Mesquita, M.K. Galuppo, A.G. Tempone, B.J. Neves, C.H. Andrade, R.L.O. Cunha, M. Uemi, P. Sartorelli, J.H.G. Lago, Antitrypanosomal activity and evaluation of the mechanism of action of dehydrodieugenol isolated from Nectandra leucantha (Lauraceae) and its methylated derivative against Trypanosoma cruzi, Phytomedicine 24 (2016) 62–67. [16] S.S. Grecco, T.A. Costa-Silva, G. Jerz, F.S. De Sousa, V.S. Londero, M.K. Galuppo, M.L. Lima, B.J. Neves, C.H. Andrade, A.G. Tempone, J.H.G. Lago, Neolignans from leaves of Nectandra leucantha (Lauraceae) display in vitro antitrypanosomal activity via plasma membrane and mitochondrial damages, Chem. Biol. Interact. 277 (2017) 55–61. [17] WHO (World Health Organization), Neglected Tropical Diseases [Internet]. [Access in 2018 may 06]. Available at:, 2018 http://www.who.int/neglected diseases/diseases/en/. [18] Brasil. Ministério da Saúde – DECIT, Doenc¸as negligenciadas: estratégias do Ministério da Saúde, Rev. Saúde Pública 44 (2010) 200–202. [19] S. Burza, S.L. Croft, M. Boelaert, Leishmaniasis, Lancet 392 (2018) 951–970. [20] L.E. Echeverria, C.A. Morillo, American Trypanosomiasis (Chagas Disease), Infect. Dis. Clin. North Am. 33 (2019) 119–134. [21] C. Beyer, S. Singh, D. Sudarshi, Neglected tropical diseases, conflict, and the right to health, in: E. Choffnes, D. Relman (Eds.), The Causes and Impacts of Neglected Tropical and Zoonotic Diseases, National Academies Press, 2011, pp. 132–155. [22] E. Stahl, U. Kaltenbach, Dünnschicht-chromatographie: VI. Mitteilung. Spurenanalyse von Zuckergemischen auf Kieselgur G-Schichten, J. Chromatogr. 5 (1961) 351–355. [23] N. Köhler, E. Chou, Y. Ito, P. Winterhalter, Development of a new preparative spiral-coil low-speed rotary countercurrent chromatographic (spiral-coil

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

11

LSRCCC) method, J. Liq. Chromatogr. Relat. Technol. 27 (2004) 2547–2560. S.S. Grecco, H. Lorenzi, A.G. Tempone, J.H.G. Lago, Update: biological and chemical aspects of Nectandra genus (Lauraceae), Tetrahedron Asymmetry 27 (2016) 793–810. L. Chen, Q. Zhang, G. Yang, L. Fan, J. Tang, I. Garrard, S. Ignatova, D. Fisher, I.A. Sutherland, Rapid purification and scale-up of honokiol and magnolol using high-capacity high-speed counter-current chromatography, J. Chromatogr. A 1142 (2007) 115–122. A. Peng, P. Hewitson, H. Ye, L. Zu, I. Garrard, I. Sutherland, L. Chen, S. Ignatova, Sample injection strategy to increase throughput in counter-current chromatography: case study of honokiol purification, J. Chromatogr. A 1476 (2016) 19–24. M.N. Vieira, F. das Neves Costa, G. Guimarães Leitão, I. Garrard, P. Hewitson, S. Ignatova, P. Winterhalter, G. Jerz, Schinus terebinthifolius scale-up countercurrent chromatography (Part I): high performance countercurrent chromatography fractionation of triterpene acids with off-line detection using atmospheric pressure chemical ionization mass spectrometry, J. Chromatogr. A 1389 (2015) 39–48. F. das Neves, M.N. Costa, I. Vieira, P. Garrard, G. Hewitson, G. Jerz, Guimarães Leitão, S. Ignatova, Schinus terebinthifolius countercurrent chromatography (Part II): intra-apparatus scale-up and inter-apparatus method transfer, J. Chromatogr. A 1466 (2016) 76–83. J.B. Althaus, G. Jerz, P. Winterhalter, M. Kaiser, R. Brun, T.J. Schmidt, The alkaloid fraction from Buxus sempervirens leaves shows strong in-vitro activity against Plasmodium falciparum, Planta Med. 79 (2013) PA2. M. Suarez, J. Bonilla, A.M.P. De Diaz, H. Achenbach, Dehydroeugenols from Nectandra polita, Phytochemistry 22 (1983) 609–610. S.S. Grecco, G. Jerz, J.H.G. Lago, P.G. Jones, Crystal structure of Dehydrodieugenol B methylether, a neolignan from Nectandra leucantha Nees and Mart (Lauraceae), Acta Cryst. E74 (2018) 518–521.

Please cite this article in press as: S. dos Santos Grecco, E. Letsyo, A.G. Tempone, et al., Electrospray mass-spectrometry guided target isolation of neolignans from Nectandra leucantha (Lauraceae) by high performance- and spiral-coil countercurrent chromatography, J. Chromatogr. A (xxxx), https://doi.org/10.1016/j.chroma.2019.460422