Optimisation of preparative HPLC separation of four isomeric kaempferol diglycosides from Prunus spinosa L. by application of the response surface methodology
Graphical abstract
Introduction
Prunus spinosa L. (blackthorn or sloe) is a large thorny shrub or small tree widespread in the temperate regions of the northern hemisphere. It has been known since ancient times as medicinal and dietary plant. Its leaves has been used traditionally for various purposes, e.g. for the treatment of urinary tract inflammations and in blood cleansing therapies (Berger, 1950). The plant material is a rich source of flavonoid glycosides (up to 4.27% dry weight), comparable to other widely applied flavonoid herbal remedies such as birch leaves (Betulae folium) and elder flower (Sambuci flos) (Olszewska et al., 2001). Nine different flavonoids have been isolated to date from the leaves of P. spinosa including kaempferol, quercetin, their 3-O-α-l-arabinofuranosides, kaempferol 3- and 7-O-α-l-rhamnopyranosides, kaempferol 3,7-di-O-α-l-rhamnopyranoside, kaempferol 3-O-α-l-arabinofuranoside-7-O-α-l-rhamnopyranoside, quercetin 3-O-(2′′-O-β-d-glucopyranoside)-α-L-arabinofuranoside and kaempferol 3-O-(2′′-E-p-coumaroyl)-α-l-arabinofuranoside-7-O-α-l-rhamnopyranoside (Olszewska and Wolbiś, 2002a). Some of these compounds, especially mono- and dipentosides, which are characteristic of P. spinosa, and relatively rare in nature, could probably be applied as analytical markers in standardisation of plant materials derived from the taxon (Olszewska and Wolbiś, 2001, Olszewska and Wolbiś, 2002a, Olszewska and Wolbiś, 2002b). However, the typical flavonoid pattern of the leaves still requires full characterisation, as no comprehensive studies on their phenolic profile has been published so far.
The flavonoid fraction of blackthorn leaves was proven in in vivo studies to significantly reduce capillary permeability, exhibit anti-inflammatory effects in the animal skin and internal organs, normalize the blood cholesterol and a cholesterol/phospholipid ratio in atherogenic rabbits, display spasmolytic effects on isolated uterine and intestinal segments from different animals, increase the amplitude of heart contractions in perfusion of isolated frog hearts, and demonstrate significant diuretic and natriuretic activity in rats (Lisevitskaya et al., 1970, Makarov, 1972, Makarov, 1970, Makarov and Khadzhai, 1969). These effects were especially pronounced for kaempferol 3-O-α-l-arabinofuranoside-7-O-α-l-rhamnopyranoside and kaempferol 3,7-di-O-α-l-rhamnopyranoside – activity of which was higher than that of rutin – suggesting that the 3,7-disubstituted kaempferol derivatives might be primary determinants of the activity of P. spinosa leaves (Makarov and Khadzhai, 1969). However, despite these promising results, published in the early 70’s of the 20th century, further in vivo studies on the target flavonoids were hindered by their commercial unavailability and lack of effective methods for isolation from natural sources.
According to our previous studies on P. spinosa, the blackthorn leaves appear rich in kaempferol dipentosides, the fraction of which is, on one hand, easy to isolate from the plant due to unique solubility, but extremely difficult to separate into components (due to their considerable structural similarities) on the other. The target fraction was composed of at least four analytes, two of which remain structurally unidentified to date. Moreover, two major constituents of the fraction (kaempferol 3-O-α-l-arabinofuranoside-7-O-α-l-rhamnopyranoside and kaempferol 3,7-di-O-α-l-rhamnopyranoside) were isolated from the leaves but with unsatisfactory yield resulting from low resolution of the applied conventional open column chromatography techniques (Olszewska and Wolbiś, 2002a).
Therefore, the aim of the present study was to develop and optimise the first preparative HPLC procedure for fast and efficient isolation of four isomeric kaempferol dipentosides from the leaves of blackthorn. Since the optimisation of HPLC separations by a trial-and-error method can be inefficient and time-consuming, and does not guarantee the optimal results, it was facilitated by the application of the response surface methodology (RSM) – a systematic experimental design and statistically-assisted construction of a response function describing relationships between the process variables and performance parameters of the chromatographic separation. The optimal conditions, based on isocratic elution with ternary mobile phase, established in an analytical scale, were directly transferred to a preparative one, and successful isolation was then achieved with a run time of 12 min. The purity of isolates and their structures were thoroughly analysed by the spectroscopic (ESI–MS, 1H NMR, 13C NMR, COSY, HMQC, HMBC) studies and hydrolytic experiments. Finally, the qualitative composition of the raw leaf extracts and the target dipentoside fraction was fully characterised by an UHPLC-PDA-ESI–MS3 assay with comparison to the currently and previously isolated standards.
Section snippets
Optimisation of the separation procedure
The target kaempferol diglycosidic fraction was isolated from the leaves of P. spinosa according to the simplified procedure described previously (Olszewska and Wolbiś, 2002a), i.e. from n-butanolic fraction of the defatted 70% (v/v) aqueous methanolic extract, separated from accompanying polymeric proanthocyanidins and quercetin diglycosides by gel permeation chromatography on Sephadex LH-20. Thus, 2.1 g of the dipentoside fraction was obtained from 100 g of the dried plant material (2.1% yield,
Conclusion
The present paper describes the application of the central composite design in combination with the response surface methodology for optimisation of preparative HPLC separation of four isomeric kaempferol diglycosides. The applied statistical approach, used to the best of our knowledge for the first time for optimisation of preparative HPLC separation, proved highly effective in finding the most suitable analysis conditions for closely structurally related flavonoid isomers. The crucial factor
General experimental procedures
HPLC-PDA experiments (both analytical and preparative) were performed on preparative Waters 2545 binary system (Waters, Milton, MA, USA) equipped with Waters 2767 autosampler and Waters 2998 PDA detector.
UHPLC-PDA-ESI–MSn analyses were performed on UHPLC-3000 RS system (Dionex, Dereieich, Germany) with a diode array detector with multiple wavelength (Thermo Fisher Scientific Inc., Waltham, MA, USA), and an ultrahigh resolution hybrid quadrupole/time-of-flight mass spectrometer (UHR-Q-TOF–MS,
Acknowledgements
The authors would like express their gratitude to the Department of Pharmacognosy and Molecular Basis of Phytotherapy, Faculty of Pharmacy, Medical University of Warsaw, for enabling access to the UHPLC–MS equipment. This work is supported by Medical University of Lodz grant Nos 503/2-022-01/503-31-001, 502-03/3-022-01/502-34-082, and 503/2-014-01/503-31-001.
References (45)
- et al.
Experimental design and optimization
Chemom. Intell. Lab. Syst.
(1998) - et al.
Four kaempferol glycosides from leaves of Cinnamomum sieboldii
Phytochemistry
(1983) - et al.
Phenolic constituents of the inflorescences of Sorbus torminalis (L.) Crantz
Phytochem. Lett.
(2011) Separation of quercetin, sexangularetin, kaempferol and isorhamnetin for simultaneous HPLC determination of flavonoid aglycones in inflorescences, leaves and fruits of three Sorbus species
J. Pharm. Biomed. Anal.
(2008)- et al.
Resolution of the enantiomers of aldoses by liquid chromatography of diastereoisomeric 1-(N-acetyl-(-methylbenzyalmino)-1-deoxyalditol acetates
Carbohydr. Res.
(1982) - et al.
Determination of flavonoids by high-performance liquid chromatography and capillary electrophoresis
J. Chromatogr. A
(2004) - et al.
Validation and application by HPLC for simultaneous determination of vitexin-2′′-O-glucoside, vitexin-2′′-O-rhamnoside, rutin, vitexin, and hyperoside
J. Pharm. Anal.
(2011) - et al.
PTP1B, (-glucosidase, and DPP-IV inhibitory effects for chromene derivatives from the leaves of Smilax china L
Chem. Biol. Interact.
(2016) - et al.
Cytotoxic flavonoid glycosides from Rapistrum rugosum L
Iran. J. Pharm. Res.
(2012) - Aubin, A., Cleary, R., 2009. Analytical HPLC to Preparative HPLC: Scale-Up Techniques using a Natural Product Extract...