WO2025218875A1

WO2025218875A1 - Method for extracting phyllodulcin including nades

Info

Publication number: WO2025218875A1
Application number: PCT/EP2024/060144
Authority: WO
Inventors: Mats KIENE; Phuong Anh Nguyen; Jakob Peter Ley; Silke HILLEBRAND; Margit LIEBIG; Johannes WELLMANN; Melanie Stürtz
Original assignee: Symrise AG
Current assignee: Symrise AG
Priority date: 2024-04-15
Filing date: 2024-04-15
Publication date: 2025-10-23
Anticipated expiration: 2026-10-15

Abstract

The present invention relates to a method for producing a plant extract comprising phyllodulcin, wherein particular natural deep eutectic solvents are used, and to the use of a natural deep eutectic solvent in such a method.

Description

Method for extracting phyllodulcin including NADES

Consumers generally have a strong preference for foodstuffs or indulgence foods, which have a large amount of high caloric sugar, in particular sucrose (saccharose), glucose, fructose or mixtures thereof, due to the pleasant sweetness and sweetness profile associated therewith. On the other hand, it is generally known that a large content of readily metabolizable carbohydrates causes a steep rise in blood sugar levels, leads to the formation of fat deposits and ultimately can result in health problems such as overweight, obesity, insulin resistance, age-onset diabetes and complications thereof. Another particular aggravating factor is that many of the above-mentioned carbohydrates can also have an adverse effect on dental health, as they are decomposed by specific types of bacteria in the oral cavity into lactic acid, for example, and can attack the enamel of milk teeth or adult teeth (caries). Therefore, it has long been an objective to reduce the high caloric sugar content of food or beverage products and replace it partly or entirely by other substances that impart a sweet taste or which can positively affect the sweet taste in a low concentration without exhibiting sweet taste itself at these low concentration (taste modulators). The use of phyllodulcin as taste modulator for sugar-reduced products, flavoring mixtures, and methods for producing such products was described in EP 2 298 084 B1.

Phyllodulcin is a natural compound occurring in subspecies of the plant Hydrangea macro- phylla. Phyllodulcin cannot be economically produced by chemical synthesis or biotechnological approaches.

The leaves of Hydrangea macrophylla are mainly used for preparing tea, particularly Ama- cha, a sweet tasting Japanese tea, which contains tannins and dihydroisocoumarins including phyllodulcin. The leaves of Hydrangea macrophylla are typically used in Japan and Korea for ceremonial purposes (Buddhas birthday).

Several methods have been described for extracting phyllodulcin from the leaves of Hydrangea macrophylla. However, when used in products for nutrition or pleasure or in pharmaceutical products, several regulatory conditions need to be met. In this regard, the obtained extract must not contain detectable amounts of certain substances, such as chloroform, which are frequently used as solvents for extraction.

Furthermore, it is necessary that the sensory characteristics of phyllodulcin are not negatively affected by the extraction process or the compounds additionally present in the obtained extract. In this regard, it has been found that in an alkaline milieu phyllodulcin is converted to an open chain form and finally to a certain stilbene carboxylic acid, which has no considerable taste characteristics. Thus, when phyllodulcin is converted to degradation products, its sweetening characteristics are lost.

Degradation of phyllodulcin

JP51002480 describes a method, in which a tar-like precipitate is obtained by salting out phyllodulcin from an alcoholic extract. The precipitate was dissolved in methanol and an extraction with chloroform was performed. The product was obtained from the organic phase and finally recrystallized in methanol. However, due to regulatory reasons, chloroform must not be used for obtaining an extract, which is used in any consumed products. Furthermore, JP51002480 does not disclose the sensory purity of the obtained product.

Jung et al., Phytochem. Anal. 2016, 27, 140-147 discloses the extraction and isolation of phyllodulcin from the leaves of Hydrangea macrophylla for analytical purposes. The method is performed by applying steps comprising ultrasound and high pressure extraction combined with preparative HPLC. However, for the industrial purposes of obtaining extracts with a high amount of phyllodulcin, this method is not suitable, as the above steps are not relevant due to economical reasons.

KR101662498 discloses a method, in which the leaves are wet fermented and subsequently extracted with methanol or ethanol. The obtained extract is purified by ion exchange and a reverse-phase solid phase and elution with acetonitrile. Finally, the product is obtained by chromatographic separation by HPLC. However, the method applies 5 steps of separation and is thus not suitable for industrial purposes. Furthermore, expensive materials such as the reverse-phase solid phase are used. Moreover, acetonitrile must not be used for obtaining an extract, which is used in any consumed products, for regulatory reasons.

KR 20210074526A discloses a method for extracting phyllodulcin, however, the possibility of reducing off-tastes, which are frequently observed, is not addressed.

Kim Min-Ji et al.: "Relative sweetness and sweetness quality of phyllodulcin [(3R)-8-Hy- droxy-3-(3-hydroxy-4-methoxyphenyl)-3,4-dihydro-1 H-isochromen-1 -one]", Food Science And Biotechnology, The Korea Soc. Of Food Science And Technology, Heidelberg, vol. 25, no. 4, 31 August 2016 discloses plant extracts comprising phyllodulcin. However, the plant extracts are described as having strong bitter notes, alcohol notes and as being rather astringent.

JPS5724745 describes the extraction of phyllodulcin with mixtures of water and alcohols. Subsequently, the alkali salt is obtained and precipitated and the product is washed with chloroform. As described above, chloroform must not be used for regulatory reasons. Furthermore, when obtaining the alkali salt, phyllodulcin is converted to the respective stilbene carboxylic acid, thus losing the sweetening characteristics of phyllodulcin. EP 2298084 describes the production of extracts containing phyllodulcin, wherein the extracts contain 14 % phyllodulcin when the extraction is performed with ethyl acetate or contain 5.3 % or, respectively, 3.2 % phyllodulcin when a hydroalcoholic extraction is performed. However, the amounts or phyllodulcin are rather low. Furthermore, it is described that the extracts have a herbaceous, fishy, liquorice-like, tea-like, woody and slightly bitter olfactory and gustatory notes. Consequently, the obtained extracts and the method of their production are not suitable for providing a flavouring mixture.

Therefore, a great need exists to provide methods for producing an extract containing phyllodulcin, wherein the extract can be used for products for nutrition or pleasure or for pharmaceutical products and wherein the sensory characteristics of phyllodulcin can be used.

WO 2023/079021 A1 describes such a method providing the required advantages, wherein the method includes a step of extraction with a solvent. WO 2023/079021 A1 mentions several exemplary solvents for this extraction step, for example deep eutectic solvents.

Natural deep eutectic solvents are mixtures of at least two components where the mixture shows a melting point much lower than either of the individual components. Preferably natural components such as sugars, alcohols, amines, organic acids or amino acids are used as such solvents.

Natural deep eutectic solvents are mixtures of typical substances in laboratories and are easily available at lower costs. Furthermore, natural deep eutectic solvents are biologically degradable and regenerative.

Solvents are typically used in high amounts in such methods and should thus be available at possibly low costs. At the same time, solvents must be non-toxic and fulfil further, regulatory requirements for a product to be used in a food product. Moreover, the selection of a particular solvent influences the amount of extracted phyllodulcin in an extraction step. Furthermore, the solvent should be a natural and regenerative solvent, wherein the amount of extracted phyllodulcin should not be reduced compared to previously used solvents.

The primary object of the present invention was thus to provide a method for producing an extract containing a possibly high amount of phyllodulcin with a non-toxic and non-expen- sive solvent. The primary object of the present invention is solved by a method for producing a plant extract comprising phyllodulcin, wherein the method comprises or consists of the following steps: i) drying the plant material to be extracted, preferably wherein the plant material predominantly consists of plant leafs, preferably wherein the plant material is of a plant of the species Hydrangea macrophylla preferably of the subspecies serrata, more preferably selected from the group consisting of the varieties Oamacha, Amacha and Amagi-Amacha, ii) moistening the dried plant material of step i) with water and wet fermenting the plant material at a temperature in the range of from 10 to 50 °C, preferably in the range of from 30 to 45 °C, for 0.5 to 72 hours, preferably for 0.5 to 16 hours, particularly preferably for 1 to 5 hours, iii) subjecting the wet fermented plant material of step ii) to an extraction with a natural deep eutectic solvent, at a temperature of from 0° C to the boiling point of the respective solvent, wherein the natural deep eutectic solvent comprises at least 2 components selected from the group consisting of lactic acid, choline chloride, Proline, preferably L-Pro- line, Betaine, 1 ,2-propane diol, glucose and malic acid, iv) collecting the supernatant (S1) of the extraction of step iii) and optionally: repeating step iii), collecting the supernatant (S2) and combining supernatants S1 and S2, and v) filtering the supernatant or combined supernatants obtained after step iv) and collecting the filtrate.

Several natural deep eutectic solvents have been tested and the amount of phyllodulcin in the extracts was compared between the used solvents. It was surprisingly found that the above natural deep eutectic solvents provided a particularly high amount of phyllodulcin in the extracts. Furthermore, it was surprisingly found that the amount of phyllodulcin in the extracts was comparable to the amount obtained with other solvents such as methanol or ethanol. The term “drying” as used herein, preferably as used in step i) of the method according to the invention describes a process, in which the water content of the material or mixture to a residual water content of less than 5 wt.-%, preferably less than 2.5 wt.-%, particularly preferably less than 1 wt.-%, based on the total weight of the material or mixture. Preferably, the step of drying in step i) of the method according to the invention is performed by a method selected from the group consisting of heat drying, sun drying, hot air drying, combined air and heat pump drying, vacuum oven drying, freeze drying.

Preferably, a water content, as described herein, is determined by dry-loss-method or Karl- Fischer titration.

It is preferred that the dry matter content of the material or mixture, i.e. the portion, which is not water, is at least 95 wt.-% preferably at least 97.5 wt.-%, particularly preferably at least 99 wt.-%, based on the total weight of the material or mixture.

The term “wherein the plant material predominantly consists of’ as used herein refers to an amount of at least 90 wt.-%, preferably at least 95 wt.-%, particularly preferably at least 98 wt.-%, especially preferably at least 99 wt.-%, based on the total weight of the plant material.

The plant varieties Amacha and Amagi-Amacha, as mentioned herein as plant varieties of Hydrangea macrophylla, are also known as varieties thunbergii or, respectively, amagiana.

It is preferred in the method according to the invention that the method includes a step ii.2), which is performed between step ii) and step iii) and which includes removing fermentation water, wherein the removal is achieved by e.g. applying mechanical or air pressure filtration, decantation, centrifugation, combined filtration and centrifugation, hot air drying, combined air and heat pump drying, vacuum oven drying, orfreeze drying. Preferably, removing fermentation water describes removing at least 50 wt.-%, preferably at least 75 wt.-%, particularly preferably at least 90 wt.-%, further preferably at least 95 wt.-% of the fermentation water, based on the mixture of the plant material and water as obtained in step ii) of the method according to the invention.

The term “boiling point” as used herein describes the boiling point at standard conditions, i.e. a pressure of 1013 mbar. In case it is referred to the boiling point of a mixture of substances, the skilled person knows that the boiling points of the respective single substances are known in literature and a boiling point can be calculated accordingly for the mixture. In this case, the term “boiling point” describes the calculated boiling point.

It is preferred in step iii) that the natural deep eutectic solvent comprises lactic acid.

It is preferred in step iii) that the natural deep eutectic solvent comprises choline chloride.

It is preferred in step iii) that the natural deep eutectic solvent comprises Proline.

It is preferred in step iii) that the natural deep eutectic solvent comprises L-Proline.

It is preferred in step iii) that the natural deep eutectic solvent comprises Betaine.

It is preferred in step iii) that the natural deep eutectic solvent comprises 1 ,2-propane diol.

It is preferred in step iii) that the natural deep eutectic solvent comprises glucose.

It is preferred in step iii) that the natural deep eutectic solvent comprises malic acid.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises lactic acid and 1 ,2-propane diol.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises lactic acid and glucose.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises lactic acid and choline chloride.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises lactic acid and Proline.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises lactic acid and L-Proline.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises malic acid and Proline. It is preferred in step iii) that the natural deep eutectic solvent is or comprises la malic acid and L-Proline.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises malic acid and Betaine.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises L-Proline and 1 ,2-propane diol.

It is preferred in step iii) that the natural deep eutectic solvent is or comprises choline chloride and 1 ,2-propane diol.

It is preferred in step iii) that the natural deep eutectic solvent further comprises water. Preferably, the water content in the natural deep eutectic solvent is in a range of from 2.5 to 50 wt.-%, preferably in a range of from 5 to 40 wt.-%, particularly preferably in a range of from 7.5 to 35 wt.-%, based on the total weight of the mixture of water and the at least 2 components selected from the group consisting of lactic acid, choline chloride, Proline, preferably L-Proline, Betaine, 1 ,2-propane diol, glucose and malic acid.

The term “total weight of the mixture of water and the at least 2 components selected from the group consisting of [...]” is to be understood such that it refers to the weight of the mixture of water and the described components, as far as present. For example, in case the natural deep eutectic solvent is a mixture of water, lactic acid and 1 ,2-propane diol, the term refers to the weight of the water, lactic acid and 1 ,2-propane diol.

It is preferred in step iii) that the molar ratio of the or of two components of the natural deep eutectic solvent is in the range of from 1 :1 to 1 :10, preferably in the range of from 1 :1 to 1 :7.5, particularly preferably in the range of from 1 :1 to 1 :5.

It is preferred that step iii) of the method according to the invention is performed at normal pressure, i.e. a pressure of 1013 mbar. Alternatively, step iii) is performed at reduced pressure. Alternatively, step iii) is performed at elevated pressure.

It is further preferred that the extraction in step iii) of the method according to the invention is performed by a method comprising or consisting of a step selected from the group consisting of Soxhlet, counter-current, percolation and maceration. As described above, step iii) of the method according to the present invention may (optionally) be repeated. Preferably, step iii) is repeated twice or three times. Preferably, in case step iii) is repeated, the same type of solvent is used each time step iii) is performed, i.e. in case step iii) is performed a first time with lactic acid and 1 ,2-propane diol as solvent, it is preferred that in each repetition, lactic acid and 1 ,2-propane diol is used as a solvent. Additionally or alternatively, the time of extraction in step iii), in a method, in which step iii) is repeated, may be shorter or longer than in a method, in which step iii) is not repeated. For example, one method according to the invention may comprise an extraction for 2 hours, wherein step iii) is repeated twice (i.e. step iii) is performed three times), whereas another method according to the invention may comprise an extraction for 4 hours, wherein step iii) is not repeated (i.e. step iii) is performed once).

It is preferred in the method according to the invention that the extract is purified by solidphase adsorption after the extraction in step iii) and/or repeated step iii).

Additionally in the above case, it is also preferred that the adsorbent used in the solidphase adsorption is selected from the group consisting of polystyrene, aliphatic methyl acrylate and mixtures thereof, preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and wherein the polystyrene is cross-linked, preferably with a divinyl benzene, particularly preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and wherein the polystyrene is cross-linked, preferably with a divinyl benzene, and wherein the polystyrene is macroporous and/or wherein the aliphatic methyl acrylate is microporous.

The term “macroporous” as used herein describes a material with pores having an average pore size of more than 2 nm, preferably a pore size of from 5 to 110 nm. The average pore size is the measured or calculated average over the size of all pores in the material. Methods for determining the pore size are well-known and include e.g. the Hg intrusion method.

The term “microporous” as used herein describes a material with pores having an average pore size of equal or less than 2 nm. The average pore size is the measured or calculated average over the size of all pores in the material. Methods for determining the pore size are well-known and include e.g. the Hg intrusion method. Preferably, the term “a divinyl benzene” describes a divinyl benzene selected from the group consisting of 1 ,2-Divinylbenzol, 1 ,3-Divinylbenzol, 1 ,4-Divinylbenzol and mixtures thereof.

It is further preferred in the method according to the invention that the extraction in step iii) is performed over a time in the range of from 0.5 to 10 hours, preferably 0.75 to 5 hours, particularly preferably from 1 to 3 hours.

It is further preferred that the method according to the invention comprises the step vi) diluting the filtrate collected in step v) with a diluent, preferably wherein the diluent is selected from the group consisting of water; protic solvents, such as alcohols, preferably methanol, ethanol or isopropanol, and carbonic acids, preferably formic acid, acetic acid, propanoic acid; and mixtures thereof

It is further preferred that the method further comprises the step vii) purifying the filtrate collected in step v) or the diluted filtrate in step vi), preferably by a method selected from the group consisting of precipitation, centrifugation, crystallization, solid-phase adsorption, filtration, reverse osmosis and combinations thereof.

Thus, the method according to the invention may comprise step vi), as well as step vii). However, the method may also comprise step vi) but no step vii). Likewise, the method may also comprise step vii) but no step vi).

It has been surprisingly found that by the purification it was possible to significantly reduce the amount of hydrangenol in the obtained extract. Furthermore, particularly high amounts of phyllodulcin were obtained.

Furthermore in the method according to the invention, it is preferred that the purification in step vii) comprises solid-phase adsorption and wherein the adsorbent used in the solidphase adsorption is selected from the group consisting of polystyrene, aliphatic methyl acrylate and mixtures thereof, preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, particularly preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, and the polystyrene is macroporous.

The term “microporous” as used herein describes a material with pores having an average pore size of equal or less than 2 nm. The average pore size is the measured or calculated average over the size of all pores in the material. Methods for determining the pore size are well-known and include e.g. the Hg intrusion method.

Preferably, the term “a divinyl benzene” describes a divinyl benzene selected from the group consisting of 1 ,2-Divinylbenzol, 1 ,3-Divinylbenzol, 1 ,4-Divinylbenzol and mixtures thereof.

Preferably, when referring to a filtration or filtering step, a filter with a mesh size of at least I at most 5 pm is used. Additionally or alternatively, the term “filtration” or “filtering step” refers to or includes nanofiltration or ultrafiltration

Preferably in the method according to the invention, the plant material comprises at least 2.5 wt.-%, preferably at least 3 wt.-%, particularly preferably at least 4 wt.-%, especially preferably at least 5 wt.-%, phyllodulcin or phyllodulcin equivalents, based on the total weight of the plant material.

Generally, the term “phyllodulcin” describes a chemical compound, which is classified as aglycon. However, in nature, phyllodulcin is often present in form of a glycoside, wherein several glycosides are known in the prior art and to a skilled person. The term “phyllodulcin equivalents” as used herein describes the aglycon as well as the phyllodulcin glycosides. Thus, if an amount of phyllodulcin equivalents is to be determined, both the aglycon as well as the phyllodulcin glycosides are to be considered, each as far as present, but preferably calculated on the base of virtually free phyllodulcin by deglycosidation. Furthermore, it is preferred in a method according to the present invention that, after the fermentation according to step ii), phyllodulcin is substantially present in its aglycon form. Typical glycosides of phyllodulcin , included in the term “phyllodulcin glycosides” are for example but not limited to:

Preferably, the term “phyllodulcin” describes a mixture of phyllodulcin enantiomers (and their glycosides), wherein the amount of the enantiomer (3R)-phyllodulcin is higher than each of the amounts of the other enantiomers, particularly preferably higher than the combined amounts of the other enantiomers.

Also preferred in the method according to the invention is that the plant material comprises 2 wt.-% or less, preferably 1 .5 wt.-% or less, particularly preferably 1 .0 wt.-% or less, espe- cially preferably 0.5 wt.-% or less, hydrangenol or hydrangenol equivalents, based on the total weight of the plant material.

It is further preferred in the method according to the invention that the weight ratio of phyllodulcin and hydrangenol in the plant material is in a range of from 1 :1 to 1 :50, preferably in a range of from 1 :2 to 1 :40, preferably in a range of from 1 :3 to 1 :30, preferably in a range of from 1 :4 to 1 :25, preferably in a range of from 1 :5 to 1 :20. Generally, the term “hydrangenol” describes a chemical compound, which is classified as aglycon. However, in nature, hydrangenol is often present in form of a glycoside, wherein several glycosides are known in the prior art and to a skilled person. In addition, hydrangenol can also be present as open chain version, so called hydrangeic acid or its glycosides. The term “hydrangenol equivalents” as used herein describes the aglycon as well as the hydrangenol glycosides and preferably also hydrangeic acid as well as its glycosides. Thus, if an amount of hydrangenol equivalents is to be determined, both the aglycon as well as the hydrangenol glycosides are to be considered, preferably also the aglycon as well as the glycosides of hydrangeic acid, each as far as present, but calculated on the base of virtually free hydrangenol by deglycosidation. Furthermore, it is preferred in a method according to the present invention that, after the fermentation according to step ii), hydrangenol and/or hydrangeic acid is/are substantially present in its/their (respective) aglycon form.

Typical glycosides of hydrangenol, included in the term “hydrangenol glycosides” are for example but not limited to:

The present invention further relates to the use of a natural deep eutectic solvent in a method according to the invention, wherein the natural deep eutectic solvent comprises at least 2 components selected from the group consisting of lactic acid, choline chloride, Proline, preferably L-Proline, Betaine, 1 ,2-propane diol, glucose and malic acid.

What was said above with regard to the method according to the invention and with regard to the natural deep eutectic solvents applies accordingly. Preferably, a plant extract obtained or obtainable by a method according to the invention comprises at least 50 wt.-%, preferably at least 60 wt.-% particularly preferably at least 65 wt.-%, especially preferably at least 70 wt.-% phyllodulcin, based on the total weight of the dry matter content of the extract.

Additionally or alternatively, a plant extract obtained or obtainable by a method according to the invention comprises less than 20 wt.-%, preferably less than 15 wt.-%, particularly preferably less than 10 wt.-%, especially preferably less than 5 wt.-% hydrangenol, based on the total weight of the dry matter content of the extract. The term “plant extract comprising at least 50 wt.-% phyllodulcin” as used herein is understood as excluding phyllodulcin as a pure substance, i.e. 100 wt.-% phyllodulcin. The term “plant extract” clarifies in this regard that phyllodulcin is obtained, i.e. extracted, from a plant source and is not the result of a chemical synthesis producing phyllodulcin from a precursor. The same applies accordingly for the further specified wt.-% indications as well as for the term “plant extract comprising less than 20 wt.-% hydrangenol” and the further specified wt.-% indications in this regard.

The term “dry matter content of the extract” describes the portion of the extract, which is not water, i.e. excludes the water content. Preferably, the dry matter content of the extract is at least 95 wt.-% preferably at least 97.5 wt.-%, particularly preferably at least 99 wt.-%, based on the total weight of the extract.

Preferably, the extract comprises at least 0.05 wt.-%, preferably at least 0.1 wt.-%, particularly preferably at least 0.5 wt.-%, more preferably at least 1 wt.-%, further preferably at least 5 wt.-%, most preferably at least 10 wt.-% of plant based components, based on the total weight of the extract.

Additionally or alternatively it is preferred that the extract comprises at most 30 wt.-%, preferably at most 25 wt.-%, particularly preferably at most 20 wt.-%, more preferably at most 10 wt.-%, further preferably at most 5 wt.-%, most preferably at most 1 wt.-% of plant based components, based on the total weight of the extract.

The term “plant based components” describes components, which are present in the plant material, from which the extract was obtained, but excludes phyllodulcin and hydrangenol (and their glycosides, as described above). Preferably, the term “plant based components” describes components selected from the group consisting of chlorophyll, rubisco, other proteins, triacylglyceride, waxes, cellulose, sugars, amino acids, flavonoids, hydroxybenzoic acids, and further aroma compounds.

Preferably, the extract comprises phyllodulcin in an amount of from 75 to 95 wt.-%, preferably 77.5 to 90 wt.-%, based on the total weight of the dry matter content of the extract.

Preferably, the extract comprises hydrangenol in an amount of from 15 to 25 wt.-%, preferably less than 20 wt.-%, preferably less than 15 wt.-%, particularly preferably less than 10 wt.-%, especially preferably less than 5 wt.-% hydrangenol, each based on the total weight of the dry matter content of the extract. Preferably, the extract comprises hydrangenol in an amount of from 0,1 to 7.5 wt.-%, preferably in an amount of from 0.5 to 6 wt.-%, preferably in an amount of from 1 to 5 wt.-%, based on the total weight of the dry matter content of the extract.

It is preferred that in the plant extract the weight ratio of phyllodu Icin and hydrangenol is at least 3:1 , preferably at least 5:1 , particularly preferably at least 7:1 , further preferably at least 10:1 , more preferably at least 15:1 , most preferably at least 25:1 , even further preferably at least 30:1 , preferably at least 35:1 , preferably at least 40:1 , preferably at least 45:1 , preferably at least 50:1 , preferably at least 55:1 , preferably at least 60:1 , preferably at least 65:1 , preferably at least 70:1 . Further aspects and advantages of the invention result from the subsequent description of preferred examples.

Examples

Example 1 : Software-based prediction of the phyllodulcin contents based on the respective natural deep eutectic solvent

The suitability of different substances as potential NADES components for extracting phyllodulcin was evaluated using a software-based prediction (COS MO- RS), which allowed to calculate the activity coefficient (y) of the compound in different NADES systems. The activity coefficient y°° ^s of solute / infinitely diluted in solvent s is defined as: where ® is the chemical potential of the solute / in the solvent s and ₍. is the chemical potential of pure solute /.

For software-based prediction the conformers of phyllodulcin, choline chloride, and betaine were calculated in BIOVIA COSMOconfX (Version 22.0.0, Dassault Systemes, Velizy-Vil- lacoublay, France) using the Becke-Perdew functional (BP) and a triple-zeta valence polarization with diffuse functions (TZVPD) and a fine grid marching tetrahedron cavity (FINE) template. They had a full geometry optimization with the density functional theory (DFT) at the BP-TZVP level, with a consecutive BP-def2-TZVPD single-point calculation and a FINE cavity for the COSMO calculation. The conformers were considered a Boltzmann-weighted mixture of conformers for the calculations, and the maximum number of conformers was set to 75. Calculated structures were verified to be true minima using vibrational frequency analysis. All other NADES components were taken from the COSMOtherm database.

For modelling the component choline chloride, the so-called ion pair approach was used. Thereby, the COSMO-RS optimized structure of choline chloride could be described as a single non-dissociated molecule (e.g. described in Diedenhofen, M.; Klamt, A. COSMO-RS as a Tool for Property Prediction of IL Mixtures — A Review. Fluid Phase Equilib. 2010, 294 (1-2), 31-38. https://doi.Org/10.1016/j.fluid.2010.02.002). The organic acids were treated as protonated compounds. NADES were treated as binary mixtures of two components (HBD and HBA) at different stoichiometric ratios within the framework of COSMO-RS.

For this approach, only the conformer with the lowest energy level of phyllodulcin, the HBAs, as well as the HBDs, were used. The software BIOVIA COSMOthermX (Version 22.0.0, Dassault Systemes) with the BP_TZVPD_FINE_22.ctd parameterization was used to calculate the activity coefficient of phyllodulcin in NADES at 25 °C, and infinite dilution with 30 wt.-% of water.

The tools applied in this approach are described e.g. in Klamt, A. Conductor-like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena. J. Phys. Chem. 1995, 99 (7), 2224-2235. https://doi.Org/10.1021/j100007a062.

Klamt, A.; Jonas, V.; Burger, T.; Lohrenz, J. C. W. Refinement and Parametrization of COSMO-RS. J. Phys. Chem. A 1998, 102 (26), 5074-5085. https://doi.orq/10.1021/jp980017s.

Eckert, F.; Klamt, A. Fast Solvent Screening via Quantum Chemistry: COSMO-RS Approach. AIChE J. 2002, 48 (2), 369-385. https://doi.Org/10.1002/aic.690480220.

BIOVIA COSMOtherm, Release 2022; Dassault Systemes. http://www.3ds.com.

With a decreasing activity coefficient, the tested combination is better able to solubilize and extract phyllodulcin. Accordingly, if the inverse value 1/ln y is considered, a higher value indicates a better ability of the NADES to extract phyllodulcin.

Based on this software-based prediction, suitable NADES are to be identified in advance without the need for laboratory work. The above combinations were identified as potential suitable NADES for extracting phyllodulcin.

Further combinations were tested, such as the combinations of L-Proline and glycerol, as well as of malic acid and fructose. However, these combinations tend to have a lower phyllodulcin activity coefficient and were thus not considered as suitable NADES systems for extracting phyllodulcin.

Example 2: Preparation of a natural deep eutectic solvent

The natural deep eutectic solvents were produced as follows:

The single components (e.g. malic acid, fructose, water) are weighed and mixed with each other. The mixture is heated to approximately 70 °C under stirring. The mixture is stirred until a clear and homogeneous mixture is obtained.

Example 3: Extraction of phyllodulcin

Leafs of Hydrangea macrophylla ssp. serrata were dried. Subsequently, the dried leaves were moistened with the double amount (w:w) water and mixed. The obtained plant material was wet fermented for 2 hours at 40 °C.

Subsequently, the obtained wet fermented leaves were extracted with the 20-fold (w:w) amount of natural deep eutectic solvent, which was prepared as described in Example 2, for 2 hours at 40 °C under stirring. The obtained mixture was then filtered with a filter with a mesh size of 400 pm (US mesh size 40).

Several different natural deep eutectic solvents were tested in parallel approaches.

Subsequently, the obtained amount of phyllodulcin was measured in the extracts with a HPLC-UV-method with external calibration.

Based on the results provided in Example 1 , the following natural deep eutectic solvents (according to the invention) were tested:

As a comparison (not according to the invention), further natural deep eutectic solvents were tested: As a further comparison (not according to the invention), other solvents were tested:

It was found that the natural deep eutectic solvents of the method according to the invention provided much higher amounts of phyllodulcin in the extracts compared to the further tested natural deep eutectic solvents. Advantageously, the amounts of phyllodulcin obtained with the natural deep eutectic solvents of the method according to the invention were comparable to those obtained using methanol or ethanol.

Interestingly, the software-based prediction was confirmed for the above natural deep eutectic solvents, as shown in Example 1 .

Example 4: Extraction of phyllodulcin - purification step vii) A plant extract was prepared as described in Example 3, using the following natural deep eutectic solvent:

The obtained extract was diluted (1 :10, v:v) with water and stored for 9 days at 3 °C. Subsequently, the mixture is centrifuged for 5 min at 8000 rpm. The supernatant is discarded. The pellet is washed twice with water, centrifuged again under the same conditions. The obtained pellet was then analysed for its phyllodulcin content with a HPLC-UV-method with external calibration.

A product comprising 56.08 wt.-% phyllodulcin was obtained.

Example 5: Extraction of phyllodulcin - purification step vii)

A plant extract was prepared as described in Example 3, using the following natural deep eutectic solvent:

The obtained extract was diluted (1 :10, v:v) with a mixture of ethanol and water (2:8, v:v). The diluted extract was applied to a column (250 x 25 mm, 100 mL), with cross-linked poly(styrene-divinylbenzene) as adsorbent. The column was washed with a mixture of ethanol and water (2:8, v:v). The column was eluted with ethanol. The ethanol was evaporated at the rotary evaporator. The residue was then analysed for its phyllodulcin content with a HPLC-UV-method with external calibration.

A product comprising 36.28 wt.-% phyllodulcin was obtained.

Example 6: Extraction of phyllodulcin - purification step vii) A plant extract was prepared as described in Example 3, using the following natural deep eutectic solvent:

The obtained extract was diluted (1 :1 , v:v) with water. The diluted extract was applied to a column (250 x 25 mm, 100 mL), with cross-linked poly(styrene-divinylbenzene) as adsorbent. The column was washed with a mixture of ethanol and water (2:8, v:v). The column was eluted with ethanol. The ethanol was partly evaporated at the rotary evaporator. Subsequently, phyllod ulcin was crystallized from the obtained concentrated ethanolic mixture. The crystals were filtered and washed with cold ethanol. The obtained product was then analysed for its phyllodulcin content with a HPLC-UV-method with external calibration.

A product comprising 88.32 wt.-% phyllodulcin was obtained.

Example 7: Comparison of sensory characteristics

The following mixtures were prepared: For the sensory evaluation the samples were evaluated by a trained test panel. The following results were obtained:

The above experiments show that with a method according to the invention a product can be obtained, which provides an advantageous sensory evaluation corresponding to a product obtained with other extraction solvents, however already at a lower amount. Example 8: Sweetness evaluation

The products of Example 5 (sample 1) and Example 6 (sample 2) were obtained and compared with the pure substance (R)-phyllodulcin with regard to their sweetness, which was evaluated by a trained test panel. The following samples were tested: The sensory evaluation of sweetness revealed that all tested samples provided a comparable sweetness.

Claims

1. Method for producing a plant extract comprising phyllodulcin, wherein the method comprises or consists of the following steps: i) drying the plant material to be extracted, preferably wherein the plant material predominantly consists of plant leafs, preferably wherein the plant material is of a plant of the species Hydrangea macrophylla preferably of the subspecies serrata, more preferably selected from the group consisting of the varieties Oamacha, Amacha and Amagi-Amacha, ii) moistening the dried plant material of step i) with water and wet fermenting the plant material at a temperature in the range of from 10 to 50 °C, preferably in the range of from 30 to 45 °C, for 0.5 to 72 hours, preferably for 0.5 to 16 hours, particularly preferably for 1 to 5 hours, iii) subjecting the wet fermented plant material of step ii) to an extraction with a natural deep eutectic solvent, at a temperature of from 0° C to the boiling point of the respective solvent, wherein the natural deep eutectic solvent comprises at least 2 components selected from the group consisting of lactic acid, choline chloride, Proline, preferably L-Proline, Betaine, 1 ,2-propane diol, glucose and malic acid, iv) collecting the supernatant (S1) of the extraction of step iii) and optionally: repeating step iii), collecting the supernatant (S2) and combining supernatants S1 and S2, and v) filtering the supernatant or combined supernatants obtained after step iv) and collecting the filtrate.

2. Method according to claim 1 , wherein after the extraction in step iii) and/or repeated step iii) according to step iv), the extract is purified by solid-phase adsorption.

3. Method according to claim 2, wherein the adsorbent used in the solid-phase adsorption is selected from the group consisting of polystyrene, aliphatic methyl acrylate and mixtures thereof, preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, particularly preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, and the polystyrene is macroporous.

4. Method according to any one of the preceding claims, wherein the extraction in step iii) is performed over a time in the range of from 0.5 to 10 hours, preferably 0.75 to 5 hours, particularly preferably from 1 to 4.5 hours.

5. Method according to any one of the preceding claims, wherein the method further comprises the step vi) diluting the filtrate collected in step v) with a diluent, preferably wherein the diluent is selected from the group consisting of water; protic solvents, such as alcohols, preferably methanol, ethanol or isopropanol, and carbonic acids, preferably methanoic acid, acetic acid, propanoic acid; and mixtures thereof.

6. Method according to any one of the preceding claims, wherein the method further comprises the step vii) purifying the filtrate collected in step v) or the diluted filtrate in step vi), preferably by a method selected from the group consisting of precipitation, centrifugation, crystallization, solid-phase adsorption, filtration, reverse osmosis and combinations thereof.

7. Method according to claim 6, wherein the purification in step vii) comprises solidphase adsorption and wherein the adsorbent used in the solid-phase adsorption is selected from the group consisting of polystyrene, aliphatic methyl acrylate and mixtures thereof, preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, particularly preferably wherein the adsorbent is polystyrene or a mixture of polystyrene and aliphatic methyl acrylate and the polystyrene is cross-linked, preferably with a divinyl benzene, and the polystyrene is macroporous.

8. Method according to any one of the preceding claims, wherein the plant material comprises at least 2.5 wt.-%, preferably at least 3 wt.-%, particularly preferably at least 4 wt.-%, especially preferably at least 5 wt.-% phyllodulcin equivalents, based on the total weight of the plant material.

9. Method according to any one of the preceding claims, wherein the plant material comprises 2 wt.-% or less, preferably 1 .5 wt.-% or less, particularly preferably 1 .0 wt.-% or less, especially preferably 0.5 wt.-% or less hydrangenol equivalents, based on the total weight of the plant material.

10. Use of a natural deep eutectic solvent in a method according to any one of the preceding claims, wherein the natural deep eutectic solvent comprises at least 2 components selected from the group consisting of lactic acid, choline chloride, Proline, preferably L-Pro- line, Betaine, 1 ,2-propane diol, glucose and malic acid.