WO2024154051A1 - Controlled release system of 1-deoxynojirimycin - Google Patents

Abstract

The invention relates to a system for the controlled release of 1-Deoxynojirimycin. More specifically, the invention refers to a method for: treating an imbalance of blood glucose levels, lowering blood glucose levels, treating a glucose metabolism disorder, treating hyperglycemia, diabetes, diabetes mellitus and prediabetic stages in a subject, treating a disorder linked to an anomalous accumulation of body fat such as obesity in a subject, said method comprising administering a composition comprising an anionic polymer and 1-DNJ or a natural extract containing 1-DNJ, wherein the combination is formulated to release a therapeutically effective amount of 1-DNJ mainly at an intestinal level.

Description

CONTROLLED RELEASE SYSTEM OF 1-DEOXYNOJIRIMYCIN

Technical field

The present invention relates to a system for the controlled release of 1 - Deoxynojirimycin (hereinafter also indicated as 1 -1 -DNJ or simply 1 -DNJ- Cas Number 19130-96-2). According to the present invention, the release system can be obtained either by using 1 -DNJ as a pure product or even with 1 -DNJ present in a natural extract (EXT in the following). The extract can be obtained according to the extraction techniques of natural products known to the expert in the field.

More specifically, the invention refers to a method for:

- treating an imbalance of blood glucose levels in a subject who needs it,

- lowering blood glucose levels into a subject who needs it,

- treating a glucose metabolism disorder in a subject who needs it,

- treating hyperglycemia, diabetes, diabetes mellitus and prediabetic stages in a subject who needs it,

- treating a disorder linked to an anomalous accumulation of body fat like obesity in a subject who needs it, said method comprising the administration to said subject of a composition comprising a combination of an anionic polymer and 1 -DNJ which contains a therapeutically effective amount of 1 -DNJ or its derivative or its salt or a natural extract containing 1 -DNJ, wherein the combination is formulated to release a therapeutically effective amount of 1 -DNJ mainly at the intestinal level or wherein the combination is formulated to release a first therapeutically effective amount of 1 -DNJ in the stomach and a second therapeutically effective amount of 1 -DNJ mainly at the Intestinal level, the first amount being lower than the second amount.

Known Art

Diabetes mellitus (DM) is a complex metabolic disorder, which affects the natural regulation of blood glucose levels. This condition is characterized by chronic hyperglycemia and carbohydrate and fat metabolism disorders, deriving from insufficient insulin secretion (type 1 DM) or insulin resistance (type 2 DM). The spread of diabetes is increasing all over the world and is becoming one of the main health concerns according to the World Health Organization, which requires new strategies and therapies with an acceptable benefit costs ratio (1 ). Several studies have shown that phytochemical substances from natural sources represent a suitable and promising approach to the management of diabetes, especially in the initial step of the disease (2). The extract of mulberry leaves (Morus spp. L.) has been used in traditional medicine in Asian countries for the treatment of hyperglycemia and DM since ancient times (3-5).

1 - Deoxynojirimycin (1 -DNJ or 1 -1 -DNJ), also called Moranolin and known in medicine as Duvoglustat, is an imino sugar present in the fruits, leaves, roots of the mulberry plants belonging to the genus Morus.

1 -1 -DNJ was demonstrated to be a powerful hypoglycemic capable of inhibiting the intestinal enzymes involved in the digestion of sugars. In addition, 1 -1 -DNJ showed marked systemic activities against hyperglycemic states. It is in fact able to reduce the expression of glucose transporters and increase the expression of mRNA of liver enzymes involved in the metabolism of glucose, thus accelerating its use on a systemic level. Finally, 1 -1 -DNJ is able to increase the sensitivity to insulin and to activate the in vivo [3-oxidation of fatty acids. All these activities, also demonstrated through clinical trials, make 1 -1 -DNJ a powerful active ingredient to counteract not only diseases related to hyperglycemic states such as diabetes, but also obesity. In addition, 1 -1 -DNJ has shown to play a pharmacological chaperon role for a-glucosidase enzyme for the treatment of a-1 ,4-glucosidase deficiency, involved in lysosomal accumulation diseases.

The well-documented inhibition of the intestinal a-glycosidases exercised by Deoxynojirimycin (1 -DNJ) makes it a successful anti-hyperglycemic agent, slowing down the conversion rate of polysaccharides into sugars (6, 7). In addition, 1 -DNJ has several positive systemic effects against the hyperglycemic state. In fact, it has been shown that it reduces the expression of glucose transporters and increases the expression of mRNA of liver enzymes involved in the metabolism of glucose, accelerating their use and disposal (5). Finally, it has been shown that 1 -DNJ is able to improve insulin sensitivity and activate in vivo [3-oxidation of fatty acids (3). The extract of Morus leaves has been shown, in addition to 1 -DNJ, to also contain other active constituents (mainly kaempferol, quercetin and chlorogenic acid) that carry out an antiglycating synergistic activity and trapping of the dicarbonilic compounds (8).

For this reason, the administration of the extract would benefit from the synergy of various components and the combined effect can help preventing long-term complications in diabetic patients. Taking into account the biological activities mentioned above, the mulberry thanks to the presence of 1 -DNJ active metabolite, can be considered one of the most promising functional foods for the management of DM. Unfortunately, the systemic effects of 1 -DNJ are reduced by its short halflife due to the high hydrophilia of the molecule. As previously reported, 1 - DNJ is quickly eliminated by the body via renal excretion (9). Therefore, constant intake by repeated administration is required to reach the effective therapeutic dose. On the other hand, drug-transporting technologies improve biodistribution of therapeutic agents, improving their biological effect. In particular, the encapsulation of molecules in specific polymers allows to control the release profile by means of the vector's chemical and physical characteristics.

As reported above, all systemic activities of 1 -1 -DNJ are significantly limited by its short half-life. Since 1 -1 -DNJ is a highly hydrophilic molecule, once absorbed through the intestinal glucose transporters, it is quickly eliminated through renal clearance. To increase its systemic activities, the chemical structure of 1 -1 -DNJ has been modified to increase its lipophilia. 1 -1 -DNJ was used as "lead compound" for the synthesis of molecules currently on the market as drugs for the treatment of type II diabetes (Miglitol) and Gaucher's lysosomal genetic disease (Miglustat). In general, a method of increasing the systemic activities of a short halflife compound is to slow down its intestinal absorption so that its blood concentration remains more constant over time.

In the literature there is a study wherein a Morus spp extract is conveyed by microparticles based on gelatin, a cationic polymer. In this study, a controlled release of 1 -DNJ has been observed, but almost 60% of the active ingredient was released in the first minutes from the intake (10).

It would be of great use to develop a system capable of releasing 1 -DNJ in the intestinal tract. This release system would allow to have a prolonged hypoglycemic effect over time.

Furthermore, and not to be overlooked, there is an important side effect of the oral intake of 1 -DNJ, that is, the one for which, when taken in significant quantities (i.e. in the absence of a slow-release mechanism) it recalls water in the intestine and can cause diarrhea, effect that could instead be avoided by the slow release at the intestinal level.

The need is therefore felt to have a release system available that is effective and at the same time it is easy to prepare and have low costs.

If not specifically excluded in the detailed description that follows, what is described in this chapter is to be considered as an integral part of the detailed description of the invention.

Summary of the invention

The present invention aims to provide a solution to the problems exposed above through the development of a system for the controlled release of 1 -Deoxynojirimycin (in the following also indicated as 1 -1 -DNJ or simply 1 -DNJ-Cas Number 19130 -96-2).

Further purpose of the invention is to provide a method to lower blood glucose levels in a person who needs it, comprising administering a composition to said subject comprising a combination of an anionic polymer and 1 -DNJ which contains a therapeutically effective amount of 1 -DNJ or its derivative or a salt or a natural extract containing 1 -DNJ, wherein the combination is formulated to release 1 -DNJ mainly in the intestine.

Further purpose of the invention is to provide a method for:

- treating an imbalance of blood glucose levels in a subject who needs it,

- lowering blood glucose levels into a subject who needs it,

- treating a glucose metabolism disorder in a subject who needs it,

- treating hyperglycemia, diabetes, diabetes mellitus and prediabetic stages in a subject who needs it,

- treating a disorder linked to an anomalous accumulation of body fat like obesity in a subject who needs it, said method comprising the administration to said subject of a composition comprising a combination of an anionic polymer and 1 -DNJ which contains a therapeutically effective amount of 1 -DNJ or its derivative or its salt or a natural extract containing 1 -DNJ, wherein the combination is formulated to release a therapeutically effective amount of 1 -DNJ mainly at the intestinal level or wherein the combination is formulated to release a therapeutically effective amount of 1 -DNJ in the stomach and a second therapeutically effective amount of 1 -DNJ mainly at the Intestinal level, the first amount being lower than the second amount.

Further purpose of the invention is to provide pharmaceutical compositions and their uses as defined in the attached claims.

Further purpose of the invention is to provide pharmaceutical compositions, functional foods or food supplements comprising a combination of an anionic polymer and 1 -DNJ as mentioned above.

Brief description of the figures

Further purposes and advantages of the present invention will be clear from the detailed description that follows, an example of realization of the same (and its variants) and by the attached figures, provided for pure explanatory and non -limiting scope, wherein: Figure 1. Overlapping the MRM transitions monitored for 1 -DNJ (A) and the related ESI-MS spectrum (B) for the mulberry leaves extract.

Figure 2. ESEM images of beads unloaded and loaded with the EXT at 2 and 3% of the ALG to the enlargement of 100 or 150x (A) and 500x (B).

Figure 3. ESEM images of SDMs unloaded (A) and SDMs loaded with the EXT (B) to the enlargement of 2000x.

Figure 4. FTIR spectra of mulberry extract (EXT), blank and Ca-ALG-2% loaded EXT beads (top) and SDM (bottom).

Figure 5. ¹H NMR spectrum and 1 -DNJ assignments in deuterated water (D2O). Chemical shifts relate to the TSP.

Figure 6. Dynamic absorption of water from the Ca-ALG beads in gastric (pH 3) and intestinal (pH 6.8) simulated fluid.

Figure 7. 1 -DNJ in vitro release from the Ca-ALG beads in gastric (0-120 min) and intestinal (120-360 min) simulated fluid. The magnification in the outlined box shows the release profile in the first minutes (booster effect).

Figure 8. In vitro release of 1 -DNJ from SDMs in simulated gastric (0-120 min) and intestinal (120-360 min) fluid.

Definitions

The term "about" as used herein when referring to a measurable value such as an amount, a temporal duration, and the like, is intended to enclose variations of ± 20%, ± 10%, ± 5%, ± 1%, or ± 0.1% of a specified value, where these variations are appropriate to perform the methods described.

The wording "near to 7" as used herein indicates the intestinal pH near neutrality, in a range between about 6 and about 8.

The term "polymer" as used herein indicates a macromolecule (high molecular weight) consisting of repetitive units linked together by a covalent bond. The term "anionic polymer" indicates a pH sensitive polymer because it consists of monomers containing acidic functional groups (such as carboxylic acids, sulphonic acids, phosphonic acids) and comprises, but is not limited to, alginic acid, carrageenan, hyaluronic acid, pectin, carboxymethylcellulose, also in the form of salt.

As used herein, the term "polysaccharide" refers to a polymeric molecule consisting of a number greater than ten of monosaccharide units linked together through glycosidic bonds.

The term "alginate, high G/M ratio" as used herein means a linear anionic polysaccharide consisting of a high amount of [3-l-guluronate (G) monomers and a lower amount of a-D-mannuronate monomers (M) (preferably the G/M ratio is about 70:30).

The term "Ca-ALG beads" as used herein means millimeter particles of calcium alginate, obtained by cross-linking sodium alginate with bivalent calcium cations (Ca²⁺).

The term "microparticle(s)/SDM/SDMs" as used herein means microparticle/s obtained by spray-drying.

An "effective amount" and "therapeutically effective amount" as used herein refers to an amount of a pharmaceutical composition or a medicinal agent that is sufficient to provide a desired effect. In some embodiments, a doctor or other healthcare professional decides the appropriate amount and dosage regime.

Detailed description of the invention

The present invention relates to a way to increase the systemic activities of a short half-life compound such as 1 -DNJ slowing down its intestinal absorption so that its blood concentration remains more constant over time.

In the present invention, a method has been developed to increase the systemic effect of 1 -DNJ which exploits a controlled release mechanism over time. The administration system that is an object of the present invention provides for the preparation of an electrostatic complex between an anionic polymer (for instance but not limited to sodium alginate) and the active ingredient 1 - Deoxynojirimycin (1 -DNJ) or its derivative or a salt or a natural extract that contains the 1 - Deoxynojirimycin or its derivative or a salt, said electrostatic complex preferably but not limited in the form of calcium-alginate particles (Ca-ALG) or SDMs.

1 -DNJ, being an imino sugar, has an imino group that is able to charge positively according to the environment wherein the molecule is placed and which can interact electrostatically with negatively charged molecules, thus creating a sort of electrostatic complex. The formation of this complex has been observed, together with other types of electrostatic interaction, through the studies of the present inventors wherein they tried to create two types of particles (for instance, but not limited to alginate) incorporating the extract of mulberry that has proven to be able to slow down the release of 1 -DNJ in gastrointestinal fluids (for instance, but not limited to, Ca-ALG beads and SDMs).

Ca-ALG beads can be obtained by ionic jellification (technique known to the expert in the field). This process, which can be conducted with room pressure and ambient temperature and a pH around neutrality, comprises the step of dripping within a solution of a precipitating agent, preferably calcium chloride (CaCh), a dry EXT of mulberry leaves and sodium alginate (Na-ALG) dissolved in water in order to promote the formation of the Ca-ALG beads. The thus obtained Ca-ALG particles, once filtered from the solution and dried for about 24h, appear with a spherical morphology and a diameter between 1 .1 -1 .5 mm.

SDMs can be obtained by means of a spray-drying process. This process involves the removal of the solvent from a hydroalcoholic EXT solution of mulberry leaves and Na-ALG in water, through the help of a spray-dryer. The particles obtained show a spherical morphology, a smooth surface and a bimodal distribution of the diameters such that, 80% of the particles, has an average diameter of 1 1 pm.

Both Ca-ALG beads and SDMs are obtained through procedures performed at a pH around neutrality. These pH conditions are such as to induce the formation of the electrostatic complex (polymer in anionic form and 1 -DNJ in protonated form). Both in the Ca-ALG beads and in the SDMs, since the starting solutions are homogeneous, the distribution of the mulberry leaves EXT in the release system is homogeneous. In addition, the release tests in simulated gastrointestinal conditions show that the fraction of 1 -DNJ present on the surface of the particles (both in the form of beads and SDMs) is released quickly (booster effect) in the first minutes; on the contrary, the fraction of 1 -DNJ present inside the particles is gradually released over time.

From the release studies of 1 -DNJ in an aqueous solvent simulating the gastrointestinal tract, it was surprisingly found that 1 -DNJ remains linked to the anionic polymer solubilized in the fluid, even when, in simulated intestinal conditions, the particles are at least eroded. In fact, the free 1 - DNJ (solubilized in the simulated intestinal fluid) was less than 43%, suggesting that 57% of 1 -DNJ remains linked to the anionic polymer despite the complete erosion and solubilization of the Ca-ALG beads and the complete solubilization of SDMs at the intestinal pH of 6.8. The complete erosion of the Ca-ALG beads is caused by the exchange of Ca²⁺ with Na⁺ which breaks the cross-linking bonds of the particles, leading to the formation of Na-ALG. The subsequent solubilization, observed also in the case of the SDMs, takes place due to the solubility of Na-ALG in an aqueous environment at pH greater than 5.

It was therefore evident that to retain 1 -DNJ in solution there is no need to use particularly complex particles or vehicles (such as the Ca-ALG beads), as the presence of a polymer or excipient negatively charged in the solution wherein 1 -DNJ is dissolved is sufficient not to make 1 -DNJ free and therefore available.

The stability and strength of the anionic/1 -DNJ polymer complex are mainly based on an electrostatic interaction ruled by pKa of 1 -DNJ and the polymer and the pH of the solvent wherein they are, conditions that determine how many basic and acidic groups respectively are present in dissociated form. The anionic polymer and 1 -DNJ interact when both are mainly or in part in their dissociated form, deprotonated as regards the anionic polymer (loss of a proton in the carboxylic groups of the polymer monomers) and protonated as regards 1 -DNJ (acquisition of a proton on the nitrogen of the imino sugar) respectively.

By definition, pKa (negative logarithmic value of the acid dissociation constant) corresponds to the value of pH to which a molecule is present for about 50% in dissociated form. Molecules containing acidic groups will mainly be in their deprotonated form with pH> pKa; on the contrary, molecules containing basic groups will mainly be in their protonated form with pH <pKa. By "mainly" we mean a percentage greater than 50%. Therefore, anionic polymers (such as alginate) with pKa close to 3 are in their deprotonated form (i.e. anionic form) for about 50% at pH 3 (which is the pH of the gastric environment) and mainly deprotonated at higher pH. Imino sugars such as 1 -DNJ, having pKa next to 8 are in their protonated form for about 50% to pH 8 and mainly protonated at lower pH.

The pH conditions indicated above mean that at a gastric level 1 -DNJ which is located on the external surface of the electrostatic complex is released for the partial protonation of the carboxylic groups of the anionic polymer.

Conversely, at the intestinal level (pH close to neutrality) thanks to the gradual restoration of the deprotonated form (i.e. anionic form) of the polymer there will be the restoration of the electrostatic complex with the remaining 1 -DNJ, that is, of that portion that has been trapped within the Ca-ALG or SMDS particle. Inventors have shown that polymers rich in acidic groups (for instance carboxylic acids or sulphate groups) are able to link through electrostatic interactions, and therefore retain, 1 -DNJ which is not free in the gastrointestinal fluid and bioavailable for subsequent absorption by intestinal epithelium in a pH range between about 3 and about 7.

However, it is necessary to consider that the pH in the stomach commonly has a value around 3. This condition of marked acidity is such as to induce the protonation of about 50% of the carboxylic groups of the anionic polymer used, with the consequent loss of electrostatic interaction with 1 - DNJ. Conversely, in the intestinal tract the pH is around neutrality and induces the anionic polymer to restore the deprotonated form. It therefore involves the restoration of all electrostatic interactions with 1 -DNJ.

So when the particles subject of the present invention, based on the electrostatic complex described above, reach the stomach, they undergo a reduction of the electrostatic interactions in the most external and more exposed to the environment particle layers, such as to justify a release of 1 -DNJ at the level of the stomach. However, considering the high number of carboxylic functionalities present on the polymer and the pH conditions not such as to induce a quantitative protonation of the carboxylates, in addition to the insolubility of the polymer itself in the gastric environment, this release remains reduced. After the stomach, the raising of the pH, however, favors the restoration of the electrostatic interactions between the anionic polymer and 1 -DNJ, which, due to the pH conditions of the entire gastrointestinal tract, will survive as iminium ion through all digestion. The reacquisition of these ionic interactions promotes the anchoring of 1 -DNJ to the anionic polymer, limiting their solubilization and allowing a prolonged release over time.

It is important to highlight how the retention capacity of 1 -DNJ or of its derivative or of its salt or any pharmaceutically acceptable form of it, in addition to being linked to environmental conditions, is strictly dependent on pKa of the anionic polymer used. In fact, as the acid dissociation constant (pKa) of the anionic polymer increases, the acidity of its carboxylic groups decreases. Said carboxylic groups, if therefore immersed in a gastric pH environment, will proton more easily causing a significant attenuation of the electrostatic interaction with the imino sugar which will therefore be released more easily. So, is an object of the present invention a controlled release system comprising an electrostatic complex consisting of 1 -DNJ or its derivative or its salt or a pharmaceutically acceptable form, and an anionic biocompatible polymer preferably chosen from the anionic polymers having preferably pKa less than 4. Preferably, pKa of the anionic polymer is less than 3.9, or preferably less than 3.8, or preferably less than 3.7, or preferably less than 3.6, or preferably less than 3.5.

The proportions between 1 -DNJ and polymer for the creation of an electrostatic complex capable of retaining the mixture therefore vary according to the polymer used, its length and therefore from how many groups are negatively charged. It has been shown here that as long as the negative groups are in excess, the active principle of opposite charge is retained.

As mentioned above, the acidity conditions in the intestinal tract are such that both the anionic polymer and 1 -DNJ are mainly found in their respective ionic forms. The electrostatic interactions that are triggered between the two species are therefore maximized as well as the retention of 1 -DNJ which, remaining complexed with the polymer, is not bioavailable for the absorption by the intestinal epithelium. The electrostatic complex, however being immersed in an aqueous environment, is subject to a spontaneous chemical equilibrium schematized below.

wherein Pa is the active ingredient and P is the polymer.

The equilibrium reaction is such as to induce a gradual release of the active ingredient 1 -DNJ which can therefore be absorbed in a moderate and constant way over time in the intestine, thus increasing its half-life and effectiveness.

In fact, when it is in this equilibrium state, Pa "sequestering" from the solution (following intestinal absorption for instance) makes it possible that part of the PaP complex "breaks" to restore equilibrium. Therefore, in vivo, 1 -DNJ free from the complex would be absorbed at the intestinal level a little at a time because the complex would gradually release it to restore the equilibrium little by little. In the case of Ca-ALG beads, this equilibrium seems to be favored also by the gradual ionic exchange of calcium with sodium at the intestinal level. In fact, since this cationic replacement involves the introduction of a monovalent cation, the cross-linking degree of the anionic polymer is lowered by promoting its opening. It therefore induces an increase in the surface exposed to the aqueous solvent. In fact, the opening of the polymer network allows swelling, i.e. the swelling of the polymer made by aqueous solvent molecules that manage to penetrate the inner areas of the particle and which favor the gradual release of 1 -DNJ molecules most retained/restrained. Differently, in the case of SDMs, since these consist of Na-ALG, no swelling is shown; the particles undergo a gradual solubilization which in any case involves a release of 1 -DNJ controlled over time.

Following the surprising evidence observed during the in vitro release tests, the electrostatic interaction between the polymer and 1 -DNJ has also been demonstrated by means of a nuclear magnetic resonance spectrometry (NMR), by recording the relaxing time of the protons of 1 - DNJ.

Therefore, in the present invention this chemical interaction was exploited using polymers rich in functional groups negatively charged in the form of acids or salts, such as alginates, carrageenan, hyaluronic acids, pectins, carboxymethyl celluloses, etc., or excipients for technological use, such as lecithins, surfactants.

To obtain this type of interaction, specific processes or specific methods are therefore not necessary. Simply, it is sufficient to mix 1 -DNJ as such or a natural extract that contains it with the anionic polymer, even without preventive solubilization in aqueous solvents. In fact, the complex is formed when 1 -DNJ and the polymer enter the solution or in any case in contact with aqueous solvents.

The relative quantities of anionic polymer and 1 -DNJ (or a natural extract) are not particularly binding, preferably the molar ratio between alginate and 1 -DNJ is 1 : 0.001 -1 :300.

The interaction between the anionic polymer and 1 -DNJ allows to prolong the systemic effect of the active ingredient. This evidence can be used to create new food supplements based on more effective Morus spp extract for the treatment of hyperglycemia conditions (diabetes mellitus and prediabetic stages) and obesity as an adjuvant to the diet.

The use of biocompatible anionic polymers for the formulation of a supplement for the administration of Morus spp extracts or 1 -DNJ or synthetic derivatives of 1 -DNJ has never been taken into consideration or demonstrated to date. The advantage of the use of anionic polymers lies in the fact that they provide prolonged action of the active ingredient on a systemic level.

In fact, the inventors have experimentally found that 1 -DNJ (as such or as a synthetic derivative or as a vegetable extract) is released in the body with a double release: in the stomach the first release step takes place with an initial boost of about 30% of the active ingredient and then in the intestine the remaining 1 -DNJ is released with slower kinetics. This mechanism entails the big advantage that, if the intake of 1 -DNJ is made during the meal, there will be an initial amount of drug, necessary at the first stage of digestion, and subsequently, in the intestine, there will be a slow and gradual release, constant over time, that enzymes deputies to sugar digestion will inhibit.

Therefore, the best dosage will be assuming the pharmaceutical complex during the meal or at the end of the meal. A further advantage must be ascribed to the formulation of the invention and is due to a collateral effect of 1 -DNJ which, when it is taken orally in significant quantities (but in the absence of the slow release mechanism described in the present invention) recalls water in the intestine and it gives diarrhea, an effect that instead is avoided by the slow release in the intestinal level; in fact, the slow release avoids high release peaks and therefore does not give diarrhea and constitutes an undeniable advantage in favor of the formulation of the invention.

According to the present invention, the system for the controlled release of 1 -Deoxynojirimycin (hereinafter also indicated as 1 -DNJ- Cas Number 19130-96-2) comprises a combination of an anionic polymer and 1 -DNJ. Preferably the anionic polymer is chosen between, but not limited to, the following polymers in the form of acids or salts: alginates, carrageenan, hyaluronic acids, pectins, carboxymethyl celluloses, etc., or excipients for technological use, such as lecithin, surfactants. pKa of the anionic polymer determines the strength of the polymer-1 -DNJ complex. In fact, the lower pKa, the more the complex with a basic molecule (like 1 -DNJ) is strong. pKa of natural polymers such as alginates, carrageenan, hyaluronic acids, pectins, carboxymethylcelluloses and lecithin and anionic surfactants are lower than 4, a necessary requirement to observe a technical effect of controlled release. Polymers with pKa > 4 are able to form an electrostatic complex with basic molecules; however, at pH 6.8 the undissociated (deprotonated) form will be less abundant than the polymers with pKa <4. Neutral or cationic polymers do not interact electrostatically with 1 -DNJ. pKa of a natural polymer may vary depending on how many acidic groups are present. In the case of pectin, pKa increases as the esterification degree of carboxylic groups increases; in the case of carboxymethylcellulose, pKa is ruled by the number of substitutions of the hydroxylic groups with carboxylic groups.

The alginate in the form of sodium salt is particularly preferred, with a molecular weight in the range between 10,000 and 600,000 g/mol. To obtain a good cross-linking through bivalent cations (preparation of Ca- ALG beads), an alginate with high G/M ratio is preferable (for instance about 70:30. pKa of G and M are respectively 3.65 and 3,38 and pKa of the polymer depends on the G/M ratio. Therefore, the sodium alginate used has a pKa less than 3.65 (12).

1 -DNJ can be a synthesis product or its derivative or salt (13) or a vegetable extract that contains 1 -DNJ or its derivative or a salt. The derivatives of 1 -DNJ on the market are for instance Miglustat (also known as N-Butyl Deoxynojirimycin) and Miglitol (ID CAS: 72432-03-2). Other synthetic derivatives maintain the same imino sugar structure with the addition of an alkyl chain replacement on the nitrogen of the ring.

Preferably 1 -DNJ is contained in natural extracts, such as mulberry extracts. Although 1 -DNJ has been detected in small quantities also in the mulberry fruit, its expression is definitely greater at the level of the cortex, the roots and in the leaves. In addition, the amount of 1 -DNJ is higher in the leaves collected during the summer compared to other periods of the year (14). From previous studies it has been observed that the concentration of 1 -DNJ in extracts is higher if a 50% hydroalcoholic extractive solution is used. The mulberry leaves extracts are currently the most used in supplements for their high 1 -DNJ content. The advantage of using a raw extract of mulberry leaves is the co-presence of polyphenols who have shown important activities in contrasting pathological conditions strictly connected to diabetes and can also synergistically increase the inhibitory action of 1 -DNJ at the intestinal level on a-glucosidase (8).

The combination of the invention allows a slow release of the active ingredient 1 -DNJ mainly at the intestinal level.

Therefore, the administration of the combination of the invention allows to adjust the glucose levels in the blood in a subject who needs it. In particular, the combination of the invention allows:

- treating an imbalance of blood glucose levels in a subject who needs it,

- lowering blood glucose levels in a subject who needs it,

- treating a glucose metabolism disorder in a subject who needs it,

- treating hyperglycemia, diabetes, diabetes mellitus and prediabetic stages in a subject who needs it,

- treating a disorder linked to an anomalous accumulation of body fat like obesity in a subject who needs it.

The combination allows a controlled release of 1 -DNJ in the stomach and at the intestinal level, as a first therapeutically effective amount of 1 -DNJ is released in the stomach and a second therapeutically effective amount of 1 -DNJ is released at the intestinal level, the first amount being lower than the second amount.

An extremely simple methodology to prepare the slow-release combination according to the invention involves mixing together an alginate aliquot with a solution containing 1 -DNJ or its derivative or pharmaceutically acceptable salt. An example of preparation is provided with the experimental part.

The mixing temperature can advantageously be the ambient temperature or less than 40°C and the pH included in the range between 5 and 10, preferably around neutrality.

The alginate can be a salt of an alkaline or alkaline earth metal. Preferred are sodium or potassium or ammonium alginate. The G/M ratio is to be selected on the basis of the formulation that is desired: a high G/M ratio (example 70:30) is to be selected to obtain a good cross-linking by means of bivalent cations (preparation of Ca-ALG beads) (1 1 ). The molecular weight of the alginate influences its solubility in aqueous solutions and the amount of 1 -DNJ which can be electrostatically linked.

The solution containing 1 -DNJ can be an aqueous or hydroalcoholic solution containing up to 70% alcohol. Alcohol is preferably ethyl alcohol.

The solution containing 1 -DNJ can be a natural extract obtained from vegetable material. The extracts of mulberry leaves are particularly preferred. The preparation of the natural extract is a procedure within the reach of the expert in the field and an example of extraction is provided with the experimental part.

Said mixture is based on anionic polymer and 1 -DNJ, and possibly surfactants/emulsifiers with hydrophilic-lipophilic balance (HLB) greater than 10 (surfactants with HLB <10 are lipophilic and therefore not suitable for the formulation of the particles that takes place in an aqueous environment) such as lecithin, polysorbate, and/or additives/technological excipients necessary for the industrial development of a finished pharmaceutical formulation (thinners, lubricants, preservatives, dyes, flavoring, anti-caking agents) can be subjected to:

- simple mixing under stirring to obtain a solution or suspension. If a hydroalcoholic mulberry extract is used, it is preferable to remove alcohol at least partially from the solution by means of low-pressure evaporator before mixing with alginate to avoid an excess of ethyl alcohol in the solution/suspension. The solution or suspension thus obtained can be assumed as it is or subjected to subsequent treatments, such as, for instance:

- drying methods to obtain a dust such as vacuum drying (lyophilization), by nebulization (example spray-drying), thermal (example in fluid or static bed) or other perse known drying techniques.

- granulation methods in the dry or wet step to obtain pharmaceutical forms in the shape of granulates. Granulation can take place through oscillating granulators, fluid bed, rotor or other drying known techniques.

Cross-linking methods to obtain particles, for instance through the use of salts containing bivalent cations (such as calcium, magnesium). The formulation of cross-linked particles is a procedure within the reach of the expert in the field and an example of formulation is provided with the experimental part.

The slow-release combination or system of release of the present invention comprises or is constituted by the combination of 1 -DNJ with an anionic polymer, preferably alginate. The molar ratio between monomers of guluronic or mannuronic acid of the alginate and 1 -DNJ is about 1 :1 or higher.

Preferably the alginate has a molecular weight of 10,000 and 600,000 g/mol and the ratio by weight between the alginate and the extract is 1 : 0.05 - 1 : 1.5.

Preferably the release of the active ingredient 1 -DNJ takes place with the following mechanism:

-a first 1 -DNJ amount is released at pH 3-5

- a second 1 -DNJ amount is released at pH 6.5- 8

The first amount is about 30% and the release takes place in the stomach; the second amount is released over time to pH close to 7 of the intestine.

The pharmaceutical or food or dietary compositions according to the invention comprise the slow-release combination of the present invention and one or more pharmaceutically or dietary acceptable excipients. The compositions can also be formulated in combination with other pharmaceutically or dietary acceptable components such as, for instance, excipients for obtaining a finished pharmaceutical form (tablets, granulates, capsules, suspensions), preservatives, flavors, dyes etc.

The various pharmaceutical or dietary forms or food supplements of the present invention can be prepared using the preparation procedures and the related equipment commonly known and used in the pharmaceutical or dietary or preparatory dietary technique; consequently, the expert technician will not have any difficulty in adopting the most suitable procedure and equipment in making the desired pharmaceutical or dietary or food form according to the type of administration and chosen dosage. Therefore, no other explanatory details are needed.

Also as regards the dosage of the active ingredient of the present invention, the expert technician can select the same by using the information and teachings known for the use of drugs or food supplements already used in the sector for this purpose. A preferred therapeutic dosage range for mulberry extract containing 1 -DNJ is typically 300 mg per day (titrated at 1 or 2% of 1 -DNJ). The dosage of supplements provides for the intake of 3 tablets per day (each containing 100 mg of mulberry extract titrated at 1 or 2% of 1 -DNJ).

Pharmaceutical, food and dietetic forms can be chosen from: substantially aqueous solution or suspension, powder, tablet, granulate, and capsules for the preparation of food and/or nutraceutical and/or pharmaceutical products intended for oral administration, for the use in the treatment of the affections indicated above.

The aforementioned pharmaceutical forms can be administered in combination with other drugs such as insulin-based drugs, semaglutide, sulfonylureas, biguanides, thiazolinidones, repaglinide, orlistat.

The supplements currently on the market are tablets consisting of mulberry extract and the following pharmaceutical or equivalent excipients based on the manufacturer: microcrystalline cellulose and magnesium stearate as stabilizing agents, silicon dioxide as filling agent, and di calcium phosphate as anti-caking agent. Following the disintegration of the tablet, 1 -DNJ is promptly solubilized by the fluids of the gastrointestinal tract because highly soluble and quickly absorbed at the intestinal level. Hypoglycemic systemic activities are compromised by the low half-life of 1 -DNJ and fast renal clearance. The administration of the extract of mulberry leaves in pharmaceutical forms consisting of anionic polymers allows to obtain a controlled release over time at the intestinal level of the active ingredient 1 -DNJ. The controlled release allows to have a slow and prolonged absorption over time capable of maintaining a constant blood concentration of 1 -DNJ. In this way, the hypoglycemic action of 1 -DNJ is thus continued over time and not disadvantaged by its low half-life.

The production of pharmaceutical forms (aqueous-based solutions or suspensions, tablets, granulates, capsules) based on anionic polymers transporting 1 -DNJ is extremely simple and scalable on an industrial level. It is in fact based on methods and excipients commonly used for the manufacture of food supplements or drugs.

The following examples are given to illustrate the invention and are not to be considered limiting the relative scope. The average technician in the field will be able to deduce and implement operational variations without these modifying the results of the invention and the scope of this application.

EXAMPLES

MATERIALS AND METHODS

Chemicals and solvents

Acetonitrile and ethanol (HPLC grade), ammonium formiate (purity> 99.0%), calcium chloride, disodium hydrogen phosphate monohydrate, deuterium oxide (D2O), 1 -DNJ reference standard (purity 95.0%), hydrochloric acid and sodium salt of 3-(trimethylsilyl) propionic acid (trimetylsilyl-2,2,3,3-D4 (TSP) were provided by Merck Life Science S.r.L, (Milan, Italy). The sodium alginate (Na-ALG, high G/M, 70:30 ratio, molecular weight 1 15,000 g/mol) was obtained by Honeywell Fluka (Charlotte, NC, USA). The water was purified using a Milli Plus'! 85 system of Millipore (Milford, MA, USA).

Extraction of mulberry leaves

The leaves of Morus Alba (L.) collected by the Nervosa cultivar were used to obtain an extract rich in 1 -DNJ (in the following, extract = EXT). The vegetable material was supplied by Centro di Ricerca per I’Agricoltura e I’Ambiente (CREA), a sericulture laboratory located in the north-east of Italy (45 @24 57 96 N lat., 11 @52 58 08 E long. Padua, Italy). The collection was made in the summer months of 2020, when the content of 1 -DNJ was the highest, as demonstrated by a previous study conducted by Marchetti and collaborators (14). The leaves were dried in the oven at 50°C until they reached a constant weight and then ground in an automatic mill. The extraction was carried out by dynamic maceration on 10 g of leaves with 500 ml of 50% ethanol (v/v) at room temperature. The procedure required three consecutive extraction steps of 2 h each, the first and second with 200 ml of extractive solution and the last with 100 ml. The first EXT was centrifuged for 5 minutes at 7200 rpm, filtered on paper, two more extractions on the residue were subsequently performed. The extracts of Morus leaves were combined in a volumetric flask and brought to 500 ml with 50% ethanol (v/v). 1 -DNJ extraction procedure described here had been previously optimized and validated to obtain the best yield (14). To obtain a stable dried EXT to be incorporated into the beads, the ethanol was evaporated under vacuum at 50°C and the remaining aqueous suspension was freeze-dried (Heraeus Lyovac GT2, Leybold GmbH, Cologne, Germany).

1-DNJ determination through HPLC-ESI-TQ-MS analysis

The content of 1 -DNJ both in the liquid EXT and in the freeze-dried one was determined by means of a HPLC Agilent 6400 Series Triple Quadrupole. The liquid EXT was suitably diluted with acetonitrile for quantification. The lyophilized EXT was dissolved in 50% ethanol (v/v) at the concentration of 1 mg/ml, filtered with a 0.22 pm cellulose acetate filter and diluted with acetonitrile before the injection. The Masshunter software (Agilent Technologies, Inc.) was used for the acquisition and processing of data. The detection was carried out using the ESI Source in positive mode. The capillary voltage was set at +3500 V. The drying gas temperature was set at 300°C; nitrogen has been used as drying gas (flow 9 l/min) and nebulizing gas (pressure 28 psi). A liquid chromatography of hydrophilic interaction was performed on a Hilic Codecs UPLC column (1.6 pm, 2.1 100 mM). The 1 - Deoxynojirimycin was eluted with a binary gradient consisting of 20 mM of ammonium formiate in water (A) and acetonitrile (B). The gradient was as follows: 0- 10 min, 18% A; 1 1 -16 min, 50% A; 16-27 min, 18% A. The flow was adjusted to 0.3 ml/min and the column temperature was maintained at 25°C. The HPLC analyzes were performed in triplicate. This method had previously been validated for linearity, precision and accuracy (14).

Preparation of Ca-ALG beads

Two different lots of beads (Ca-ALG-2% and Ca-ALG-3%) were obtained starting from 2% and 3% respectively (P/V) of Na-ALG aqueous solution, degassed by ultrasonic bathroom (Bandelin Electronic GmbH, Berlin, Germany). The Ca-ALG beads were prepared by ionic jellification, according to the method proposed by lannuccelli et al. (15) with some changes. To prevent the presence of ethanol in the final product, the beads were loaded with freeze-dried EXT, wherein the organic solvent had been completely removed. The loaded beads (EXT-Ca-ALG Beads-2 and 3%) were obtained by melting 100 mg of dry EXT in 5 ml of Na-ALG solution. To reduce the possible loss of 1 -DNJ during the formulation process, the dry EXT was also dissolved in the reticulating (cross-linking) solution of CaCh 2% (w/v), so that the concentration of 1 -DNJ in the solution with ALG was the same as the solution with the cross-linking agent. The Na-ALG solution was dripped through a silicone tube (2 mm internal diameter) by a height of 6 cm in the CaCh solution, with a slight magnetic stirring. The beads that form instantly were left for 30 minutes in contact with the vehicle under stirring, then were recovered, washed with Milli-Q water, and dried at room temperature for at least 24 h. The unloaded beads (empty) were prepared in the same way, without the addition of the EXT. All the formulations have been prepared in triplicate. To prevent the absorption of humidity, the particles have been kept in a dryer until the time of analysis. The efficiency of performance of the formulation process has been calculated as follows:

Preparation of alginate microparticles via spray-dryer (SDMs)

One hundred ml of ethanolic EXT at 50% of mulberry leaves were mixed under slight magnetic stirring with 100 ml of Na-ALG aqueous solution to 2% (w/v) to obtain a final solution at the concentration of 1% of Na-ALG. The solution was degassed through an ultrasonic bath and then dried by spray-drying (Buchi 190 Mini-Spray Dryer, Buchi Labortechnik, Flawil, Switzerland) with the following operating conditions: inlet temperature, 140°C; outlet temperature, 65-70°C; pump speed, 3 ml/min; nebulization flow, 600 Nlh^-1; aspirator setting, 15; Auger cap diameter, 0.5 mm. The unloaded SDM (empty) were obtained in the same operating conditions starting from a 1% solution of Na-ALG in water. All the formulations have been prepared in triplicate. To reduce the absorption of humidity, the particles were kept in a dryer before the analyzes. The efficiency of performance of the formulation process has been calculated as follows:

The weight of the EXT has been calculated considering the average weight of the dry extraction obtained from the freezing process.

Particle size analysis

The size of the empty and loaded beads was calculated by analyzing the images obtained from a digital camera. The diameter of the beads has been measured with the Imaged software (National Institutes of Health, USA), which allows to transform pixels into length units (mm) measuring a caliber section. The relative frequency distribution of each sample has been calculated and the data were adapted with a Gaussian equation on Graphpad Prism 8.4.3 (Graphpad Software, San Diego, Ca). Since the diameters are normally distributed, the homogeneity of the beads (Polydisperion index, PDI) has been calculated as follows:

™ = (%)²

Where o and d represent the standard deviation and the average diameter respectively.

The size of the SDM particles was determined through a laser diffractometer (Matarsizer Hydro 2000 MU, Malvern Panalytical, Malvern, UK) suspending about 20 mg of SDM in 25 ml of isopropanol under magnetic stirring. The homogeneity was assessed by calculating the Span factor as follows:

Wherein d¹⁰, d⁵⁰ and d⁹⁰ represented the fine, medium and coarse fractions of the particles respectively.

Morphological analysis

Morphological analyzes were performed by ESEM microscopy (Environmental Scanning Electron Microscopy), (Quana 200, Fei, Hillsboro, Oregon, USA). Before the analysis, the Ca-ALG beads and the SDMs were placed in a dryer overnight to remove residual humidity. The beads and the loaded and empty Ca-ALG SDMs have been fixed on an aluminum support using a type of double-side carbon tape and then coated under vacuum with gold-palladium in an argon atmosphere for 1 minute (Sputter Coater Emitech K550, Ethech Ltd., Ashford, Kent, United Kingdom).

Loading of 1-DNJ and Encapsulation efficiency

To determine the amount of 1 -DNJ encapsulated in Ca-ALG beads and SDMs, 10 mg of each formulation were introduced in 1 ml of 0.2 M HCL solution. The strong acid solution was selected to break down the interaction between 1 -DNJ and the polymer. After 24 hours, the suspension was mixed by vortex, diluted with ethanol and analyzed by HPLC-ESI-TQ-MS, as described in point 2.3. The actual loading or drug loading (DL) and the efficiency of encapsulation (EE%) have been calculated with the following equations:

Fourier's transform infrared spectroscopy (FTIR)

The infrared spectra in attenuated total reflectance (ATR-FTIR) of Ca-ALG beads and SDMs (and the corresponding unloaded formulations) were obtained using a Spectrum Two spectrophotometer equipped with a universal ATR sampling accessory (Perkin Elmer, Milan, Italy). The beads were pulverized in a mortar to obtain fine powder. The acquisition spectrum was between 4000-450 cm’¹ with 16 total scans and a resolution of 4 cm’¹.

Relaxometry with proton nuclear magnetic resonance

The interactions between 1 -DNJ and ALG have been studied through relaxometry with proton nuclear magnetic resonance (¹H-NMR). The NMR relaxation measure was widely used for the study of interactions between molecules and their movements. Two relaxation times (T 1 and T1 p) of the protons of 1 -DNJ have been considered for this purpose. After an impulse, the spin-lactice relaxation time (T1 ) is the time necessary for the recovery of magnetization along the direction of the main magnetic field (z axis). Consequently, T1 determines the speed at which a sequence of impulses can be repeated. Instead, the relaxation time Ti p is the time necessary for the decay of magnetization along the radio frequency field of a "spinlocking" impulse. The value of Ti p is measured by first applying a 90° radio frequency impulse to an equilibrium magnetization vector. A second impulse is then applied, which effectively blocks the magnetization vector in the transversal plane (xy). During the spin-locking impulse, the magnetization vector decays to its equilibrium value, with a time constant equal to T1 p. T1 p is known to be more sensitive than T 1 to slow molecular fluctuations, typical of in vivo processes such as chemical exchanges with macromolecules (16-18). To evaluate whether there was a sort of interaction, the Ti p values of 1 -DNJ were measured in the absence and presence of ALG, in a molar ratio 50: 1 .

Subsequently, to better understand the results, the specific interaction constants between the ligand (1 -DNJ) and the polymer (ALG) were calculated at different concentrations, according to the method proposed by Di Cocco et al. (19). In this regard, 1 -DNJ- ALG solutions were prepared by adding 1% (w/v) ALG to 1 -DNJ solutions in D2O at different concentrations (4, 5, 10, 15, 20, 25 mM). The association constants (Ka) were determined following the evolution of 1 -DNJ relaxation times, increasing the concentration of 1 -DNJ and keeping the ALG constant. The NMR experiments were carried out on an FT-NMR AVANCE III HD 600 MHz Bruker spectrometer (Bruker Biospin GmbH Rheinstetten, Karlsruhe, Germany), at 298°K. The chemical displacement values were expressed in ppm compared to the TSP, used as a reference. The relaxation times of Ti p have been measured using the Bruker sequence "t1 rho_esgp2d". The acquisition parameters were as follows: number of spectral points (time domain), 32 k; fake scans, 4; number of scans, 16; amplitude of the pulse, 12.03 ps (90°); acquisition time, 2.50 s; "delay time", 5 s; spectral amplitude, 1 1 ppm (6602 Hz), FID resolution, 0.40 Hz; digitization, baseopt. Total acquisition time: 32 min and 30 s. In total, 14 values of T for calculations were used, between 10 ms and 6 s. The ¹H-NMR experiments for the calculation of T1 were performed by means of the "t1 ir_pr" Brucker sequence, with the pre-saturation of the residual signal of the water. The acquisition parameters were as follows: number of spectral points, 128 k; fake scans, 0; number of scans, 4; amplitude of the pulse, 1 1 .84 ps (90°); acquisition time, 4.96 s; "delay time", 5 s; spectral amplitude, 22 ppm (13204 Hz), FIDA resolution, 0.2 Hz; digitization, baseopt. Total acquisition time: 7 min and 42 s. The spin-lactice relaxation times were measured through an "inversion-recovery pulse" sequence (180°-T-90°) and fitted through an exponential regression analysis of the curves of the recovery times of the longitudinal magnetization of the protons. In total 8 values of T from 100 ps to 20 s were used for calculation.

Study of the swelling of the Ca-ALG beads

The absorption behavior of the water by the Ca-ALG beads has been studied gravimetrically (20). The pre-weighted beads (about 100 mg) were placed in 10 ml of simulated gastric liquid (HCI solution, pH 3) and saline phosphate buffer 0.1 M (PBS, pH 6.8) in sink conditions at 37°C. The beads were taken at fixed time intervals and superficially dried to remove the weakly linked excess water; then carefully weighed and put back in the solutions. All the experiments were carried out in triplicate. The dynamic weight variation (%) of the beads was determined according to the following expression: 100

wherein Ws is the weight of the swollen beads and Wi is the initial weight.

1-DNJ in vitro release

The release of 1 -DNJ from the different formulations has been studied in sink conditions in gastrointestinal simulated fluids. As regards the release of 1 -DNJ from the Ca-ALG beads, about 40 mg were placed in 4 ml of HCI solution (pH 3) under magnetic stirring at 100 rpm and 37°C. After 2 hours the beads were filtered on paper and placed in 4 ml of 0.1 M PBS (pH 6.8), for an additional 3 hours under continuous stirring. As for the SDMs, a 40 mg amount was placed in a dialysis membrane (cut off 12,000/14,000 Da) and immediately wet with 10 ml of HCI solution (pH 3.0) under magnetic stirring (100 rpm) a 37°C. After 2 hours, the dialysis membrane has been moved into 10 ml of PBS (pH 6.8) for a further 3 hours, under stirring at 37°C. For both release studies, the amounts of the solutions were taken at fixed time intervals and the initial volume was restored with fresh simulated fluid. The solutions were suitably diluted with the mobile phase and analyzed using HPLC-ESI-TQ-MS. The experiments were performed in triplicate on several lots.

Statistical analysis

Significant differences within the groups were determined at P <0.05 using the Student test and the analysis of the variance (Anova) followed by Tukey post hoc test. All the analyzes were performed using Graphpad Prism 8.4.3 (Graphpad Software, San Diego, Ca).

RESULTS AND DISCUSSION

In this experimental work, a new approach to the prolonged release of the main active constituent of the mulberry has been studied and verified using ALG particles. 1 -DNJ has attracted a lot of attention to potential use as a functional food or supplement to control postprandial hyperglycemia and type 2 DM in its initial stage. The main objective was to develop a delivery system capable of trapping the compound and delay its release in the intestine, where it could act directly as a glucosidase inhibitor and then reach systemic circulation and perform systemic effects.

The encapsulation of small hydrophilic molecules generally is a great challenge and often leads to poor results in terms of efficiency of trapping and unwanted release profile. Inventors have tried to overcome these problems using ALG, which in addition to being well tolerated and biocompatible can benefit from electrostatic interactions with 1 -DNJ, with consequent better retention of the compound. Two different systems were designed and developed, and their release characteristics have been compared. Millimetric Ca-ALG microparticles (beads) have been selected for their low surface-area-volume ratio, which can reduce the release of small hydrophilic drugs (21 ).

On the contrary, SDM microparticles have been taken into consideration for their ease of production and scalable technology on an industrial level. Spray drying technology is in fact quick, continuous, highly reproducible and allows obtaining a product with low humidity.

As part of this study, the ALG polymer with a high G/M ratio has been selected. In fact, the specific conformation G guarantees a high degree of coordination of the bivalent ions, leading to the formation of a more rigid gel. Therefore, the particles composed of alginates rich in guluronic acid could guarantee the obtaining of more effective systems for a controlled release (22.23).

1-DNJ determination through HPLC-ESI-TQ-MS analysis

Due to the short duration of this analysis and its high sensitivity, the HPLC- MS has proven to be one of the best methods for the determination of 1 - DNJ (10,14,24). The molecular ion (m/z 164.09) and fragments (m/z 146.08, 128.07, 1 10.06) generated by the subsequent loss of water molecules (18 Da), were used for the selective determination of 1 -DNJ, through the multiple reaction monitoring (MRM) method. The quantification of 1 -DNJ in the samples was obtained with the external standard method. The calibration curve was built by analyzing six concentrations of the reference compound in the range 0.005-0.5 pg/ml. In our experimental conditions, the retention time of 1 -DNJ was 4.50 min. Figure 1 shows the MRM chromatogram of mulberry leaf ethanolic EXT and the related ESI-MS spectrum.

The amount of 1 -DNJ in the ethanolic EXT was 15.93 ± 0.19 pg/ml and 1 .93 ± 0.04 mg/g in the freeze-dried product. The yield of the freezing process was 4.50 ± 0.52 mg/ml.

Dimension, morphology and content of 1-DNJ in the particles.

The dimensional parameters of the particles and the content of active compound in the Ca-ALG beads and in the SDMs are shown in Table 1 . The content of 1 -DNJ is expressed as an encapsulation efficiency (EE%) and loading (DL).

Table 1. Dimension, homogeneity, yield, encapsulation efficiency (EE%) and loading of 1 -DNJ (DL) of the Ca-ALG beads and SDMs.

The data are expressed as an average ± standard deviation of three independent experiments. EE, efficiency of encapsulation; DL, drug loading; PDI, polydispersion index.

The Ca-ALG beads had a spherical geometry with a diameter between 1 .10 ± 0.16 and 1 .55 ± 0.15 mm in the dry state. As previously reported, many factors can influence the formation, the size of the particles and the properties of the ALG particles obtained from ionic jellification. The size of the particles is positively correlated with the ALG concentration, the crosslinking time and the diameter of the drip system (25-27). On the contrary, the dimensions are adversely related to the stirring speed and the CaCh concentration, due to the formation of a more rigid gel network and greater electrical conductivity in the presence of a higher concentration of the cross-linking agent (28- 31 ). A good dimensional homogeneity has been observed for the Ca-ALG beads, as indicated by the values of PDI lower than 0.1 , which implies a monodisperse particles population. Both the 2% unloaded beads and those at 3% showed a slightly higher average diameter than the respective formulation loaded with the EXT. However, the homogeneity of the dimensions of the beads has not been significantly influenced by the addition of the EXT, maintaining a narrow distribution for both concentrations.

No significant difference was observed in the size between 2 and 3% Ca- ALG beads, suggesting that the small variation in the concentration of ALG did not result in a significant increase in size. To obtain maximum efficiency in 1 -DNJ encapsulation, the dry EXT was also added to the CaCI₂ solution, in the same proportion compared to the ALG solution (34). However, due to the very high solubility in water, hydrophilic nature and low molecular weight, 1 -DNJ showed a rather low EE (%) in the case of beads. A critical loss of the compound may have been caused by the final washing step to remove the excess of the cross-linking agent (25). In addition, 1 -DNJ may have been partially lost during the drying process, when the water incorporated into the jellified beads was removed and partially absorbed on the filter paper. In the case of Ca-ALG-3% beads, a higher EE% value was found (P <0.05) due to the higher concentration of Na-ALG used in the formulation process.

In fact, the increase in the concentration of ALG provided a greater availability of bond sites for Ca²⁺ ions, leading to the formation of a more compact gel membrane with reduced pores size (28). Therefore, in this condition the loss of 1 -DNJ during the formulation process was moderate (32, 33). At the same time, the highest concentration of ALG led to the significant reduction of DL from 0.52 ± 0.02 to 0.43 ± 0.04 pg/mg, respectively for the Ca-ALG beads-3% and those 2% (P <0.05). The morphological analysis confirmed the sphericality of the beads and highlighted some differences in the surface structure mainly depending on the concentration of the polymer in the unloaded formulations. In fact, the surface of the unloaded particles of Ca-ALG-2% appeared wrinkled and less compact compared to that of the beads of unloaded Ca-ALG-3%. The encapsulation of the EXT therefore seemed to give a more compact surface structure (Figure 2).

As for the SDMs, the yield was rather low, around 60%. In fact, the conventional spray-dryer involves a remarkable loss of particles on the walls of the drying chamber and the cyclone (34). The unloaded SDMs showed a mono-dispersed population with a low span value. On the contrary, the SDMs loaded with the EXT showed a bimodal distribution (table 1 ), with the smaller size (1 1 ± 5 pm) which was the most abundant (about 80%). This evidence can be attributed to a greater tendency of the particles to form agglomerates in the presence of the EXT, due to the natural presence in it of sugars, glycosides and other gluey components (35, 36). As has been previously underlined, microparticles that encapsulate extracts of plants usually tend to establish bridges to connect together by absorbing humidity from the environment. In addition, an agglomeration phenomenon can take place following the collision of particles during drying, which can lead to the coexistence of small and larger particles (36). ESEM analysis also highlighted the non-homogeneity of the SDMs (Figure 3). Some morphological differences have been observed between unloaded SDMs (Figure 3A) and loaded with the EXT (Figure 3B).

The loaded microparticles (EXT-SDM) showed a spherical shape and a smooth surface. On the contrary, some empty SDMs had a collapsed appearance or appeared like hollow spheres. According to the hypothesis proposed by Ameri et al., the formation of a polymer film on the outside of the droplet during the initial drying step caused by the rapid evaporation of the solvent can explain the presence of introflexions in the structure. The further increase in the concentration of the polymer on the surface could then hinder the spread of the water on the periphery of the droplet and cause a storage of steam pressure inside the particle. Finally, the outbreak of the film would cause punched and hollow particles (37). As for the encapsulation of 1 -DNJ in the SDMs, the efficiency was almost complete, suggesting that the ratio between polymer and EXT and the parameters of the spray dryer were optimal (38). The SDM DL was significantly higher (P <0.01 ) than the bead DL. Although the SDMs have exhibited an EE% twice higher (99 ± 3%), the DL was of the same order of magnitude as that of the beads. This fact can be explained considering that the preparation of the SDM has led to the use of the ethanolic EXT, with a lower absolute amount of 1 -DNJ, without any preliminary concentration (obtained by freezing in the case of beads). Therefore, the ratio between 1 -DNJ and ALG in the SDMs was lower than that of the Ca- ALG beads.

Fourier's transform infrared spectroscopy The ATR-FTIR spectra of the mulberry leaves EXT, the Ca-ALG beads, the SDMs and the related unloaded formulations have been acquired (Figure 4). Since no differences have been observed between the Ca- ALG-2% and -3% beads, only the spectra of the Ca-ALG-2% beads have been reported below. The spectrum of the lyophilized EXT showed several large peaks, mainly attributable to functional groups of phenolic acids and flavonoids. In particular, the spectral bands at 3274, 1038 and 990 cm ¹ were representative of the stretching of O-H and C-0 of the alcohol, while the peaks at 1367 and 1272 cm^-1 were induced respectively by bending O-H vibrations of phenols and of secondary O-H in plane. Finally, the signals at 1722 and 1580 cm’¹ were attributable to the C=O carboxylic and carbonylic groups of ketones or esters (39, 40). As for the SDMs and beads of unloaded Ca-ALG, the main signals around 3258, 1604, 1412 and 1028 cm’¹ have been induced by stretching the vibrations of O-H, C=O (symmetrical and asymmetrical) of carboxylic groups e C-O-C bonds respectively (41 ). These spectra were similar with slight shifts of about 20 crrr¹ with lower wavelengths in the case of the Ca-ALG beads for the vibrations of hydroxyls and carboxyls. The shifts were caused by the interaction of these functional groups with calcium ions (41 , 42). The loading of the EXT in both the Ca-ALG and SDM formulations did not change the IR spectra of the polymer, and the typical bands of the EXT were not detected. The same observation was previously reported by Bagheri et al. and confirmed that the encapsulation of the EXT took place successfully (41 ).

Study of the interaction by NMR relaxometry

This study was conducted to analyze the nature of the interaction between 1 -DNJ and ALG. In particular, monitoring and measurement of "spinlattice" relaxation times can provide essential information on the mobility of the ligands in the presence or absence of macromolecules (19, 43, 44). The chemical shifts of ¹H NMR, the multiplicity of the signal and 1 -DNJ assignments are shown below in Figure 5 and Table 2.

Table 2. ¹H NMR assignments, chemical shifts, and multiplicity of the signals of 1 -DNJ in deuterated water.

Table 3. Spin-lattice relaxation time in the Rotating Frame (Tip) of 1 - DNJ protons in the absence and presence of alginate.

As can be seen from Table 3, the presence of ALG has caused a significant decrease in the values of Tip of the protons of 1 -DNJ, especially for H2, H3, H4 and H5. This decrease is indicative of the reduction in the mobility of the ligand (45, 46), suggesting the formation of electrostatic interactions, hydrogen bonds or other weak intermolecular forces between the protons of 1 -DNJ and the carboxylic groups of ALG. Having assumed that the interaction was present, the NMR was further used for the determination of the association constants, according to the method proposed by Di Cocco et al. (19). When a ligand (L) interacts with a macromolecule (M) an equilibrium occurs, and is expressed by the following equation:

If the complex is kept together by weak intermolecular forces, the equilibrium constant is defined as an association constant (K_a) and the product has chemical characteristics that still resemble strongly to nonassociated (or free) molecules. K_a can be obtained as:

Where [L], [M], and [LM] are the molar concentrations of the free ligand, macromolecule and complex, respectively.

The relaxation index (R) is the reciprocal of T1 , and it is the experimental NMR parameter that best explains the dynamic coupling between L and M. The experimentally determined relaxation index of the ligand in the presence of the macromolecule (Robs) can be expressed as:

pf: fraction of free ligand pb: fraction of associated ligand

Rf: relaxation index in the free state

Rb: relaxation index in the associated state

Assuming that the fraction of free ligand is much higher than the associated one (pb «< 1 ), and that pb + pf = 1 : (1)

The K_a association constant can be expressed as:

Wherein [Mo] is the initial concentration of the macromolecule.

Since pb is the fraction of the associated ligand, then:

Finally, replacing (3) in (4) and then in (1 ), the following equation is obtained:

The last equation (5) shows a linear relationship for 1/ R and [L]. In practice, the T1 values have been measured at different concentrations of 1 -DNJ (4-25 mM) in the absence and presence of ALG (1% w/v). So the difference with the reciprocal of T1 has been calculated (R). The value (1/R) was then plotted against the concentrations of 1 -DNJ, and a linear regression was obtained for each proton. For 1/ AR = 0, -1/ Ka was extrapolated (Table 4-5).

Table 4. Spin-Lattice relaxation times (Ti ) of 1 -DSL protons in the absence and presence of ALG.

Table 5. Association constants (K_s) of 1-DNJ pretons, drained from the difference in the values of the spin-lattice relaxation times (T-i) in the absence and presence of ALG.

t is generally accepted that the determinations of the Ka values based on NMR are reliable only if in the interval 10-10⁴ M-¹ (48). The values obtained here were all within the range of acceptability and consistent with those provided by another study that evaluated the interaction between ALG and different amino acids, very similar to 1 - DNJ from a chemical point of view (19). The highest interaction constants were observed for H6 (109 M’¹), H3 (120 M’¹), H2 (141 M’¹) e H4 (151 M’¹). On the basis of these results, we can conclude that a specific interaction involved the above- mentioned protons of 1 -DNJ and the ALG, due to the most significant decrease in the relaxation time among those observed. The data relating to T1 measurements (Table 5) also agree with the values of T1 p experimentally determined and reported in Table 3. In fact, 1 -DNJ is a base (pKa 8.06) which could be electrostatically linked to the carboxylic groups of guluronic and mannuronic acids. This electrostatic interaction can contribute to the stabilization of the particle loaded with EXT and contribute to the achievement of a prolonged release of the active compound, regulated by a dynamic dissociation equilibrium from the polymer.

Study of the swelling of the Ca-ALG beads

Since the beads are formed by an ALG cross-linked matrix, to use them as a form of oral transporting, it is essential to evaluate their swelling in simulated gastric and intestinal fluids. For this reason, the swelling of the Ca-ALG beads was conducted in the same experimental conditions as the in vitro release. During the swelling process, two distinct mechanisms occurred. The first consisted in the hydration of the hydrophilic groups of ALG (47), due to the penetration of water through the surface of the bead and the filling of the pores between the polymer chains. The second mechanism was linked to the exchange between the Ca²⁺ and the Na⁺ ions present in the environment, or the simulated intestinal liquid. The swelling of the beads was strongly dependent on the pH and the ionic force of the solution wherein the particles were placed (Figure 6). All beads placed in simulated gastric liquid moderately increased their weight (about 40-45%), reaching the maximum absorption of water in about 60 min. All formulations showed more significant swelling rates when placed in the simulated intestinal liquid. Here, the beads showed a first absorption of water in the initial 60 minutes, then they began to lose weight due to the progressive external erosion, due to the dissolution of the ALG chains induced by the ionic exchange. The first absorption of water was mainly linked to the exchange between Na⁺ and Ca²⁺ ions linked to external chains of poly-mannuronate (48). When the Ca²⁺ ions are progressively replaced, the polymer chains undergo a relaxation, making the beads swell and absorb water. The second massive absorption of water (which is observed from 90-120 min until the end) prevails over erosion and this time involves the exchange of Na⁺ from the vehicle with Ca²⁺ ions linked to carboxylic groups of the poly- guluronate units distributed inside, as previously described by Bajpai and collaborators (48). In light of the above, the presence of Na⁺ in the environment is responsible for the greater extent of the swelling process and the subsequent degradation of the beads, which on the other hand does not occur during the 120 minutes of exposure to the solution of hydrochloric acid. The results suggest that dried beads swell slightly in the gastric environment and then undergo significant absorption of water in the upper tract of the intestine. This behavior indicates that the beads are stable in the acid medium and are possibly able to free 1 -DNJ in the intestine in a controlled way over time, due to the erosion of the particles. The post hoc Test of Tukey and the Anava One-Way analysis have highlighted significant differences between all samples at 180 minutes in the simulated intestinal liquid (P <0.01 ). In particular, the highest degree of swelling has been observed for the unloaded Ca-ALG beads-3%, indicating that the concentration of the polymer is positively correlated with a greater relaxation of the ALG chains. On the other hand, in the Ca-ALG beads loaded with the EXT the overall swelling has been significantly reduced (P <0.0001 ), supporting the hypothesis that the electrostatic interaction between ALG and 1 -DNJ can stabilize the structure, hardening the polymer chains.

1-DNJ in vitro release

The release of 1 -DNJ from Ca-ALG beads and SDMs has been studied in simulated gastrointestinal fluids (figures 7-8). The pH of aqueous media was selected to simulate the post-prandial conditions of the gastrointestinal tract. In the case of the Ca-ALG beads, a quick initial release (called booster effect) was observed in the first 10 minutes following contact with gastric liquid, reaching 28 ± 4 and 36 ± 6% of the cumulative release for concentration of ALG of 3% and 2%, respectively. The initial release of 1 - DNJ from the Ca-ALG beads-3% was slightly faster than the one from the Ca-ALG beads-2%, although higher concentrations of ALG have generally been considered more effective in delaying the release of drugs (49-51 ). The booster effect is caused by the rapid permeation of the water between the polymer chains, a phenomenon already observed during the swelling studies. Therefore, the explanation lies in the rapid dissolution in the simulated gastric fluid of 1 -DNJ absorbed on the external polymer chains of the beads through electrostatic interaction (26). Subsequently, the release of 1 -DNJ in acid conditions is slow and constant up to 2 h, which is the approximate time of stay in the stomach. Since the swelling of the Ca-ALG beads in acid conditions is quite limited, the constant release of 1 -DNJ for 2h is probably due to a process of diffusion from the insoluble particle matrix in the acid medium (52).

On the contrary, the release of 1 -DNJ in the intestinal fluid seems to be regulated by the complex process of "swelling-dissolution-erosion" (52). In fact, in the intestinal fluid, Ca²⁺ was replaced by Na⁺ present in the environment, and consequently the beads visibly lost their compact shape after 60 minutes, quickly degrading. This evidence is also confirmed by the significant weight loss recorded during swelling studies. Interestingly, even if the progressive disintegration and solubilization of beads has occurred, the amount of 1 -DNJ released has never increased beyond 27 ± 6 and 37 ± 6% for Ca-ALG-2% and CA -ALG-3% beads, respectively. Considering that 1 -DNJ is an extremely soluble compound characterized by a small chemical imino sugar structure, its low concentration in free form in solution could be explained by the strong interaction with the polymer demonstrated through NMR studies.

In light of the evidence observed during the release of 1 -DNJ from the beads, the same experiment was also performed for the SDMs in the same operating conditions. In this case, the release of 1 -DNJ was made using the dialysis membrane to avoid taking microparticles together with the fluid to be submitted to quantitative analysis. By comparing the release profiles of beads and SDMs (Figures 7-8), it is clear that the encapsulation of EXT by means of the ALG cross-linking with calcium ions (beads) did not play a central role in delaying the release of 1 -DNJ.

Normally, the release of Na-ALG drugs depends on the solubility of the drug and the polymer, their interaction, and the concentration of the polymer (53). The main factor involved in the control of the release of 1 -DNJ from the SDM seemed to be the interaction between the polymer and the imino sugar. In addition, even the pH of simulated fluids may have played an important role in controlling the release of 1 -DNJ. In fact, depending on the pH, the sodium alginate (Na-ALG) meets structural changes. In acid conditions (pH <5), carboxylic groups are protonated, and Na-ALG is therefore mostly converted into alginic acid (undissociated form). In its undissociated form, the polymer is shrinked and completely insoluble. So, during the release experiments at pH 3, the SDMs are not altered or degraded at all by the simulated gastric fluid (54). This characteristic of the ALG polymer has meant that after a quick release of 1 -DNJ in the first 30 minutes, the cumulative release was constant and did not exceed 20% in the following 2h. As with Ca-ALG beads, the initial booster effect may have affected 1 -DNJ placed on the surface of the SDMs. These results therefore showed that the electrostatic interaction between 1 -DNJ and ALG is stronger than the formation of salt between 1 -DNJ and HCI present in the simulated gastric environment. For this reason, although 1 -DNJ is highly hydrophilic, its complete release in the gastric environment has not happened. During the subsequent simulation step of the intestinal fluid, further release was observed (Figure 8). The transition of the pH from 3.0 to 6.8 may have induced the rearrangement of the polymer structure followed by the progressive solubilization of the SDMs, leading to a quick release of 1 -DNJ in the first 30 minutes. In fact, in the intestinal environment the alginic acid reacts with the sodium salts of the phosphoric acid of the PBS buffer. In these conditions Na-ALG is soluble in aqueous environment and the particles dissolve. When the equilibrium has been restored again around 150 min, the release of 1 -DNJ reached a new plateau reaching about 43% of the cumulative release.

The incomplete release of 1 -DNJ observed in this study is partially in contrast with the results obtained by Gavini et al., who evaluated the in vitro release of metoclopramide from SDM (55). The authors reached the complete release of the drug rich in basic groups in 3 h (pH 7), even if there was a possible electrostatic interaction between ALG and the drug (55). This difference can be explained considering that metoclopramide is a higher molecular weight drug than 1 -DNJ and that therefore it is more difficult to "get stuck" between the chains of the ALG polymer.

Therefore, the results of the in vitro simulated release suggested that Ca-ALG beads and SDMs could maintain their integrity in the stomach acid environment and be able to control the release in the intestinal environment. The controlled release of 1 -DNJ seems to be mainly ruled by the electrostatic interaction between the active compound and the carboxylic groups of ALG. Therefore, 1 -DNJ encapsulation in structured support such as Ca-ALG beads may not be necessary to achieve the goal of prolonged release. In fact, this study demonstrates that the co-presence of ALG and 1 -DNJ in solution is sufficient to obtain a controlled release over time thanks to the proven electrostatic interaction and supported by all the experiments conducted.

CONCLUSIONS

In the current work, two different strategies for 1 -DNJ encapsulation have been developed and characterized to prolong its release and therefore improve its therapeutic efficacy. Both systems have shown to delay and successfully control the release of 1 -DNJ in the gastrointestinal environment, but the maximum encapsulation efficiency has been reached for SDMs. The in vitro release study underlined that the release is ruled by the strong electrostatic interaction between 1 -DNJ and the carboxylic groups of ALG, demonstrated through the NMR experiments. Between the two tested, 1 -DNJ transporting strategy based on Na-ALG cross-linking with CaCh has shown to be the less convenient option to achieve the goal. In fact, the Ca-ALG beads require a greater amount of EXT in the jellification solution to obtain the same loading than the SDMs. In conclusion, the formulation of the SDMs was the simplest, most promising and scalable strategy to convey 1 -DNJ contained in the extracts of mulberry leaves, even if the product yield must be improved. On the basis of these results, the active compound could be released slowly and gradually from the stomach to the intestine, allowing to persist in situ and act for a prolonged period of time. This phenomenon would guarantee a good probability that an effective dosage of the loaded compound passes into the blood circulation, ensuring high bioavailability and the optimal therapeutic result. This work is of the utmost importance to overcome the problems related to the low bioavailability of 1 -DNJ and other highly soluble drugs in aqueous environments.

Although the invention has been illustrated and described in detail in the figures and in the previous description, this illustration and description are to be considered illustrative or exemplary and non-limiting; the invention is defined by the attached claims and therefore not limited to the embodiments described. Variations to the embodiments described can be understood and carried out by the experts in the field who perform the claimed invention, by a study of the figures, description and attached claims. REFERENCES

1 . World Health Organization, Global Report on Diabetes (2016).

2. Y. P. Naveen, A. Urooj, and K. Byrappa, J Herb Med 28, 2210-8033 (2021 ).

3. K. Thakur etal., Trends Food Sci Technol89, 88-99 (2019).

4. Z. Wang, J. Wang, and P. Chan, Evidence-based Complementary and Alternative Medicine 2013 (2013).

5. Y. G. Li et al., Sci Rep 3, 1377 (2013).

6. L. Milella et al., J Nat Prod 79(8), 2104-2112 (2016).

7. K. Yatsunami, Y. Saito, E. Fukuda, S. Onodera, and K. Oshigane, Food Sci Technol Res 9(4), 392-394 (2003).

8. L. Marchetti et al., Food Funct 3(8), 4344-4359 (2022).

9. K. Gao et al., Molecules 21 (11 ), 1600 (2016).

10. C. Vichasilp et al., Food Chem 134(4), 1823-1830 (2012).

12. T. Yilmaz, L. Maldonado, H. Turasan, and J. Kokini, J Food Eng 254, 42-50 (2019).

13. H. Wang, Y. Shen, L. Zhao, and Y. Ye, Curr Med Chem 28(3), 628-643 (2021 ).

14. L. Marchetti etal., Plants 1Q(8), 1-13 (2021 ).

15. V. lannuccelli, G. Coppi, and R. Cameroni, Int J Pharm 143(2), 195-201 (1996).

16. F. V. C. Kock, T. Monaretto, and L. A. Colnago, Int J Biol Macromol 98, 228-232 (2017).

17. S. D. Serai, Pediatr Radiol 52(2), 217-227 (2022).

18. W. Fieber et al., Macromolecules 40(15), 5372-5378 (2007).

19. M. E. Di Cocco, C. Bianchetti, and F. Chiellini, http://dx.doi.org/10. 1177/08839110303701518(4), 283-296 (2016).

20. G. Pasparakis and N. Bouropoulos, Int J Pharm 323(1-2), 34-42 (2006).

21 . M. Khorram, M. Samimi, A. Samimi, and H. Moghadam, J Appl Polym Sci 132(4), 41334 (2015).

22. T. Ostberg, L. Vesterhus, and C. Graffner, Int J Pharm 97(1-3), 183-193 (1993).

23. F. Acarturk and S. Takka, J Microencapsul 16(3), 291-301 (1999).

24. K. Nakagawa et al., Anal Biochem 404(2), 217-222 (2010).

25. L. E. Kurozawa and M. D. Hubinger, Curr Opin Food Sci 15, 50-55 (2017).

26. Q. Li, X. Li, and C. Zhao, Front Bioeng BiotechnolQ (2020).

27. C. Jeong, S. Kim, C. Lee, S. Cho, and S. B. Kim, Foods 9(2) (2020).

28. N. T. T. Uyen, Z. A. A. Hamid, N. X. T. Tram, and N. Ahmad, Int J Biol Macromol 153, 1035-1046 (2020).

29. S. F. Lin et al., PLoS One 11 (4), e0153685 (2016). A. R. Cho, Y. G. Chun, B. K. Kim, and D. J. Park, J Mater Sci 49(13), 4612-4619 (2014). S. Shukla, D. Jain, and K. Verma, Asian J Pharm 4(4) (2010). K. Essifi et al., Polymer Bulletin 2020 78:1078(10), 5789-5814 (2020). K. Essifi et al., J Chem 2021 (2021 ). A. Sosnik, Int Sch Res Notices 2014, 1-17 (2014). T. Esposito et al., Molecules 25(6), 1273 (2020). S. C. Lourengo, M. Moldao-Martins, and V. D. Alves, Foods 9(6) (2020). M. Ameri and Y. F. Maa, Drying Technology 24(6), 763-768 (2007). C. I. Pinon-Balderrama et al., Processes 8(8), 889 (2020). J. Coates, Encyclopedia of Analytical Chemistry 1 , 1-23 (2006). D. Das, R. Ghosh, and P. Mandal, SN Appl Sci 1 (5) (2019). L. Bagheri, A. Madadlou, M. Yarmand, and M. E. Mousavi, Food Research International 62, 1113-1119 (2014). H. Daemi and M. Barikani, Scientia Iranica 19(6), 2023-2028 (2012). M. Mic, A. Pirnau, C. G. Floare, and M. Bogdan, Int J Biol Macromol 147, 326- 332 (2020). J. M. Jaski et al., Ind Crops Prod 129, 662-672 (2019). L. Fielding, Chemlnform 31 (47), no-no (2010). K. Oku etal., J Am Chem Soc 125(42), 12739-12748 (2003). A. S. Hoffman, Adv Drug Deliv Rev 54(1 ), 3-12 (2002). S. K. Bajpai and S. Sharma, React Funct Polym 59(2), 129-140 (2004). S. Basu, S. Mandal, S. Senthil Kumar, B. Krishnamoorthy, and S. Kumar Basu, Article Brazilian Journal of Pharmaceutical Sciences 46 (2010). M. F. Aldawsari et al., Mar Drugs 19(8), 467 (2021 ). V. Pillay, C. M. Dangor, T. Govender, K. R. Moopanar, and N. Hurbans, Drug Deliv 5(1 ), 25-34 (1998). H. H. Tonnesen and J. Karlsen, Drug Dev Ind Pharm 28(6), 621-630 (2002). N. Hussein et al., Pharm Dev Tech no! 25(3), 290-299 (2020). J. J. Chuang et al., Int J Polym Sci 2017 (2017). E. Gavini, G. Rassu, V. Sanna, M. Cossu, and P. Giunchedi, J Pharm Pharmacol 57(3), 287-294 (2005).

Claims

1 . A controlled release administering system of an active ingredient, said administering system being constituted or essentially consisting of an electrostatic complex between an anionic polymer and an active ingredient in a cationic form, wherein the anionic polymer is about 50% in its dissociated form at gastric pH and at intestinal level is mainly in its dissociated form, and the active ingredient is 1 -Deoxynojirimycin or its salt or a natural extract that contains 1 -Deoxynojirimycin or its salt; the system, being able to release a first amount of the active ingredient at gastric pH and a second amount of the same active ingredient at intestinal pH, wherein the first amount is less than the second amount.

2. The controlled release administering system according to claim 1 wherein the anionic polymer is chosen between polymers with functional groups negatively charged in the form of acids or salts, such as alginate, carrageenin, hyaluronic acids, pectins, carboxymethylcellulose, or lecithin and surfactants.

3. The controlled release administering system according to claims 1 -2 wherein the anionic polymer is alginate, preferably an alginate with a ratio of [3-L-guluronate and a- D-mannuronate units of 70:30

4. The controlled release administering system according to anyone of claims 1 -3 wherein the natural extract is an extract obtained from plants of the genus Morus, preferably from mulberry leaves.

5. The controlled release administering system according to anyone of claims 1 -4 wherein the anionic polymer is an alginate preferably having a molecular weight from 10,000 to 600,000 g/mol and wherein the molar ratio between alginate and 1 -DNJ is 1 : 0.001 -1 : 300.

6. The controlled release administering system according to anyone of claims 1 -5 wherein:

-a first amount of 1 -DNJ is released at a pH between 3-5.

-a second amount of 1 -DNJ is released at a pH between 6.5-8. and wherein the first amount is about 30% and the second amount is released over time at the pH of the intestine, preferably close to pH 7.

7. The controlled release administering system according to anyone of claims 1 -5 which is in liquid form or powder or granules.

8. A composition comprising the controlled release administering system according to anyone of claims 1 -7.

9. The composition according to the previous claim which further comprises one or more pharmaceutically or dietary acceptable excipients.

10. The composition according to anyone of claims 8-9 which is a pharmaceutical composition, a food supplement or a functional element.

11 . The composition according to anyone of claims 8-10 formulated as a substantially aqueous-based solution or suspension, powder, tablet, granulate.

12. The composition according to anyone of claims 8-1 1 which is administered with a dosage of the mulberry extract containing 1 -DNJ of 300 mg per day, titrated at 1 or 2% of 1 -DNJ.

13. The composition according to anyone of claims 8-12 that is administered during or immediately after meals.

14. The composition according to anyone of claims 8-13 that is administered in combination with other drugs such as insulin-based drugs, semaglutide, sulfonylurea, biguanides, repaglinide.

15. The controlled release administering system or the composition according to anyone of claims 1 -14 for use in a method for the treatment or prevention of:

- an imbalance of blood glucose levels in a subject who needs it,

- a glucose metabolism disorder in a subject who needs it,

- hyperglycemia, diabetes, diabetes mellitus and prediabetic stages in a subject who needs it,

- a disorder linked to an anomalous accumulation of body fat such as obesity in a subject who needs it, or to lower blood glucose levels in a subject who needs it.

16. The controlled release administering system or the composition according to anyone of claims 1 -15 wherein said method comprises the administration to said subject of a combination of the anionic polymer and 1 -DNJ, said combination comprising a therapeutically effective amount of 1 -DNJ or a salt or a natural extract containing 1 -DNJ, wherein the combination is formulated to release a therapeutically effective amount of 1 -DNJ mainly at the intestinal level or wherein the combination is formulated to release a therapeutically effective first amount of 1 -DNJ in the stomach and a second therapeutically effective amount of 1 -DNJ mainly at the intestinal level, the first amount being lower than the second amount.