Method for producing fruit and / or vegetable juice

Abstract

The invention relates to a method for improving the filterability of a fruit and / or vegetable preparation, comprising a step of fermenting the fruit and / or vegetable preparation with a lactic acid bacterial strain.

Description

[0001] DESCRIPTION

[0002] TITLE: Process for manufacturing fruit and / or vegetable juice

[0003] Technical field of the invention

[0004] The present invention relates to a process for manufacturing fruit and / or vegetable juice and the fruit and / or vegetable juice obtained by this process.

[0005] State of the art

[0006] Fruit and / or vegetable juices are naturally cloudy due to the presence of more or less marked plant residues in suspension (cellulose, hemicellulose, starch and lipids) and dispersed macromolecules (pectins, proteins, soluble parts of starch, certain polyphenols and their oxidized or condensed derivatives).

[0007] Fruit and / or vegetable juice manufacturing processes typically include a clarification step. The clarification step primarily aims to completely or partially remove the compounds responsible for product turbidity, which plays a significant role in its sensory qualities (color, texture, appearance) and its suitability for further processing (standardized product, storage stability) (C. Bhattacharjee et al., 2017). Clarification also standardizes the pulp content, removing larger elements consisting of debris or cell clumps.

[0008] Juice clarification is followed by a microbiological stabilization step aimed at reducing the microbial load of the juice and extending its shelf life. Currently, heat treatments remain the most widely used methods for stabilizing fruit juices and beverages, particularly high-temperature treatments (> 80 °C) such as pasteurization, sterilization, and canning (L. Petruzzi, D. Campaniello, et al., 2017).

[0009] However, these high temperatures lead to the degradation of heat-sensitive compounds in the juice (anthocyanins, carotenoids, vitamins, bioactive proteins, etc.), resulting in negative impacts on their nutritional and sensory properties (G. Rajauria et al., 2018). Furthermore, these techniques require heating procedures that can be energy-intensive (L. Petruzzi, D. Campaniello, et al., 2017).

[0010] While these conventional techniques have demonstrated their technological and economic feasibility, growing consumer demand for more natural products has driven manufacturers and researchers to develop processes that are more respectful of the product and the environment. Innovative techniques have thus emerged, such as the application of pulsed electric fields (A. Cendres, 2010), high-pressure processes, and membrane processes. These new low-temperature stabilization processes have enabled the emergence of new products such as freshly squeezed juices, guaranteeing the absence of heat treatment (Union Interprofessionnelle des Jus de fruits et Nectar, 2009) (A. Cendres, 2010). Among these alternative techniques, membrane processes have appeared for several decades as one of the most efficient, both for the production of fruit juices and for the valorization of waste inherent in their production.

[0011] Nevertheless, membrane clogging remains a major problem with these techniques, especially when filtering juices with high viscosities, such as tropical fruit juices like mango, papaya, or guava, or cucurbit juices like watermelon or melon. Indeed, membrane clogging increases filtration time and cost, and reduces membrane lifespan.

[0012] To limit the clogging of filtration membranes, enzymatic compositions containing pectinases are used. These enzymatic compositions both clarify the juices and improve their filterability. While these enzymatic treatments are effective and widely used to reduce juice turbidity, they nevertheless have some drawbacks (enzyme recovery, discontinuous and time-consuming processes, regulatory restrictions).

[0013] Therefore, there remains a need for new techniques for processing fruit and / or vegetable juices in order to improve their filterability.

[0014] Summary of the invention

[0015] The inventors discovered that fruit and / or vegetable preparations with a high risk of clogging, such as mango puree, had their filterability improved by a fermentation step with a strain of lactic acid bacteria such as Lactiplantibacillus (Lpb.) plantarum prior to a microfiltration step.

[0016] The fermentation step using a lactic acid bacteria strain modifies the physical properties of sieved, crushed, or pressed fruit and / or vegetable preparations, thus improving their filterability. Fermentation with a lactic acid bacteria strain specifically increases the filtration flow rate, decreases viscosity, reduces particle size, and lowers the concentration of insoluble solids in suspension.

[0017] Furthermore, the fermentation stage using a lactic acid bacteria strain eliminates the need for a clarification step. In addition, it turns out that sterilization, particularly by pasteurization, of the preparation prior to the microfiltration stage is unnecessary. Thus, the fruit and / or vegetable juice production process is faster, consumes less energy, and is therefore more economical.

[0018] The invention therefore relates to a method for improving the filterability of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with at least one strain of lactic acid bacteria, preferably with a strain of lactic acid bacteria, so as to obtain a fermented fruit and / or vegetable preparation.

[0019] According to different embodiments of the filterability improvement process according to the invention, which may be taken together or separately: - the fruit and / or vegetable is chosen from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture of these;

[0020] - the fermented fruit and / or vegetable preparation has a filtration flux at a volume reduction factor of 2 (FRV2) that is at least 20% higher than the FRV2 filtration flux of the fruit and / or vegetable preparation before fermentation;

[0021] - the preparation of fruits and / or vegetables is fruits and / or vegetables that have been sieved, crushed and / or pressed, preferably sieved and / or crushed, more preferably crushed;

[0022] - the process does not include a sterilization, pasteurization and / or clarification step before the fermentation step;

[0023] - the process does not include a sterilization, pasteurization and / or enzymatic or physical clarification step;

[0024] - the process includes a step of inoculating the fruit and / or vegetable preparation with only one or more strains of lactic acid bacteria, more preferably with a single strain of lactic acid bacteria;

[0025] - the process includes a step of inoculating the fruit and / or vegetable preparation with at least one strain of lactic acid bacteria, preferably a lactic acid bacteria strain, at an inoculation rate greater than 10 5 UFC / ml of fruit and / or vegetable preparation;

[0026] - at least one strain of lactic acid bacteria is a strain of genus chosen from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc, preferably at least one strain of lactic acid bacteria is chosen from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum, more preferably at least one strain of lactic acid bacteria is a strain chosen from the group consisting of Weissella cibaria, Leuconostoc pseudomesenteroides and Lactiplantibacillus plantarum; even more preferably, at least one strain of lactic acid bacteria is a Lactiplantibacillus plantarum strain;

[0027] - the fermentation of the fruit and / or vegetable preparation is for less than 72 hours, preferably between 5 and 72 hours, more preferably between 8 and 48 hours, even more preferably between 24 and 48 hours.

[0028] One aspect of the invention also relates to a method of filtering a fruit and / or vegetable preparation, characterized in that it comprises improving the filterability of the fruit and / or vegetable preparation according to the method of improving the filterability according to the invention and a step of filtering the fermented fruit and / or vegetable preparation to a cut-off threshold of less than 22 µm.

[0029] Another aspect of the invention is the use of at least one strain of lactic acid bacteria, preferably a strain of lactic acid bacteria, to improve the filterability of a fruit and / or vegetable preparation.

[0030] According to different embodiments of the use according to the invention, which may be taken together or separately: - the use includes the steps of inoculating a fruit and / or vegetable preparation with a strain of lactic acid bacteria and fermenting the fruit and / or vegetable preparation, so as to obtain a fermented fruit and / or vegetable preparation;

[0031] - the fruit and / or vegetable is chosen from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture of these;

[0032] - the fermented fruit and / or vegetable preparation has a filtration flux at a volume reduction factor of 2 (FRV2) that is at least 20% higher than the FRV2 filtration flux of the fruit and / or vegetable preparation before fermentation;

[0033] - the preparation of fruits and / or vegetables is fruits and / or vegetables that have been sieved, crushed and / or pressed, preferably sieved and / or crushed, more preferably crushed;

[0034] - at least one strain of bacteria is at least 10 5 UFC / ml of fruit and / or vegetable preparation;

[0035] - at least one strain of lactic acid bacteria is a strain of genus chosen from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc, preferably at least one strain of lactic acid bacteria is chosen from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum, more preferably at least one strain of lactic acid bacteria is a strain chosen from the group consisting of Weissella cibaria, Leuconostoc pseudomesenteroides and Lactiplantibacillus plantarum; even more preferably, at least one strain of lactic acid bacteria is a Lactiplantibacillus plantarum strain;

[0036] - the use does not include a sterilization, pasteurization and / or clarification step before the fermentation step;

[0037] - the use does not include any sterilization, pasteurization and / or enzymatic or physical clarification steps;

[0038] - the inoculation stage has an inoculation rate greater than 10 5 CFU / ml of fruit and / or vegetable preparation

[0039] - the inoculation step is only with one or more strains of lactic acid bacteria, more preferably with a single strain of lactic acid bacteria;

[0040] - the fermentation of the fruit and / or vegetable preparation is for less than 72 hours, preferably between 5 and 72 hours, more preferably between 8 and 48 hours, even more preferably between 24 and 48 hours.

[0041] The invention also relates to a method for manufacturing fruit and / or vegetable juice characterized in that it comprises the steps of:

[0042] - fermentation of a fruit and / or vegetable preparation with at least one strain of lactic acid bacteria, preferably a strain of lactic acid bacteria, for less than 72 hours in order to obtain a fermented fruit or vegetable preparation and

[0043] - Filtration of the fermented fruit and / or vegetable preparation to a cut-off point below 22 µm. According to various embodiments of the fruit and / or vegetable juice manufacturing process according to the invention, which may be considered together or separately:

[0044] - the fruit and / or vegetable is chosen from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture of these;

[0045] - the preparation of fruits and / or vegetables is fruits and / or vegetables that have been sieved, crushed and / or pressed, preferably sieved and / or crushed, more preferably crushed;

[0046] - at least one strain of lactic acid bacteria is a strain of genus chosen from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc, preferably at least one strain of lactic acid bacteria is chosen from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum, more preferably at least one strain of lactic acid bacteria is a strain chosen from the group consisting of Weissella cibaria, Leuconostoc pseudomesenteroides and Lactiplantibacillus plantarum, even more preferably at least one strain of lactic acid bacteria is a Lactiplantibacillus plantarum strain ;

[0047] - the process does not include a sterilization, pasteurization and / or clarification step before the fermentation step;

[0048] - the process does not include a sterilization, pasteurization and / or enzymatic or physical clarification step;

[0049] - the process includes a step of inoculating the fruit and / or vegetable preparation with at least one strain of lactic acid bacteria at an inoculation rate greater than 10 5 CFU / ml of fruit and / or vegetable preparation

[0050] - the inoculation step is only with one or more strains of lactic acid bacteria, more preferably with a single strain of lactic acid bacteria;

[0051] - the fermentation of the fruit and / or vegetable preparation is between 5h and 72h, more preferably between 8h and 48h, even more preferably between 24h and 48h.

[0052] Another object of the invention is a fruit and / or vegetable preparation fermented by at least one strain of lactic acid bacteria, preferably a strain of lactic acid bacteria, characterized in that it has a filtration flux at FRV2 greater than 10h^.m' 2 and / or a viscosity at a shear gradient of 500 s 1 less than 0.125 Pa.s and / or a median diameter D50 less than 110 pm and / or a concentration of insoluble solids in suspension less than 25g / 100g.

[0053] According to different embodiments of the fermented fruit and / or vegetable preparation according to the invention, which may be taken together or separately:

[0054] - the fruit and / or vegetable is chosen from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture of these;

[0055] - at least one strain of lactic acid bacteria is a strain of genus chosen from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc, preferably at least one strain of lactic acid bacteria is chosen from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum, more preferably the strain of lactic acid bacteria is a strain chosen from the group consisting of Weissella cibaria, Leuconostoc pseudomesenteroides and Lactiplantibacillus plantarum, even more preferably at least one strain of lactic acid bacteria is a Lactiplantibacillus plantarum strain;

[0056] - the fermented fruit and / or vegetable preparation includes at least 10 5UFC of the said at least one strain of lactic acid bacteria / ml of fruit and / or vegetable preparation,

[0057] - the fermented fruit and / or vegetable preparation includes only one or more strains of lactic acid bacteria, more preferably a single strain of lactic acid bacteria.

[0058] Another aspect of the invention is a strain of Lpb. plantarum deposited at the National Collection of Microorganism Cultures (CNCM) 25-28 rue du Docteur Roux, Paris, France under the Treaty of Budapest on November 8, 2024 under the reference CNCM I-6139.

[0059] The present invention also relates to:

[0060] - a composition including the CNCM 1-6139 strain, preferably at least 10 5 CFU / ml of said strain; the composition being preferably a fermented product, more preferably a fermented fruit and / or vegetable preparation

[0061] - the use of said strain to ferment a preparation of fruits and / or vegetables; and / or

[0062] - the use of said strain to increase the filterability of a fruit and / or vegetable preparation.

[0063] Another aspect of the invention is a strain of Weissella Cibaria deposited at the

[0064] National Collection of Microorganism Cultures (CNCM) 25-28 rue du Docteur Roux, Paris, France under the Treaty of Budapest on December 10, 2025 under the reference CNCM 1-6287.

[0065] The present invention also relates to:

[0066] - a composition including the CNCM 1-6287 strain, preferably at least 10 5 UFC / ml of said strain; the composition being preferably a fermented product, more preferably a fermented fruit and / or vegetable preparation;

[0067] - the use of said strain to ferment a preparation of fruits and / or vegetables; and / or

[0068] - the use of said strain to increase the filterability of a fruit and / or vegetable preparation.

[0069] Another aspect of the invention is a strain of Leuconostoc Pseudomesenteroides deposited at the National Collection of Microorganism Cultures (CNCM) 25-28 rue du Docteur Roux, Paris, France under the Treaty of Budapest on December 10, 2025 under the reference CNCM 1-6288.

[0070] The present invention also relates to: - a composition comprising the CNCM 1-6288 strain, preferably at least 10 5 UFC / ml of said strain; the composition being preferably a fermented product, more preferably a fermented fruit and / or vegetable preparation;

[0071] - the use of said strain to ferment a preparation of fruits and / or vegetables; and / or

[0072] - the use of said strain to increase the filterability of a fruit and / or vegetable preparation.

[0073] Brief description of the figures

[0074] [Fig 1] represents a filtration flow of diluted 50 / 50 mango puree (puree / water; w / w) at the cutoff threshold 11pm. M3G corresponds to unfermented mango puree (range 3). M3G LF corresponds to fermented mango puree (range 3);

[0075] [Fig 2] is a histogram representing the values ​​of the filtrate fluxes (J in Lh _1 .m -2) to a volume reduction factor of 2 (FRV2) depending on the ranges, pretreatments, dilution, and filtration cutoff threshold. M1G corresponds to range 1 mango, M3G to range 3 mango, M1G LF to fermented range 1 mango, and M3G LF to fermented range 3 mango. 50 / 50 corresponds to a 50 / 50 dilution (puree / water; w / w), and 60 / 40 to a 60 / 40 dilution (puree / water; w / w). The different letters above the histograms represent the groups obtained by the ANOVA analysis and reflect significantly different values ​​(p-value < 0.05).

[0076] [Fig 3] represents the particle size distribution of control mango purees, enzymatically treated and fermented for the Cogshall variety.

[0077] [Fig 4] represents the median diameter (D50, in pm) of the granulometric distributions (particle size) of the control, enzymatically treated and fermented mango purees for the Cogshall and José varieties.

[0078] [Fig 5] represents the concentration of insoluble solids in suspension (SIS, in g / 100g) of control, enzymatically treated and fermented mango purees for the Cogshall and José varieties.

[0079] [Fig 6] represents the evolution of viscosity (p, Pa.s) as a function of the shear gradient (y, s 1 ) control mango purees, enzymatically treated and fermented for the Cogshall variety.

[0080] [Fig 7] is a histogram representing the viscosity at a shear gradient of 500 s-1 (p, in Pa.s) of control, enzymatically treated and fermented mango purees for the Cogshall and José varieties.

[0081] [Fig 8] represents the evolution of the elastic (G', Pa) and viscous (G", in Pa) moduli as a function of the deformation (y, %) of control mango purees, enzymatically treated and fermented for the Cogshall variety.

[0082] [Fig 9] is a histogram representing the elastic modulus in the linear viscoelastic region (LVE G', in Pa) of control, enzymatically treated, and fermented mango purees of the Cogshall and José varieties. [Fig 10] is a histogram representing the filtration flux at FRV 2 (J, in Lh^.m'). 2 ) control mango purees, enzymatically treated and fermented for the Cogshall and José varieties, during filtration on an Amicon® cell, laboratory scale, PES membrane, pore size 0.1pm, AP = 2 bars, shear gradient = 250 s 1 , dilution 60 mango / 40 water (w / w)).

[0083] [Fig 11] represents the resistance to filtration (R, in m 1) as a function of time (t, in s) in a tangential configuration of control mango purees, enzymatically treated and fermented for the Cogshall and José varieties. (Pilot-scale filtration, ceramic tubular membrane, pore size 0.1pm, AP = 2 bars, shear gradient = 3700 s 1 , dilution 60 mango / 40 water (w / w)).

[0084] [Fig 12] represents the resistance to filtration (R, m 1 ) as a function of time (t, s) in a quasi-frontal configuration of control mango purees, enzymatically treated and fermented for the Cogshall and José varieties. (Pilot-scale filtration, immersed hollow fiber membranes, pore size 0.1pm, mean J = 4.5 Lh^.m') 2 shear gradient = 47 s 1 , dilution 60 mango / 40 water (w / w)).

[0085] [Fig 13] represents the evolution of the mass of filtrate (in g) as a function of time (in s) of control mango puree, fermented with Weissella Cibaria (strain 21) and fermented with Leuconostoc Pseudomesenteroides (strain 56), diluted 60 / 40 (puree / water; w / w) and filtered at the cutoff threshold 0.7pm.

[0086] Detailed description of the invention

[0087] Definitions

[0088] Unless otherwise specified, all technical and scientific terms have and should have the same meaning as that commonly understood by a person skilled in the art in the field to which this invention belongs.

[0089] As used here, the term "UFC" or "UFC" is an abbreviation for "colony forming unit".

[0090] For the purposes of this application, the terms "greater than" or "less than" also include the starting point of the open range. For example, an increase greater than 20% means any increase equal to or greater than 20%.

[0091] As used here, the term "CNCM I-" followed by a 4-digit number designates a strain deposited at the National Collection of Microorganism Cultures (CNCM), 25-28 rue du Docteur Roux, Paris, France, under the Budapest Treaty, with a serial number corresponding to said 4-digit number, e.g., CNCM I-....

[0092] As used here, the reference to a bacterial strain or species should be understood to include functionally equivalent bacteria derived from it, such as, but not limited to, mutants, variants, or genetically transformed bacteria. These mutants or genetically transformed strains may be strains in which one or more endogenous genes of the parental strain have been mutated, for example, to modify some of their metabolic properties (e.g., their ability to ferment sugars, their resistance to acidity, their survival during transport in the gastrointestinal tract, their post-acidification properties, or their production of metabolites).These strains can also result from the genetic transformation of the parent strain to add one or more genes of interest, for example, to confer additional physiological characteristics to the genetically transformed strains, or to enable them to express proteins of therapeutic or prophylactic interest that one wishes to administer through said strains. These mutants or genetically transformed strains can be obtained from the parent strain using conventional techniques of random or targeted mutagenesis and genetic transformation of bacteria, or using the technique known as "genome shuffling."

[0093] In this text, strains, mutants, and variants derived from a parental species or strain shall be considered as encompassed by reference to said parental species or strain; e.g., the expressions "Lactiplantibacillus plantarum" and "CNCM I-6139" shall be considered as including the strains, mutants, and variants derived from them. Accordingly, as used here, a reference to a bacterial strain specified by a serial number shall be considered as encompassing variants of that strain having at least 95% identity (see: Stackebrandt & Goebel, 1994, Int. J. Syst. Bacteriol. 44:846-849). In a particularly preferred embodiment, said variant has at least 97% identity with the 16S rRNA sequence of said specified strain, more preferably at least 98% identity, more preferably at least 99% identity, or higher.

[0094] Improved filterability

[0095] The fermentation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria makes it possible to obtain a fermented fruit and / or vegetable preparation with improved filterability compared to the fruit and / or vegetable preparation before fermentation, also referred to here as unfermented fruit and / or vegetable preparation.

[0096] The filterability of fruit and / or vegetable preparations can be assessed by studying at least one of the following parameters: filtration flux, viscosity, particle size and / or concentration of insoluble solids in suspension.

[0097] The filtration flux, also called permeate flux or filtrate flux, corresponds to the flow rate of permeate passing through the filtration membrane per unit area. It can be determined as shown in the examples. Thus, the filtration flux for a given membrane (for example, with a cutoff threshold of 0.1 µm, 6 µm, 11 µm, or 20 µm) can be determined at a constant transmembrane pressure (for example, 2 bar) and a shear gradient (for example, a shear gradient of 250 s⁻¹). The filtration flux decreases over time due to membrane fouling. Therefore, to ensure its comparability, it is determined at a volume reduction factor of 2 (VRF 2), that is, at the point where 50% of the initial volume has been filtered.

[0098] Thus, according to one embodiment, the improvement in filterability corresponds to a significant increase in the FRV2 filtration flux of the fermented fruit and / or vegetable preparation compared to the fruit and / or vegetable preparation before fermentation. This increase in the FRV2 filtration flux can be greater than 20%, 30%, 50%, 75%, 100%, or 125%.

[0099] According to one embodiment, the fermented fruit and / or vegetable preparation has a filtration flow at a volume reduction factor of 2 that is at least 20% higher than the FRV2 filtration flow of the unfermented fruit and / or vegetable preparation.

[0100] An alternative to filtration flux, for example when filtration is carried out with hollow fibers, can be filtration resistance.

[0101] The filtration resistance can be determined as shown in the examples with a membrane having a cutoff threshold of 0.1 lpm, a transmembrane pressure of 2 bar and a shear gradient of 3700 s⁻¹ 1 in tangential mode and 47s 1 in frontal mode.

[0102] The resistance to filtration is preferably determined at FRV 2.

[0103] According to one embodiment, the improvement in filterability corresponds to a significant reduction in the FRV2 filtration resistance of the fermented fruit and / or vegetable preparation compared to the fruit and / or vegetable preparation before fermentation. This reduction in FRV2 filtration resistance may be greater than 20%, 30%, 50%, 75%, 100%, or 125%.

[0104] Reducing viscosity facilitates the flow of the preparation and is therefore a parameter that improves filterability. The viscosity of the fruit and / or vegetable preparation can be determined as shown in the examples, for example with a shear rate between 0.1 and 500 s⁻¹. 1 , more preferably at 500 s 1 .

[0105] Thus, according to one embodiment, improved filterability corresponds to a significant decrease in the viscosity of the fermented fruit and / or vegetable preparation compared to the fruit and / or vegetable preparation before fermentation. This decrease in viscosity can be greater than 10%, 30%, 50%, or 75%.

[0106] Reducing the size of suspended particles is also a parameter that improves filterability.

[0107] To study the particle size distribution of the preparation, the median diameter D50, which corresponds to the particle size such that half the volume of the particles in the preparation is larger than D50, is chosen as the reference value. D50 can be measured using a laser diffraction particle size analyzer, as shown in the examples. The device measures the diffraction pattern generated by the passage of particles suspended in a fluid through a laser and, through a mathematical transformation based on Mie diffraction theory, obtains the particle size distribution and the volumetric concentration of the particles in question in the fluid. Preferably, D50 is determined with an obscuration index between 20% and 30% and / or a stirring speed of 1500 RPM.

[0108] Thus, according to one embodiment, the improvement in filterability corresponds to a significant decrease in the Δ50 of the fermented fruit and / or vegetable preparation compared to the fruit and / or vegetable preparation before fermentation. This decrease in Δ50 can be greater than 10%, 15%, 25%, or 30%.

[0109] In addition to particle size, the concentration of insoluble suspended solids (SIS) can also be studied. SIS are particles larger than the membrane cutoff threshold commonly used in food microfiltration and are therefore considered the main clogging compounds. These particles originate from the solid parts of the fruit or vegetable. Larger than 1 µm, these particles are the elements likely to deposit on the microfiltration membrane, forming a clogging deposit that impedes the passage of the filtrate through the membrane.

[0110] As detailed in the examples, the fraction of insoluble solids in suspension is isolated by centrifugation at 18,000 g for a time calculated according to Stokes' Law to adapt the centrifugation schedule to the viscosity of the preparation in question. The supernatant is removed. The pellet is weighed. The concentration of insoluble solids in suspension (in g / 100 g of preparation) is thus determined.

[0111] Thus, according to one embodiment, improved filterability corresponds to a significant reduction in the fraction of insoluble solids in suspension in the fermented fruit and / or vegetable preparation compared to the fruit and / or vegetable preparation before fermentation. The reduction in D50 can be greater than 10%, 15%, 25%, 30%, or 40%.

[0112] The improvement in filterability can be determined by one or more of the parameters chosen from the group consisting of increasing the filtration flux, decreasing the resistance to filtration, decreasing the viscosity, decreasing the median diameter D50 and decreasing the concentration of insoluble solids in suspension.

[0113] According to a preferred embodiment, the invention relates to a method for increasing the filtration rate, preferably to FRV2, of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with a strain of lactic acid bacteria, so as to obtain a fermented fruit and / or vegetable preparation. The invention also relates to the use of a strain of lactic acid bacteria to increase the filtration rate, preferably to FRV2, of a fruit and / or vegetable preparation characterized in that it comprises the steps of:

[0114] - inoculation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria and - fermentation of the fruit and / or vegetable preparation.

[0115] According to one embodiment, the invention relates to a method for reducing the filtration resistance, preferably to FRV2, of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with a lactic acid bacteria strain, so as to obtain a fermented fruit and / or vegetable preparation. The invention also relates to the use of a lactic acid bacteria strain to reduce the filtration resistance, preferably to FRV2, of a fruit and / or vegetable preparation characterized in that it comprises the steps of:

[0116] - inoculation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria and

[0117] - fermentation of fruit and / or vegetable preparations.

[0118] According to one embodiment, the invention relates to a method for reducing the viscosity of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with a strain of lactic acid bacteria, so as to obtain a fermented fruit and / or vegetable preparation. The invention also relates to the use of a strain of lactic acid bacteria to reduce the viscosity of a fruit and / or vegetable preparation, characterized in that it comprises the steps of:

[0119] - inoculation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria and

[0120] - fermentation of fruit and / or vegetable preparations.

[0121] According to one embodiment, the invention relates to a method for reducing the median diameter D50 of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with a lactic acid bacteria strain, so as to obtain a fermented fruit and / or vegetable preparation. The invention also relates to the use of a lactic acid bacteria strain to reduce the median diameter D50 of a fruit and / or vegetable preparation characterized in that it comprises the steps of:

[0122] - inoculation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria and

[0123] - fermentation of fruit and / or vegetable preparations.

[0124] According to one embodiment, the invention relates to a method for reducing the concentration of insoluble solids in suspension of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting the fruit and / or vegetable preparation with a strain of lactic acid bacteria, so as to obtain a fermented fruit and / or vegetable preparation. The invention also relates to the use of a strain of lactic acid bacteria to reduce the concentration of insoluble solids in suspension of a fruit and / or vegetable preparation, characterized in that it comprises the steps of:

[0125] - inoculation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria and

[0126] - Fermentation of the fruit and / or vegetable preparation. The nature and properties of the fruit and / or vegetable preparation, the conditions under which the fermentation and filtration steps are carried out, as well as any preceding or following steps, are described in more detail below. They relate to the process for preparing fruit and / or vegetable juice, but also to other processes, uses, and products according to the invention.

[0127] Fruit and / or vegetable juice production

[0128] The process for preparing fruit and / or vegetable juice according to the invention comprises the steps of:

[0129] - fermentation of a fruit and / or vegetable preparation with a strain of lactic acid bacteria for less than 72 hours in order to obtain a fermented fruit and / or vegetable preparation and

[0130] - filtration of the fermented fruit and / or vegetable preparation with a filter having a cut-off threshold of less than 22pm.

[0131] Fruit and / or vegetable preparation

[0132] Advantageously, the fruit and / or vegetable is chosen from the group consisting of tropical fruits, fruits of the Rosaceae family, citrus fruits, and fruits of the Cucurbitaceae family, leafy vegetables, or a mixture thereof. Tropical fruits may include mango, pineapple, papaya, or guava. Fruits of the Rosaceae family may include peach, apricot, apple, or plum. Fruits of the Cucurbitaceae family may include melon or watermelon. Leafy vegetables may include cabbage.

[0133] Also, the fruit and / or vegetable may be chosen from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate, and nightshade, or a mixture thereof. Preferably, the fruit and / or vegetable is a fruit chosen from the group consisting of mango, pineapple, papaya, guava, apple, or a mixture thereof; more preferably, from the group consisting of mango, pineapple, papaya, and guava; most preferably, the fruit is mango, for example, a José or Cogshall mango.

[0134] According to one embodiment, the fruit and / or vegetable is sieved, crushed and / or pressed and optionally diluted before the fermentation step in order to obtain a fruit and / or vegetable preparation.

[0135] Preferably, the fruit and / or vegetable preparation is a fruit and / or vegetable purée. According to this preferred method, the fruit(s) and / or vegetables may be sieved or ground.

[0136] Fruit and / or vegetable preparation can be simple, i.e., obtained from a single type of fruit, or mixed, i.e., obtained by mixing two or more different types of fruit and / or vegetables.

[0137] Dilution can be carried out at a preparation / water mass ratio of between 2 and 0.4, preferably between 1.5 and 1. Depending on the embodiment, the preparation of fruit and / or vegetables (unfermented) before or after the dilution step has:

[0138] - a filtration rate at FRV2 strictly greater than 3 Lh 1 .m -2 5 Lh 1 .m -2 - 7.5 Lh _1 .m -2 ,

[0139] 10 Lh 1 .m -2 , 12.5 Lh1 .m -2 or 15 Lh 1 .m -2 for a 0.1 lpm membrane; and / or

[0140] - a viscosity at a shear gradient of 500 s 1 strictly greater than 0.075

[0141] Pa.s, 0.1 Pa.s, 0.125 Pa.s or 0.15 Pa.s; and / or

[0142] - a median diameter D50 strictly greater than 50 pm, 75 pm, 100 pm, 110 pm, or 125 pm; and / or

[0143] - a concentration of insoluble solids in suspension strictly greater than 0.2g / 100g, 1g / 100g, 10g / 100g, 20g / 100g, 25g / 100g or 30g / 100g.

[0144] According to one embodiment, the process for manufacturing fruit and / or vegetable juice according to the invention does not include a clarification step prior to the fermentation step.

[0145] According to one embodiment, the process for manufacturing fruit and / or vegetable juice according to the invention does not include a sterilization step prior to the fermentation step.

[0146] According to one embodiment, the process for manufacturing fruit and / or vegetable juice according to the invention does not include a pasteurization step prior to the fermentation step.

[0147] According to one embodiment, the process for manufacturing fruit and / or vegetable juice according to the invention does not include a pasteurization step prior to the fermentation step.

[0148] According to one embodiment, the process for manufacturing fruit and / or vegetable juice according to the invention, the process does not include a sterilization and / or pasteurization and / or enzymatic clarification (for example by pectinase) or physical clarification (for example by ultrasound).

[0149] According to one embodiment, the fruit and / or vegetable preparation does not include any component / ingredient other than the fruit and / or vegetable(s). In particular, preferably, it does not include any ingredient selected from the group consisting of added sugar, syrup, an acidifying agent, an enzyme, and a dairy product.

[0150] Fermentation

[0151] The strain used for fermentation is at least a strain of lactic acid bacteria, preferably a strain of lactic acid bacteria.

[0152] According to one embodiment, the lactic acid bacteria strain is a strain of genus selected from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc, preferably of genus selected from Lactiplantibacillus, Weissella and Leuconostoc.

[0153] According to a preferred embodiment, the lactic acid bacteria strain is selected from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum, preferably from Lactiplantibacillus plantarum, Lactiplantibacillus lactis, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, Leuconostoc pseudomesenteroides, and Leuconostoc citreum, more preferably selected from the group consisting of Lactiplantibacillus plantarum, Weissella cibaria, and Leuconostoc pseudomesenteroides, even more preferably in a strain of Lactiplantibacillus (Lpb.) plantarum.

[0154] According to the preferred embodiment of the invention, the Lpb. Plantarum strain is the Lpb. Plantarum strain which was deposited on November 8, 2024 at the National Collection of Microorganism Cultures (CNCM) (Institut Pasteur, 25-28 Rue du Docteur Roux, Paris, France) in accordance with the Budapest treaty under the reference CNCM 1-6139.

[0155] According to the embodiment of the invention, the lactic acid bacteria strain is the Weissella cibaria strain which was deposited on December 10, 2025 at the National Collection of Microorganism Cultures (CNCM) (Institut Pasteur, 25-28 Rue du Docteur Roux, Paris, France) in accordance with the Budapest treaty under the reference CNCM 1-6287.

[0156] According to the embodiment of the invention, the lactic acid bacteria strain is the Leuconostoc pseudomesenteroides strain which was deposited on December 10, 2025 at the National Collection of Microorganism Cultures (CNCM) (Institut Pasteur, 25-28 Rue du Docteur Roux, Paris, France) in accordance with the Budapest treaty under the reference CNCM 1-6288.

[0157] According to one embodiment of the invention, the lactic acid bacteria strain is selected from the group consisting of Lpb. plantarum CNCM 1-6139, W. cibaria CNCM 1-6287, and Ln. pseudomesenteroides CNCM 1-6288. In one embodiment, the methods or use according to the invention comprise a step of inoculating the fruit and / or vegetable preparation with a lactic acid bacteria strain at an inoculation rate greater than 10 5 greater than 10 6 or greater than 10 7 UFC / ml of fruit and / or vegetable preparation.

[0158] Preferably, the inoculation rate is between 10 5 - 10 8 UFC / ml of fruit and / or vegetable preparation, preferably between 10 e - 10 7 CFU / ml.

[0159] According to a preferred embodiment, the fruit and / or vegetable preparation is inoculated with only one or more strains of lactic acid bacteria. According to a more preferred embodiment, the fruit and / or vegetable preparation is inoculated with a single strain of lactic acid bacteria, even more preferably only with a strain of W. cibaria, L. pseudomesenteroides, or Lpb. plantarum, even more preferably only with a strain of Lpb. plantarum.

[0160] Advantageously, the fermentation step is carried out at a temperature between 15°C and 40°C, preferably between 30°C and 40°C, more preferably between 30°C and 37°C. According to one embodiment, the fermentation step is carried out for less than 72 hours, preferably less than 60 hours, more preferably less than 48 hours.

[0161] According to one embodiment, the fermentation step is carried out between 5h and 72h, preferably between 8h and 48h, more preferably between 24h and 48h.

[0162] In one embodiment, the fermentation step is carried out until a pH of 5, 4.7, 4.5, 4.3, 4.2, 4, or preferably 3.8 or less is reached. In another embodiment, the fermentation step is carried out until a pH between approximately 4 and approximately 3.5, and preferably between approximately 3.8 and approximately 3.5, is reached. The pH can be adjusted by controlling the fermentation by the microorganism and stopping it if necessary, for example, by cooling.

[0163] In one embodiment, the fermentation step is carried out until a titratable acidity greater than 0.9 is reached. In another embodiment, the fermentation step is carried out until a titratable acidity of between approximately 0.7 and approximately 1.25, and preferably between approximately 1 and approximately 1.1, is reached.

[0164] Fermented fruit and / or vegetable preparation

[0165] The present invention also relates to a fermented fruit and / or vegetable preparation.

[0166] The present invention also relates to a fermented fruit and / or vegetable preparation obtained by the manufacturing process according to the invention.

[0167] Typically, fruit and / or vegetable preparation involves fruits and / or vegetables that have been sifted, crushed and / or pressed.

[0168] Preferably, the fruit and / or vegetable preparation is a fruit and / or vegetable purée. According to this preferred method, the fruit and / or vegetables may be sieved or ground.

[0169] According to one embodiment, the fermented fruit and / or vegetable preparation has a filtration flux at FRV2 greater than 10.h Tnr 2 , 12, 5L.lv 1 . nr 2 , 15L.lv 1 . nr 2 or 20L.h 1 .nv 2 for a cutoff threshold of 0.lpm.

[0170] Another object of the invention is a preparation of fruit and / or vegetables fermented by a strain of lactic acid bacteria characterized in that it has:

[0171] - a viscosity at a shear gradient of 500 s 1 less than 0.125 Pa.s, 0.10 Pa.s, 0.075 Pa.s, 0.05 Pa.s; and / or

[0172] - a median diameter D50 less than 110 pm, 100 pm or 95 pm; and / or a concentration of insoluble solids in suspension less than 25 g / 100 g, 22.5 g / 100 g, 20 g / 100 g.

[0173] According to one embodiment, the fermented fruit and / or vegetable preparation does not include any component / ingredient other than the fruit and / or vegetable(s). In particular, preferably, it does not include any ingredient selected from the group consisting of added sugar, syrup, an acidifying agent, an enzyme, and a dairy product.

[0174] According to a preferred embodiment, the fermented fruit and / or vegetable preparation does not include ethanol.

[0175] According to one embodiment, the fermented fruit and / or vegetable preparation comprises at least 10 5 at least 10 6 or at least 10 7 CFU of lactic acid bacteria / ml of fruit and / or vegetable preparation.

[0176] According to a preferred embodiment, the fermented fruit and / or vegetable preparation comprises at least one strain of lactic acid bacteria, preferably selected from the group consisting of W. cibaria, Ln. pseudomesenteroides and Lpb. Plantarum, even more preferably from the group consisting of Lpb. Plantarum CNCM 1-6139, W. cibaria CNCM 1-6287 and L. pseudomesenteroides CNCM 1-6288.

[0177] Filtration

[0178] According to one embodiment, the methods and uses of the invention do not include any step for improving filterability other than fermentation prior to the filtration step. In particular, according to this embodiment, the methods and uses of the invention do not include any step involving the enzymatic degradation, for example with a pectinase, of the fruit and / or vegetable preparation, whether before or after its fermentation.

[0179] According to one embodiment, filtration is carried out at a cut-off threshold below 20pm, 10pm, 5pm, lpm, 0.45pm, 0.22pm, 0.1pm, 0.05pm or 0.01pm.

[0180] According to another embodiment, filtration is carried out at a cut-off threshold between 20pm and 0.05pm, between 10pm and 0.05pm, between 1pm and 0.05pm, between 0.45pm and 0.1pm or between 0.45pm and 0.20pm.

[0181] According to another embodiment, filtration is carried out at a cut-off threshold between 0.1pm and 0.001pm or between 0.05pm and 0.001pm.

[0182] Advantageously, the filtration is either microfiltration or ultrafiltration.

[0183] Filtration can be carried out in tangential, frontal, or quasi-frontal mode, or any other suitable filtration mode.

[0184] Filtration can be carried out at room temperature.

[0185] The operating parameters and characteristics of the filtration membranes (chemical composition, structure, pore size, geometry) are part of the general knowledge of a person skilled in the art, who will be able to determine those suitable for the processes and uses according to the invention (T. Urosevic et al. 2017). The process may further include a stabilization step following the filtration step. Stabilization can be achieved by high-temperature treatment (> 80 °C) such as pasteurization, sterilization, or canning.

[0186] The process may further include a step of adding an ingredient, for example, chosen from the group consisting of another juice and / or another purée, sugar, syrup, acidifying agent, dairy product, and / or texturizing agent (e.g., alginate). Preferably, this addition step occurs after the fermentation step, and more preferably after the filtration step.

[0187] The process may also include a vacuum evaporation-concentration step of the fruit and / or vegetable juice to reduce its volume.

[0188] Examples

[0189] Example 1: Fermentation efficiency for small-scale (laboratory scale) microfiltration

[0190] Materials and methods

[0191] Preparing the purees

[0192] The mangoes were pitted, trimmed and cut into approximately 3.5 cm cubes. Inedible parts were removed during preparation.

[0193] The fruit pieces were ground in batches of 1 kg in the Thermomix® (Vorwerk TM 31, Wuppertal, Germany) according to the following operating conditions: 1st grinding at speed 7 (1 min), homogenization, 2nd grinding at speed 7 (1 min).

[0194] All batches of puree were homogenized and then packaged in 75g pill bottles or 300g food trays. The samples were stored in a freezer at -18°C before further handling.

[0195] Two ranges of mango purees, hereinafter referred to as range 1 and range 3, corresponding to mangoes of different stages of maturity, were considered in order to verify the applicability of the method for various raw materials.

[0196] Fermentation

[0197] The following steps are carried out under sterile conditions.

[0198] Bacterial pre-cultures, inoculation and optimized fermentation

[0199] The selected lactic acid bacteria strain (Lpb. Plantarum CNCM 1-6139 deposited on November 8, 2024 at the CNCM) stored at -80°C is reactivated in 9 mL of MRS medium (deMan, Rogosa, Sharpe) and incubated at 30°C for 48 h. The reactivation is repeated to obtain a spontaneously growing LAB culture.

[0200] From bacterial pre-cultures in MRS liquid medium, the optical density is measured for each strain using a spectrophotometer at 600 nm in order to determine the bacterial concentration in uDO per mL.

[0201] The bacterial culture is then centrifuged at 5000 x g for 5 min at 10°C, and the cell pellet is washed twice successively with 10 mL of sterile double-distilled water, then resuspended in a controlled volume of sterile double-distilled water (10 mL). From this suspension, the inoculation volume is calculated to obtain a bacterial density of 0.05 µDO per mL of puree, which corresponds to 1 x O 7cells / mL according to the McFarland scale.

[0202] In this trial, for 550 g of puree, 1.7 mL of microorganism solution were inoculated and then the preparations were homogenized.

[0203] The inoculated and uninoculated purees are then incubated at 30°C for 48 hours for fermentation. The uninoculated puree is used as a control.

[0204] Under these working conditions, it was possible to do away with the pasteurization step usually required before inoculating the purees.

[0205] Fermentation monitoring

[0206] A count for each fermentation time (initial and final) is carried out on MRS-agar medium and on CGA-agar medium in triplicate for each sample, by serial dilutions of 1 / 10 in sterile peptone water.

[0207] The populations of lactic acid bacteria and yeast and mold are thus calculated from the bacterial colonies counted after incubation at 30 °C for 48 h.

[0208] Acidification monitoring of samples is also carried out for each fermentation time by measuring pH and titratable acidity.

[0209] At the end of fermentation, the purees were packaged in food trays and then stored at -20°C until microfiltration.

[0210] Microfiltration

[0211] The filterability tests of the purees were carried out in an Amicon® filtration cell (Merck Millipore, Burlington, USA).

[0212] A constant transmembrane pressure (TMP) of 2 bars and agitation (corresponding to a shear gradient of 250 s 1) were applied. A mass acquisition of the filtrate during filtration was carried out via a connected balance (Précisa 321LX, Précisa gravimétries AG, Dietikon, Switzerland) and Weight Transfer software in order to determine the density of the filtration flow.

[0213] Mango purees, treated or not by fermentation, were diluted with distilled water before microfiltration (different dilution factors were tested: 60 / 40 and 50 / 50; mass puree / mass water).

[0214] Filtrations were carried out with 3 organic membranes with average pore sizes of 20-25 pm, 11 pm and 6 pm (Whatman Grade 41, Grade 1, Grade 3, Whatman International Ltd, Maidstone, UK).

[0215] Results

[0216] Impact of fermentation on filtration performance Figure 1 shows the evolution of the filtration flux (J, Lh _1 .m -2) as a function of time, the main indicator of the productivity of the microfiltration process, for the different mango purees. This classic curve shape reflects the difference in kinetics (expressed as filtration flux as a function of time) between untreated purees and fermented purees.

[0217] The histogram in Figure 2 shows the average flow rates of the filtrations at a volume reduction factor (VRF) of 2, corresponding to the point at which half of the purée has been filtered. These values ​​allow us to visualize the average filtration behavior for the different purées. We observe that fermentation leads to a significant increase in the average flow rate, regardless of the dilution and the membrane cutoff threshold. This increase in flow rate, and consequently the reduction in membrane fouling, highlights the increased productivity resulting from this pretreatment. Furthermore, the filtration flow rates of the diluted fermented mangoes are not significantly different depending on the nature of the initial purée. This underscores the effectiveness of fermentation, which standardizes the initially heterogeneous raw materials into a more homogeneous product that is better suited to filtration.Finally, the membrane cutoff threshold does not appear to impact the average filtration flow rate. Under these operating conditions, neither the range nor the dilution appears to have a significant effect on filtration productivity.

[0218] Example 2: Efficiency of fermentation to modify the physical properties of mango purees and their filterability at pilot scale

[0219] In this second study, the aim was to demonstrate the impact of fermentation on the physical properties, particularly rheological and granulometric, of the treated purees.

[0220] A classic enzymatic treatment was also carried out on the mango purees to have a comparison with the results obtained following the fermentation treatment.

[0221] Two varieties of mango were studied: Cogshall and José. The José mango is a firm fruit, rich in fiber and very sweet, while the Cogshall mango is less firm, less fibrous and less sweet.

[0222] Three types of purée were obtained for each variety: an untreated control purée, a purée treated by fermentation, and a purée treated enzymatically. Only one stage of maturation was considered: the overripe stage, called M3G in Example 1.

[0223] Materials and methods

[0224] Grinding and packaging of mango purees

[0225] To produce the purees, the mangoes were trimmed, roughly chopped, and the inedible parts were removed. The purees were processed in a Thermomix® (Vorwerk TM 31, Wuppertal, Germany) in 1.2 kg batches under standardized conditions: an initial puree was processed at 1100 RPM for 1 minute, repeated twice, followed by a puree at 4400 RPM for 1 minute, and finally a last puree at 10200 RPM for 1 minute.

[0226] Grinding was carried out on several batches until approximately 40 kg of puree was obtained for each variety. After homogenization of all batches, the purees were divided into 800g trays and stored at -20°C. Before each test, the purees were thawed at room temperature for 12 hours and brought to a temperature of 25°C before the characterization and filtration tests.

[0227] Pre-treatment of mango purees

[0228] Two different pretreatments were carried out on the mango purees, one enzymatic and the other by fermentation.

[0229] For the enzymatic treatment, 3.2 kg of purée were divided between two Thermomix® units (1.6 kg of purée per Thermomix®). The purées were heated to 50°C using the Thermomix® heating system, with a stirring speed of 350 RPM to maintain consistent homogenization. An enzyme / purée concentration of 100 g / T was used, as recommended by the supplier. The enzyme used was Peclyve A32 Super, marketed by the Soufflet Biotechnologies® group, which contains a pectinase and an arabinase. After meticulous homogenization of the purées, the enzyme was incubated at the optimal time / temperature (1 hour / 50°C) in both Thermomix® units with a stirring speed of 350 RPM. Finally, the enzymatically treated mango purees were placed in the freezer at -20°C to preserve them and to slow down or even inhibit the action of the enzyme until they were used for filtration.

[0230] The fermentation of the mango purees was carried out in the same way as described in example 1.

[0231] Before each laboratory or pilot scale filtration, a dilution with 60% by mass of mango puree and 40% by mass of distilled water was carried out to make the operation possible (dilution retained according to the results of example 1).

[0232] Laboratory-scale microfiltration

[0233] The filterability tests of the mango purees were carried out in a semi-frontal, pressurized, and agitated Amicon®-type filtration cell (Merck Millipore, Burlington, USA), using the same method as that presented in Example 1. Filtrations were performed using a polyethersulfone membrane (pore size 0.1 µm and filtration area of ​​17 cm²). 2 ).

[0234] Pilot-scale microfiltration

[0235] Pilot-scale filtrations were performed using a modular filtration pilot (multipurpose mini-pilot, Firmus, France). In this study, two filtration modes were tested: tangential (internal-external), which is commonly used in fruit juice production, and quasi-frontal (external-internal), which is less frequently used in the food industry. Tangential filtration is often chosen for relatively high production rates, while quasi-frontal filtration offers advantages in terms of energy consumption, membrane cost, and ease of membrane cleaning. Quasi-frontal filtration can be considered more sustainable than tangential filtration, although the latter remains more efficient to date. Filtration tests were conducted at 23°C ± 2°C with a volume of 4 L of diluted mango puree.

[0236] In a tangential configuration, a monotubular membrane (0.1 µm; 0.008 m 2 ; TiO2-ZrO2; Rm = 1.2.1012 m' 1 (Orelis, France) was used. A traffic speed corresponding to a shear rate of 3700 s 1 was imposed, as well as a constant transmembrane pressure (AP) of 2 bar. The mass of the filtrate was acquired throughout the filtration using a connected balance (Précisa 321LX, Précisa gravimétries AG, Dietikon, Switzerland) and Weight Transfer software.

[0237] In a quasi-frontal configuration, a hollow fiber membrane (0.08 µm; 0.1 m 2 ; Polyethersulfone PES; Rm = 1.1.1012 m 1 (Polymem, France) in submerged mode was used. Filtration was performed by suction of the filtrate, using a peristaltic pump (520S IP31, Watson-Marlow, Massachusetts, USA) at a filtration flow rate set at 4.5 L / h. 1 .m -2The volumetric pump in the equipment allows recirculation of the retentate at low shear rates (47 s 1 ). Transmembrane pressure was acquired throughout the filtration process via a data acquisition system (Almemo 2690-8, Ahlborn GmbH, Germany).

[0238] Characterization of the physical properties of purees

[0239] Particle size distribution:

[0240] The size of suspended particles in mango puree was measured using a Mastersizer 3000 laser particle size analyzer (Malvern Instruments Limited, Worcestershire, UK). Measurements were performed in wet mode with an obscuration index between 20% and 30% and a stirring speed of 1500 RPM. The refractive indices of the particles and the dispersing phase (water) were 1.73 and 1.33, respectively. The absorption index of the particles was 0.1. The mean surface diameter D[3;2] and the median diameter D50 were chosen as indicators of particle size.

[0241] Insoluble solids in suspension (SIS):

[0242] The fraction of insoluble solids in suspension (SIS) was isolated by centrifuging 30 mL of a 18,000 g puree sample at 25°C (Sigma, 3-18K, Sigma GmbH, Osterode am Harz, Germany) and removing the supernatant to obtain the SIS concentration (g / 100 g of product, determined by weighing). The centrifugation time was calculated according to Stokes' Law (between 18 and 77 min) to adapt the centrifugation parameters to the viscosity of the puree. This adaptation was necessary to isolate particles larger than 1 µm regardless of the viscosity of the analyzed puree.

[0243] To characterize the rheological properties of the purées, the Physica MCR 301 rheometer (Anton Paar GmbH, Graz, Austria) was used, coupled with a thermal control system (VISCOTHERM VT2, Anton Paar GmbH, Graz, Austria) to ensure measurements at 25°C. Data acquisition was performed using StartRhéoplus software (RHEOPLUS / 32 V3.40, Anton Paar GmbH, Ostfildern, Germany). A coaxial cylinder geometry with a 5 mm air gap was used. This geometry consists of a rotor (CC 18.92-40 / 118 / S, Anton Paar GmbH, Graz, Austria) with a diameter of 18.9 mm and a fixed cup (stator) with a diameter of 28.9 mm.

[0244] Viscosity:

[0245] Viscosity (η, Pa.s) was measured as a function of the shear rate (y, s⁻¹) varying from 0.1 to 500 s⁻¹ 1 The Ostwald de Waele model was used to describe the flow behaviors:

[0246] [Math. 1]

[0247] ( = K x ÿn-1

[0248] With K being the consistency index (Pa.s n ) and n the flow index ( / )

[0249] The viscoelastic behavior of purees was measured through strain amplitude scanning which allows to represent the evolution of solid or elastic (G', Pa) and viscous (G", Pa) moduli as a function of the applied strain.

[0250] The samples first underwent a 50 s pre-shearing phase 1 during

[0251] 1 min, then 5 min of relaxation, finally a deformation sweep in amplitude from 0.1% to 1000% with a constant frequency of f = 1 Hz.

[0252] Results

[0253] Impact of fermentation on the physical properties of purees

[0254] The results obtained for key physical parameters, the modification of which can directly impact the performance of the microfiltration operation, are detailed below. Parameters related to the rheology of the purees, their particle size, and their concentration within the puree will be presented. Regarding rheological parameters, the viscosity of the purees will be considered, as it directly impacts the matrix's ability to flow across the surface of the microfiltration membrane. Viscoelastic behavior, particularly the solid or elastic modulus, will also be presented, as several previous studies have demonstrated a link between this parameter and the filterability of fruit-based suspensions.This parameter, which represents the elastic, or solid, component of the suspension, provides information on the microstructure of the product to be filtered, on the strength of the interactions that exist between the particles and the ability of the suspension to behave like a solid under mechanical stresses.

[0255] Particle size distribution

[0256] The results presented were obtained by laser granulometry measurements.

[0257] Figure 3 shows the particle size distributions of the control Cogshall mango purees and those that underwent enzymatic and fermentation pretreatments. For all three purees, the distribution is unimodal. The control puree has a large volume population around 120 µm. There is also a population of small particles between 1 and 40 µm, which represents a significant portion of the volume distribution of the untreated puree. Regarding the pretreatments, it is important to note that the use of the reference pretreatment (enzymatic pretreatment) and the use of the fermentation pretreatment result in the same particle size distribution for the mango puree. Indeed, for the two pretreated purees, the distribution was shifted towards smaller particle sizes, with the maximum peak value around 100 µm and a greater proportion of small particles smaller than 100 µm.This information is confirmed by Figure 4, which represents the median diameter of the particle size distribution (Dx(50), pm), i.e., the diameter below which 50% of the population falls, for mashes from the Cogshall and José varieties, both control and pre-treated. For both the Cogshall and José varieties, the median diameter of the control mash is greater than the median diameters of the enzymatically pre-treated or fermented mashes. Similarly, the fermentation treatment is as effective as the enzymatic treatment in reducing this median diameter value.

[0258] These results show that fermentation pretreatment is effective in modifying the particle size distribution of mango puree in the same way as enzymatic treatment. It is worth noting that fermentation treatment is equally effective on both varieties studied. The reduction in particle size of the fermented purees is expected to have a positive effect on the filterability of these suspensions. This effect will be validated in the final part of the study.

[0259] Concentration of insoluble solids in suspension (SIS)

[0260] Figure 5 represents the concentration of insoluble solids in suspension (SIS, g / 100g) in the puree.

[0261] The results show that untreated purees of both mango varieties have a higher concentration of membrane-bound particles than pretreated purees. For this parameter, although the reference enzymatic treatment is slightly more effective, fermentation significantly reduces the SIS content.

[0262] Viscosity

[0263] Figure 6 shows the evolution of viscosity (p, in Pa.s) as a function of the shear gradient (y, s 1) for control and pretreated Cogshall mango purees. For this type of suspension, shear-thinning behavior is commonly observed, meaning that the viscosity of the puree decreases with increasing shear gradient. The results for the Cogshall variety show that regardless of the shear gradient, the viscosity of the fermented puree is lower than the viscosity of the enzymatically treated puree, which is itself lower than the viscosity of the untreated puree.

[0264] Figure 7, which represents the viscosity at a shear gradient fixed at 500 s 1This confirms the observations for the Cogshall variety and shows that pretreatments decrease viscosity in José mango purees, with a slightly more pronounced effect for the enzymatic treatment in this case. This figure also highlights the differences in physical properties between the two mango varieties, given the differences in viscosity values, which confirms that fermentation treatment is effective even on purees from very different mango varieties.

[0265] These results demonstrate that both fermentation and enzymatic treatment reduce the viscosity of mango purees, with fermentation being more effective for the Cogshall variety. It is well known that reducing the viscosity of a suspension significantly improves its filterability, facilitating flow through the pilot filtration unit and also improving the passage of the filtrate through the membrane.

[0266] Figure 8 shows the evolution of the elastic (G', in Pa) and viscous (G", in Pa) moduli, which reflect the viscoelastic behavior, as a function of deformation (y, in %) for control and pre-treated Cogshall purees. The shape of the curves is typical for this type of suspension: for small deformations, the evolution of the moduli is linear and the elastic modulus predominates over the viscous modulus, while for large deformations, the evolution is no longer linear and the viscous modulus becomes predominant. The viscoelastic behavior, visible in these curves, provides information on the structural state of the suspension and is related to the particle size of the suspension as well as the process productivity. The lower the viscoelastic modulus values, the fewer large particles the suspension contains and / or the weaker the interactions between these particles.Here, it is important to note that the values ​​of the elastic and viscous moduli decrease with the pretreatments, with an even more marked decrease for fermented Cogshall mash.

[0267] This decrease is confirmed by Figure 9, which shows only the value of the elastic modulus in the linear viscoelastic region (LVE G', in Pa). As for the viscosity at 500 s 1 The difference in LVE G' value between the two mango varieties highlights their microstructural differences, linked to their physical, physico-chemical and biochemical characteristics.

[0268] The results demonstrate that both fermentation and enzymatic treatment reduce the solid nature of the purees (a characteristic unfavorable to filtration). Furthermore, fermentation proves more effective than enzymatic treatment for the Cogshall variety.

[0269] In summary, the results showed that fermentation of mango purees, for two varieties, refines the particle size distribution of the purees, reduces their suspended particle content, and lowers their viscosity and solidity. All these changes bring the mango purees to physical properties more suitable for microfiltration.

[0270] Impact of pretreatment on the filterability of purees

[0271] In this part of the study, filtration experiments were conducted to evaluate the effectiveness of fermentation in improving the filterability of mango purees. Three filtration operating conditions were considered. First, laboratory-scale tests were performed to validate the effectiveness of this pretreatment on a small scale, using organic membrane pore sizes of 0.1 µm. This is a standard practice in the fruit juice industry, resulting in a filtrate free of microbial contamination. Once this step was validated, pilot-scale tests were carried out using mineral and organic membranes with pore sizes of 0.1 µm. This approach allows for a more industrial scale and confirms the trends observed in the laboratory.The study focused on a filtration method commonly used in the juice industry: internal-external tangential flow filtration with ceramic tubular membranes. Tests were also conducted with a less common filtration method in the juice sector: external-internal quasi-frontal filtration with immersed organic hollow fiber membranes, which has the advantage of being a less expensive and more sustainable process.

[0272] Laboratory-scale filtration

[0273] Figure 10 shows the filtration flux values ​​at a volume reduction factor of 2 (VRF 2) (J, Lh _1 .m -2This is for filtrations conducted on a laboratory scale. Since the filtration flux decreases over time due to membrane clogging by suspension particles, the value at FRV2, which corresponds to the point where 50% of the initial volume has been filtered, is fixed to allow for comparison between samples. It is generally accepted that this laboratory-scale experiment provides information on the filterability of the suspension: the higher the filtration flux obtained, the better the filterability of the product.

[0274] The results here show that for both varieties, pretreatment by fermentation of the mango puree improves its filterability, given that the fluxes at FRV2 are higher. The effect of fermentation is very pronounced for both varieties, clearly demonstrating that this pretreatment is as effective as, or even more effective than, enzymatic pretreatment.

[0275] Pilot-scale filtration

[0276] Figure 11 shows the filtration resistance (R, in m 1 The treated and untreated purees of both varieties were analyzed as a function of time (t, s) in a tangential configuration. Filtration resistance was the chosen parameter because it allows for a comparison of the two pilot filtration methods, which behave differently. Thus, the filtration resistance of the suspension is inversely proportional to its filterability: the higher the resistance, the less filterable the puree.

[0277] The results for the classic tangential configuration show that fermentation does indeed increase the filterability of the purées for both varieties, as the filtration resistance values ​​are lower for the treated purées. For the José variety, the curve for the fermented purée even overlaps with the curve for the enzymatically treated purée, clearly demonstrating the effectiveness of this pretreatment.

[0278] Figure 12 shows the filtration resistance of untreated and treated purees, for both varieties, as a function of time (t, s) and in a quasi-frontal configuration, which is the least classically used filtration method in the food industry.

[0279] The results for this configuration also show that fermentation effectively increases the filterability of the purées for both varieties, as the filtration resistance values ​​are lower for the treated purées. For the José variety, the curve for the fermented purée even overlaps with the curve for the enzymatically treated purée, confirming that fermentation appears particularly well-suited to increasing the filterability of José mango purée.

[0280] In summary, the study of the physical properties of mango puree showed that implementing the innovative fermentation pretreatment for both varieties counteracts the properties of these purees that make them unsuitable for microfiltration. Fermentation thus refines the particle size distribution, reduces the suspended particle content, lowers the viscosity and the solid behavior of the purees, and therefore improves their filterability regardless of the filtration method used.

[0281] Example 3: Fermentation efficiency for small-scale (laboratory-scale) microfiltration with Weissella cibaria or Leuconostoc pseudomesenteroides strains

[0282] Materials and methods

[0283] The materials and methods used in this example are the same as those in Example 1, with the exception of the lactic acid bacteria strain used.

[0284] The lactic acid bacteria strain Lpb. Plantarum CNCM 1-6139 has been replaced by the Weisella Cibaria strain CNCM 1-6287 deposited on December 10, 2025 at the CNCM or the Leuconostoc Pseudomesenteroides strain CNCM 1-6288 deposited on December 10, 2025 at the CNCM).

[0285] Cogshall mangoes at stage 3 TBR2 were used.

[0286] The filterability tests of the purees were carried out in an Amicon® filtration cell (Merck Millipore, Burlington, USA).

[0287] A constant transmembrane pressure (TMP) of 2 bars and agitation (corresponding to a shear gradient of 250 s 1 ) were applied. A mass acquisition of the filtrate during filtration was carried out via a connected balance (Précisa 321LX, Précisa gravimétries AG, Dietikon, Switzerland) and Weight Transfer software in order to determine the density of the filtration flow.

[0288] Mango purees, treated or not by fermentation, were diluted with distilled water at a puree mass / water mass dilution factor of 60 / 40 before microfiltration.

[0289] The filtrations were carried out with a glass microfiber membrane with an average pore size of 0.7 pm.

[0290] Results

[0291] Impact of fermentation on filtration performance

[0292] Figure 13 shows the change in filtrate mass (in g) over time (in s) for mango puree: control, fermented with Weissella Cibaria, and fermented with Leuconostoc Pseudomesenteroides, diluted 60 / 40 (puree / water; w / w) and filtered through a 0.7 µm cutoff. An increase in filtrate mass over time is observed for the purees fermented with Weissella Cibaria and Leuconostoc Pseudomesenteroides compared to the unfermented control.

[0293] The average filtration fluxes (J, Lh_1 .m' 2 ) are as follows:

[0294] Table 1]

[0295] We observe that, as with fermentation with Lpb. Plantarum, fermentation with W. Cibaria and Ln. Pseudomesenteroides allows a significant increase in the average filtration flux.

[0296] References

[0297] T. Urosevic, D. Povrenovic, P. Vukosavljevic, I. Urosevic, S. Stevanovic, Recent developments in microfiltration and ultrafiltration of fruit juices, Food Bioprod. Process. 106 (2017) 147-161. doi:10.1016 / j.fbp.2017.09.009.

[0298] C. Bhattacharjee, VK Saxena, S. Dutta, Fruit juice processing using membrane technology: A review, Innov. Food Sci. Emerg. Technol. 43 (2017) 136-153. doi:10.1016 / j.ifset.2017.08.002.

[0299] A. Cendres, Innovative process for extracting fruit juice by microwave: manufacturing viability and nutritional quality of juices, University of Avignon, 2010.

[0300] S. Chatterjee, S. Chatterjee, B. P. Chatterjee, A.K. Guha, Clarification of fruit juice with chitosan, Process Biochem. 39 (2004) 2229-2232. doi:10.1016 / j. procbio.2003.11.024.

[0301] R.C.C. Domingues, S.B. Faria Junior, R.B. Silva, V.L. Cardoso, M.H.M. Reis, Clarification of passion fruit juice with chitosan: Effects of coagulation process variables and comparison with centrifugation and enzymatic treatments, Biomass-Deriv. Pentoses. 47 (2012) 467- 471. doi:10.1016 / j.procbio.2011.12.002.

[0302] L. Petruzzi, D. Campaniello, B. Speranza, M.R. Corbo, M. Sinigaglia, A. Bevilacqua, Thermal Treatments for Fruit and Vegetable Juices and Beverages: A Literature Overview: Heat treatment for juices and beverages..., Compr. Rev. Food Sci. Food Saf. 16 (2017) 668-691. doi: 10.1111 / 1541-4337.12270.

[0303] G. Rajauria, BK Tiwari, eds., Fruit juices: extraction, composition, quality, and analysis, Academic Press, London, 2018. (original in electronic form)

[0304] (This sheet is not part of the international application nor does it count as a sheet thereof)

[0305] (original in electronic form)

[0306] (This sheet is not part of the international application nor does it count as a sheet thereof)

[0307] RESERVED FOR THE RECEIVING OFFICE (original in electronic form)

[0308] (This sheet is not part of the international application nor does it count as a sheet thereof)

[0309] FOR INTERNATIONAL OFFICE USE ONLY

Claims

1. DEMANDS 1. A method for improving the filterability of a fruit and / or vegetable preparation, characterized in that it comprises a step of fermenting a fruit and / or vegetable preparation with at least one strain of lactic acid bacteria, so as to obtain a fermented fruit and / or vegetable preparation.

2. A method for filtering a fruit and / or vegetable preparation, characterized in that it comprises: - the improvement of the filterability of the fruit and / or vegetable preparation according to the process of claim 1 and - a filtration step of the fermented fruit and / or vegetable preparation to a cut-off threshold of less than 22pm.

3. Use of at least one strain of lactic acid bacteria to improve the filterability of a fruit and / or vegetable preparation.

4. Use according to claim 3 characterized in that it comprises the steps of: - inoculation of a fruit and / or vegetable preparation with at least one strain of lactic acid bacteria and - fermentation of the fruit and / or vegetable preparation, so as to obtain a fermented fruit and / or vegetable preparation.

5. A method for improving filterability according to claim 1, or a filtration method according to claim 2, or a use according to claim 3 or 4, characterized in that the fruit and / or vegetable preparation is fruit and / or vegetables that have been sieved, crushed and / or pressed.

6. A method for improving filterability according to claim 1 or 5, or a filtration method according to claim 2 or 5, or a use according to any one of claims 3 to 5, characterized in that it does not include a sterilization, pasteurization and / or clarification step before the fermentation step.

7. A method for improving filterability according to any one of claims 1 or 5 to 6, or a filtration method according to any one of claims 2 or 5 to 6 or a use according to any one of claims 3 to 6, characterized in that it does not include a sterilization, pasteurization and / or enzymatic or physical clarification step.

8. A method for improving filterability according to any one of claims 1 or 5 to 7, or a filtration method according to any one of claims 2 or 5 to 7 or a use according to any one of claims 3 to 7, characterized in that the fruit and / or vegetable is selected from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture thereof.

9. A method for improving filterability according to any one of claims 1 or 5 to 8, or a filtration method according to any one of claims 2 or 5 to 8, or a use according to any one of claims 3 to 8, characterized in that it comprises a step of inoculating the fruit and / or vegetable preparation with a strain of lactic acid bacteria at an inoculation rate greater than 10 5 UFC / ml of fruit and / or vegetable preparation.

10. A method for improving filterability according to any one of claims 1 or 5 to 9, or a filtration method according to any one of claims 2 or 5 to 9 or a use according to any one of claims 3 to 9, characterized in that the lactic acid bacteria strain is a strain of genus selected from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc.

11. A method for improving filterability according to any one of claims 1 or 5 to 10, or a filtration method according to any one of claims 2 or 5 to 10, or a use according to any one of claims 3 to 10, characterized in that the lactic acid bacteria strain is selected from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum.

12. A method for improving filterability according to any one of the claims I or 5 to 11, or filtration process according to any one of claims 2 or 5 to II or use according to any one of claims 3 to 11, characterized in that the lactic acid bacteria strain is a strain selected from the group consisting of Lactiplantibacillus plantarum Weissella cibaria and Leuconostoc pseudomesenteroides.

13. A process for manufacturing fruit and / or vegetable juice characterized in that it comprises the steps of: - fermentation of a fruit and / or vegetable preparation with at least one strain of lactic acid bacteria for less than 72 hours in order to obtain a fermented fruit or vegetable preparation and - filtration of the fermented fruit and / or vegetable preparation to a cut-off threshold of less than 22pm.

14. Manufacturing process according to claim 13, characterized in that it comprises a step of inoculating the fruit and / or vegetable preparation with only one or more strains of lactic acid bacteria.

15. Manufacturing process according to claim 13 or 14, characterized in that the fruit and / or vegetable preparation is fruit and / or vegetables that have been sieved or ground.

16. A manufacturing process according to any one of claims 13 to 15, characterized in that it does not include a sterilization, pasteurization and / or clarification step prior to the fermentation step.

17. A manufacturing process according to any one of claims 13 to 16, characterized in that it does not include a sterilization, pasteurization and / or enzymatic or physical clarification step.

18. A manufacturing process according to any one of claims 13 to 17 characterized in that the fruit and / or vegetable is selected from the group consisting of mango, pineapple, papaya, guava, peach, apricots, apple, plum, melon, watermelon, cabbage, pomegranate and nightshade or a mixture thereof.

19. A manufacturing process according to any one of claims 13 to 18, characterized in that it comprises a step of inoculating the fruit and / or vegetable preparation with a strain of lactic acid bacteria at an inoculation rate greater than 10 5 UFC / ml of fruit and / or vegetable preparation.

20. A manufacturing process according to any one of claims 13 to 19, characterized in that the lactic acid bacteria strain is a strain of genus selected from the group consisting of Lactococcus, Pedicoccus, Lactiplantibacillus, Weissella and Leuconostoc.

21. A manufacturing process according to any one of claims 13 to 20 characterized in that the lactic acid bacteria strain is selected from the group consisting of Lactococcus lactis, Pediococcus pentosaceus, Lactiplantibacillus plantarum, Lactiplantibacillus paraplantarum, Weissella cibaria, Weissella confusa, Weissella paramesenteroides, Leuconostoc pseudomesenteroides, Leuconostoc mesenteroides, and Leuconostoc citreum.

22. A manufacturing process according to any one of claims 13 to 21, characterized in that the lactic acid bacteria strain is a strain selected from the group consisting of Lactiplantibacillus plantarum, Weissella cibaria and Leuconostoc pseudomesenteroides.

23. Preparation of fruit and / or vegetables fermented by a strain of lactic acid bacteria characterized in that it has a filtration flux at FRV2 greater than 10h^.m' 2 and / or a viscosity at a shear gradient of 500 s 1less than 0.125 Pa.s and / or a median diameter D50 less than 110 pm and / or a concentration of insoluble solids in suspension less than 25g / 100g.

Images