Influence of the Type of Sauerkraut Fermentation with Probiotics Strains on Folate Content, Antioxidant Activity and Sensory Analysis

Abstract

Sauerkraut (SK) is a fermented food of plant origin recognised for its nutritional properties and health benefits. It is traditionally produced through spontaneous fermentation, carried out by the native microflora of fresh cabbage, which includes Gram-negative bacteria, moulds, yeasts and finally lactic acid bacteria (LAB), responsible for transforming natural sugars into lactic acid. However, spontaneous fermentation can also promote the growth of undesirable microorganisms, leading to risks of sensory or food safety alterations, such as the production of biogenic amines. To prevent these limitations, the use of LAB starter cultures is presented as a promising alternative. This study evaluated the fermentation of cabbage at 18 °C and 25 °C, comparing spontaneous fermentation with controlled fermentations using probiotic cultures (Lactiplantibacillus plantarum 229v and Lacticaseibacillus rhamnosus GG). Due to its nutritional importance, the folate content of different types of SK has been studied. Spontaneous fermentation showed positive results in all studied parameters; however, L. plantarum 229v was the inoculum with the highest and fastest acidifying efficacy, reducing the pH to below 4.0 after 7 days at both temperatures. At 25 °C, L. plantarum 229v achieved LAB counts higher than those of L. rhamnosus GG (7.02 vs. 6.00 log CFU·g⁻¹) and exerted the most effective control over enterobacteria and moulds/yeasts, reaching undetectable values (0 log CFU·g⁻¹) on day 7 under both conditions. Antioxidant activity after 42 days of fermentation was higher at 18 °C with L. rhamnosus GG, which showed the highest values (up to 3.50 mg CE·g⁻¹ FW), followed by L. plantarum 229v and spontaneous fermentation. In terms of total folate content (TFC) retention, spontaneous fermentation was particularly effective at both temperatures after 42 days of fermentation (794.741 µg/100 g TFC at 18 °C and 586.542 µg/100 g TFC at 25 °C). In sensory analysis, spontaneous fermentation, in general acceptance, was rated highest (6.2), followed by L. plantarum 229v (5.5) and L. rhamnosus GG (5.3). Considering all the factors studied, SK fermentation with the probiotic strain L. plantarum 229v has proven to be the most suitable.

1. Introduction

Sauerkraut (SK) is a fermented vegetable food of great importance, with significant gastronomic and cultural value, and one of the most widely consumed fermented products worldwide. Although often associated with German cuisine, SK is produced throughout Asia and Central and Eastern Europe, as well as in the United States [1,2]. SK is produced by the lactic fermentation of finely chopped and salted white cabbage (Brassica oleracea). This process is driven by lactic acid bacteria (LAB) that are naturally present on the surface of the cabbage. LAB metabolise sugars into lactic acid, lowering the pH of the medium, which ensures its preservation and gives it its characteristic flavour. LAB also improve the bioavailability of nutrients compared to fresh cabbage [3,4,5]. Key bioactive compounds in SK include folates (vitamin B9), phenolic compounds and glucosinolates [6]. Phenolic compounds and glucosinolates contribute to the aroma and provide antioxidant and anticarcinogenic properties, while folates are vital for DNA synthesis and repair, cell division and foetal neurological development.

In spontaneous fermentation, the initial microflora is heterogeneous and includes LAB, as well as other bacteria, yeasts and moulds. This microbial diversity in the early stages can allow the proliferation of undesirable microorganisms, leading to risks of sensory alterations or food safety issues, such as the production of biogenic amines (BAs) [7,8]. Although LAB usually dominate in the later stages due to the combined effects of salt and pH reduction, this change is not immediate. This increases the likelihood of BA formation, which can affect product safety and sensory attributes.

To address these safety and quality challenges in spontaneous fermentation, the use of starter cultures is an effective strategy to standardise the process, guide microbial succession, improve sensory and nutritional characteristics, and reduce the formation of BAs. In this way, it is possible to obtain a more predictable, safer and higher value-added fermentation [9]. Among these, LAB starter cultures are of particular interest when, in addition to their technological function, they have probiotic properties, as they provide additional functional benefits. This dual potential is particularly relevant in the context of the growing demand for non-dairy probiotic foods, driven by lactose intolerance, milk protein allergies and the adoption of vegetarian or vegan diets [10]. Various plant-based materials, such as cereals, fruits and legumes, have been successfully used as probiotic carriers [10].

In this context, Lactiplantibacillus plantarum is considered one of the most versatile starters for plant fermentation, as it has been isolated from foods such as kimchi, olives, pickles, sauerkraut and fermented animal products [11]. This species can tolerate NaCl concentrations of up to 4% and has probiotic properties, such as reducing the risk of inflammatory bowel disease and stimulating immune responses. Lacticaseibacillus rhamnosus is one of the most studied probiotics, with documented benefits for gastrointestinal, respiratory, allergic and cardiovascular health [12]. However, the use of probiotic microorganisms as starter cultures in cabbage fermentation has not been thoroughly investigated [13].

The objective of this study was to compare spontaneous fermentation with fermentations inoculated with L. plantarum 229v and L. rhamnosus GG at two temperatures (18 °C and 25 °C). The evaluation included starter implantation, sensory characteristics, antioxidant capacity and folate content, with the aim of identifying the optimal conditions for producing safer SK with higher nutritional quality and greater probiotic potential.

2. Materials and Methods

2.1. Sample Preparation

Fresh cabbages were purchased from local supermarket in Murcia, Spain. To produce SK, the outer leaves of the cabbages were removed and then cut into thin slices of 2–3 mm using a food processor (Moulinex cuisine companion model HF800A13, Ecully Cedex, France) for 30 s at rotation speed of 7. The crushed cabbage was mixed with 0.9% salt and transferred to 250 mL glass jars. Before closing the glass jars, the crushed cabbage was gently pressed until a little exudate of the cut plant material was obtained. The jars with the crushed cabbage were kept closed in a dry and dark space at two different storage temperatures: 25 °C and 18 °C.

2.2. Starter Culture Preparation

L. rhamnosus GG was obtained from commercial lyophilised Bivos Lactobacillus GG (10 sachets, 1.5 g) (Ferring, S.A, Madrid, Spain), and L. plantarum 229v was obtained from commercial lyophilised Protansitus LP (SALVAT, S.A, Barcelona, Spain).

To prepare the inoculum, an aliquot of the lyophilised sample was added to 5 mL of tryptic soy broth (TSB) (Cultimed, Barcelona, Spain) and seeded with a microbiological loop onto MRS agar plates (Aplichem, Darmstadt, Germany), which were incubated overnight at 37 °C to obtain isolated colonies. The L. plantarum 229v and L. rhamnosus GG inoculum strains were prepared by transferring a single colony obtained from the MRS agar plates to TSB, which was incubated for 24 h at 25 °C before being stored at −20 °C in a solution of 40% TSB and 60% glycerol until use. Fresh cultures for the experiments were prepared by incubating a loopful of pure culture in TSB for 24 h at 25 °C. The strains were standardised by dilution in TSB until an optical density (OD) of 0.1 at 600 nm was reached (Nicolet Evolution 300 UV–VIS spectrophotometer, Thermo Electron Corporation) to obtain a concentration of probiotic strains of 10⁶ cfu mL⁻¹. For the preparation of the pure culture that constitutes the inoculum, 1 isolated colony was taken and inoculated in 5 mL of beer wort and incubated at 37 °C for 24 h. The starters were inoculated at 10⁶ colony-forming units (cfu mL⁻¹).

2.3. Fermentation

The fermentation experiments were carried out in the 250 mL glass jars containing the cabbage, which were divided into three groups (spontaneous fermentation, controlled fermentation with L. rhamnosus GG and controlled fermentation with L. plantarum 229v. The spontaneous fermentation jars were not inoculated. Each jar of the other two batches was inoculated with 1 mL of liquid culture of L. rhamnosus GG and L. plantarum 229v, respectively. Each group was divided into two sub-groups for fermentation for 42 days at temperatures of 18 °C or 25 °C, respectively. The experiments were performed in triplicate.

2.4. Microbiological Analysis and pH Determination

Sampling was carried out on days 0, 7, 14, 21, 28 and 42. The pH was measured immediately after opening the jars. For microbiology and pH measurement analyses, the brine was removed aseptically in a laminar flow hood. The pH was measured directly from the glass jar by a pH meter (CRISON, 507, Barcelona, Spain). Three fermentation glass jars were opened at each time point, and each jar was measured three times.

For microbial analysis, four different media were applied to measure growth by enumerating log cfu g⁻¹ of sauerkraut. The total mesophilic aerobic bacteria, LAB, enterobacteria, and moulds and yeasts were determined in standard methods agar (PCA), De Man–Rogosa–Sharpe agar (MRS), violet red bile glucose agar (VRBG), and PCA + chloramphenicol agar, respectively.

For each sample, 25 g of SK was taken and transferred to a stomacher bag with 225 mL of peptone water, and from that solution the corresponding dilutions were made and harvested on triplicate agar plates. Then, the plates were incubated at 37 °C for 48 h for mesophilic aerobes and LAB, at 25 °C for 72 h for moulds and yeasts, and at 37 °C for 24 h for enterobacteria.

2.5. Determination of Folates

Folates were extracted from samples following a previously described procedure [14]. Enzymatic deconjugation and purification of samples were carried out following the methodology previously described [15]. To achieve this, a 5 mL aliquot was incubated for 3 h at 37 °C under a nitrogen atmosphere with 1 mL of conjugase prepared from fresh pig kidneys, according the method described above [16]. To stop the enzymatic activity, the samples were boiled for 5 min and then cooled on ice. The samples were then filtered through 25 mm Ø nylon disposable filters with a 0.45 µm pore size (3 mL/500 mg quaternary amine N⁺, counterion CL⁻, no. 52664; Whatman, Inc., Florham Park, NJ, USA). Finally, folates were purified by means of SAX cartridges connected to a Supelco 12-port vacuum manifold (Supelco, Bellefonte, PA, USA).

The eluted samples were weighed, and the purified extracts were kept under refrigeration for no longer than 2 h before they were placed in the cooled autosampler and injected onto the high-performance liquid chromatography (HPLC) column. The extraction, deconjugation and purification procedures were carried out under subdued light to prevent photodegradation of the folates.

2.6. HPLC-MS/MS Analysis

The analysis was carried out on an HPLC-MS/MS system consisting of an Agilent 1100 Series HPLC (Agilent Technologies, Santa Clara, CA, USA) equipped with a µ-wellplate autosampler and a capillary pump and connected to an Agilent Ion Trap XCT Plus mass spectrometer (Agilent Technologies, Santa Clara, CA, USA) using an electrospray (ESI) interface.

Chromatographic conditions and mass spectrometer variables were as described by Serrano-Amatriain et al., 2016 [15]. Different folate isoforms were quantified, specifically folic acid (FA), dihydrofolate (DHF), tetrahydrofolate (THF), 5-methyl tetrahydrofolate (5MTHF) and 5-formyl tetrahydrofolate (5FTHF).

2.7. Antioxidant Activity

The scavenging activity of cabbage against the DPPH radical was assessed according to the method of Blois (1958) with some modifications [17]. Briefly, 2.5 g of diluted sample cabbage (1/10 in methanol) was taken, and 0.5 mL of DPPH (1,1-diphenyl 2-picrylhydrazyl) was added. The reaction mixture was vortexed thoroughly and left in the dark at room temperature (25 ± 1 °C) for 30 min. The decrease in the absorbance (due to the proton donating activity) was measured at 517 nm using a UV–VIS spectrophotometer (Thermo Electron Corporation, Nicolet evolution 300, Waltham, MA, USA). Catechin was used as a reference, expressed as mg of catechin equivalents per g of fresh weight (mg CE g⁻¹ FW). The samples were analysed in triplicate.

2.8. Sensory Evaluation

To evaluate the acceptability of spontaneously fermented SK and SK fermented with the probiotic strains L. rhamnosus GG and L. plantarum 229v, seven sensory parameters (appearance, aroma, texture, flavour, colour, acidity and overall rating) were evaluated in the different SK samples. Before the sensory evaluation, the SK samples were refrigerated (8 °C) and then labelled with a random code, and approximately 20 g of each type of SK was served to the panellists. Between samples, the panellists drank still bottled water (Lanjarón, Spain) to cleanse their palates. A panel of five trained panellists was used to perform the sensory analysis of SK, of which two were women and three were men. The samples (n = 6) were evaluated by the panellists using a 10-point hedonic scale, where 0 meant totally unpleasant and 10 meant they liked it very much. A score of 5 was used as the neutral threshold for acceptance of the SK samples [18].

2.9. Statistical Analysis

Before conducting this analysis, data normality was verified using the Shapiro–Wilk test, and homoscedasticity was assessed using Bartlett’s test. Statistical significance was set at p < 0.05. All analyses were generated using the R statistical computing language and environment.

For each temperature and time condition, differences among the three fermentation types studied—for each variable included in the study—were tested using simple linear regression. To define the differences between the three different fermentations, a multiple comparisons test with Holm correction was used.

3. Results and Discussion

3.1. Changes in pH Value During Fermentation

Figure 1 shows the evolution of pH during the fermentation process of cabbage samples inoculated with L. rhamnosus GG, L. plantarum 229v and spontaneous fermentation, which serves as a control. At 18 °C (Figure 1a), a progressive decrease in pH was observed in all treatments during the fermentation process. However, the dynamics of acidification varied depending on the treatment used. On day 0, the initial pH values were very similar in all the cases (6.0 for spontaneous fermentation, 6.1 for L. rhamnosus GG and 5.9 for L. plantarum 229v). In the spontaneous fermentation, the pH decreased quickly between days 0 and 7, reaching 3.5 and 3.7 by day 21. The pH drop of L. plantarum 229v was similar to that of spontaneous fermentation, but a little slower since it takes 14 days to lower the pH to below 4. In contrast, the L. rhamnosus GG treatment exhibited a slower and more gradual acidification, maintaining a pH of 5.0 on day 21. By the end of the fermentation period (day 42), all treatments reached similar pH values, ranging from 3.5 to 3.9. These results highlight differences in acidification capacity among LAB strains. L. plantarum has been documented as one of the most efficient acidifying species in vegetable fermentation due to its high lactic acid production and rapid pH decrease [12,19]. Moreover, the use of 18 °C, although below the optimal growth temperature for L. plantarum, allows for a controlled and rapid pH decrease, enhancing microbial safety by inhibiting undesirable bacteria [20].

At 25 °C (Figure 1b), fermentation proceeded with a faster and more uniform acidification compared to 18 °C. All three treatments began with similar pH values, 6.0 for spontaneous and 5.9 for L. rhamnosus GG and L. plantarum 229v. From day 7 onwards, the pH dropped rapidly in all treatments, reaching values between 3.5 and 4.3. This pattern reflects the higher metabolic activity of LAB at temperatures above 20 °C, promoting faster acidification. This accelerated decrease in pH is indicative of a successful establishment of LAB as a starter culture in a given medium [21]. Days 14 and 21 followed a similar trend. However, on days 28 and 42, L. rhamnosus GG and L. plantarum 229v showed pH values of 2.8 and 3.0. This suggests similar acidification behaviour by both strains towards the end of the fermentation process. A study conducted by [21] where sauerkraut was fermented at room temperature with L. plantarum 229v reported successful strain adaptation, achieving pH values comparable to those found in our study.

3.2. Microbiological Changes During Fermentation

As shown in Figure 2, at 18 °C, the initial lactic acid bacteria count (LAB) (Figure 2a) during the first week (1 to 8 days) in all four sauerkraut (SK) treatments ranged between 3 and 4 log cfu g⁻¹. Previous studies have shown that during the early stages of fermentation, the native microbiota can temporarily dominate, diminishing the effect of starter inocula [22]. On day 15, spontaneous fermentation showed the highest LAB count, with a mean value of 8.38 log cfu g⁻¹, followed by L. plantarum 229v (6.67 log cfu g⁻¹) and L. rhamnosus GG with the lowest count of 6.33 log cfu g⁻¹. However, these differences decreased on day 22, when all treatments reached similar LAB counts, with an average of between 6.4 and 7.5 log cfu g⁻¹, and with spontaneous fermentation in the lead. This trend remained stable until day 29 and persisted until the end of fermentation (day 43), when LAB counts were higher for spontaneous fermentation (8.08 log cfu g⁻¹), with similar values for L. plantarum 229v and L. rhamnosus GG (5.53 log cfu g⁻¹ and 6.2 log cfu g⁻¹, respectively).

At 25 °C (Figure 2b), LAB growth followed a pattern like that observed at 18 °C. The initial LAB counts were similar in all treatments, ranging between 2 and 4 log cfu g⁻¹. However, L. plantarum 229v showed much more vigorous growth during the first 15 days, which corresponds to a decrease in pH. On day 29, the LAB counts of the three fermentations at 25 °C remained high and stable in all groups, ranging between 7.0 and 8.2 log cfu g⁻¹, with spontaneous fermentation showing the highest count. On day 43, spontaneous fermentation maintained the highest average count (7.87 log cfu g⁻¹), followed by L. plantarum 229v (7.02 log cfu g⁻¹) and L. rhamnosus GG (6.00 log cfu g⁻¹). Furthermore, spontaneous fermentation was the treatment that registers the highest microbial growth. The literature agrees that spontaneous fermentation is more adaptable to the environment, reaching LAB levels of between 8 and 9 log cfu g⁻¹. The species L. plantarum has shown great adaptability in vegetable media and especially in cabbage fermentation for the production of sauerkraut or kimchi, presenting values similar to our study. L. rhamnosus GG has very good probiotic properties, but under the conditions studied, it has not acted as well as L. plantarum 229v as an SK starter culture [23,24].

The growth of aerobic mesophilic bacteria (AMB) (Figure 2c) throughout the fermentation of SK at 18 °C showed a progressive increase in AMB in all treatments. The initial count ranged between 3.9 and 4.5 log cfu g⁻¹ for all treatments, increasing progressively between days 8 and 15 of the experiment, showing differences between spontaneous fermentation and fermentation controlled by inoculants. In the days that followed and until the end of the experiment, the counts ranged between 6 and 7 log cfu g⁻¹. The growth of AMB (Figure 2d) throughout the fermentations of SK at 25 °C showed a progressive increase in AMB in all groups during the early stages of the process (1 to 15 days). The average count on day 1 ranged between 3.30 and 4.20 log cfu g⁻¹, continuing with a growth trend during the following days. The maximum values were reached on day 22, with spontaneous fermentation showing significantly higher values (8.83 log cfu g⁻¹) than fermentation with the strains L. rhamnosus GG and L. plantarum 229v (average of 6.2 log cfu g⁻¹). This behaviour can be attributed to the ability of LAB species, such as L. plantarum and L. rhamnosus, to grow under partially aerobic conditions during the initial stages of fermentation, before more stringent anaerobic conditions are established, where they reach optimal growth conditions [24], which can be attributed to the increase in AMB in the days following until the end of fermentation. These results indicate that fermentation at 25 °C facilitates the sustained proliferation of mesophilic aerobes, regardless of the inoculation strategy. In contrast, previous findings suggest that moderately lower temperatures (15–20 °C) promote a more controlled and sequential microbial succession, characterised by a gradual decline in aerobic competitors, such as enterobacteria, and greater dominance of lactic acid bacteria (LAB) [25]. These conditions have been associated with greater biochemical stability, effective acidification kinetics and superior organoleptic properties in fermented vegetables, particularly SK [25]. It is important that the AMB count is not much higher than the LAB count, as this implies that the inoculum is the predominant culture, and therefore there is good establishment and control of the microbial flora.

As shown in Figure 2e, during fermentation at 18 °C, enterobacteria were only detected in the treatment with L. plantarum 229v on day 1, with a count of 3.82 log cfu g⁻¹, but they quickly controlled the population from the start to the end of fermentation. In contrast, L. rhamnosus GG allowed the growth of enterobacteria up to 8 days, which is not adequate. However, spontaneous fermentation did not show growth of enterobacteria. The initial presence of enterobacteria, especially in inoculated treatments, can be attributed to early microbial competition between LAB and the native microflora. This is especially plausible at 18 °C, where LAB metabolic activity is slower compared to fermentations at 25 °C or higher [26]. At 25 °C, L. plantarum 229v and spontaneous fermentation showed exactly the same trend as at 18 °C. In contrast, L. rhamnosus GG at 25 °C, despite yielding good results in terms of pH reduction and LAB growth, was not able to adequately control the growth of enterobacteria until day 15, which poses a risk of pathogen growth or organoleptic alterations [26]. Park et al., 2014 [26] observed a similar trend and reported an initial increase in enterobacteria during the fermentation of vegetables such as kimchi and cabbage, followed by their disappearance between days 7 and 14 at 15–20 °C, coinciding with the growth of LAB.

Yeast and mould counts at 18 and 25 °C (Figure 2g,h) showed growth detected only in the treatment with L. rhamnosus GG. In contrast to the other two treatments, which effectively suppressed mould and yeast growth within the first 24 h, L. rhamnosus GG required a longer period to achieve control. In contrast, no growth of mould and yeast was observed in spontaneous fermentation or in the treatment with L. plantarum 229v. From day 8 until the end of fermentation (day 42), yeast and mould counts remained undetectable in all treatments. In addition, Satora et al., 2020 [27], investigated the effect of fermentation temperature (4, 10 and 20 °C) and packaging conditions on the diversity of white colony-forming yeasts in kimchi. Their findings showed that lower fermentation temperatures significantly reduced the growth of these yeasts, suggesting that lower temperatures may limit the proliferation of undesirable yeasts, promote the dominance of LAB and improve the microbiological safety of the final product [27].

3.3. Antioxidant Activity Results

As shown in Figure 3a, during SK fermentation at 18 °C, antioxidant activity was initially low in all treatments on day 0. The experiment started with values of 0.93 mg CE g⁻¹ FW for spontaneous fermentation, 0.71 mg CE g⁻¹ FW for L. rhamnosus GG and 0.56 mg CE g⁻¹ FW for L. plantarum 229v. However, by day 7, spontaneous fermentation showed a significant increase in catechin equivalent content (2.30 mg CE g⁻¹ FW), significantly higher than the values observed in the L. rhamnosus GG and L. plantarum 229v treatments. This increase could be attributed to the enzymatic activities of the variety of spontaneous microbiota, which can release phenolic compounds from plant cell walls [28]. On day 14, spontaneous fermentation maintained the highest catechin equivalent content, with levels of 2.43 mg CE g⁻¹ FW, while inoculated treatments continued to show significantly lower values. From day 21 to day 28, an overall decrease in antioxidant activity was recorded in all treatments. This trend could reflect oxidative degradation or microbial metabolism of catechins and other phenolic compounds during fermentation, a phenomenon previously described in both vegetable and beverage fermentations [29]. At the end of the experimental period (day 42), all treatments showed a recovery in antioxidant activity values, ranging from 2.90 to 3.50 mg CE g⁻¹ FW. These findings suggest that while spontaneous fermentation may initially enhance phenolic compounds release, the long-term dynamics of phenolic content are governed by a combination of microbial activity, enzymatic action and compound stability throughout the fermentation process [30].

During the early stages of fermentation at 25 °C (Figure 3b), antioxidant activity values remained below 1 mg CE g⁻¹ FW in all treatments. From day 7 onwards, a clear divergence emerged: spontaneous fermentation showed a significantly higher increase in antioxidant activity compared to the treatment inoculated with L. plantarum 299v. On day 14, this trend intensified, with spontaneous fermentation reaching the highest concentration of catechin equivalent content (1.43 mg CE g⁻¹ FW1), significantly exceeding the levels recorded in both inoculated treatments (p < 0.05). This suggests that the metabolic heterogeneity and enzymatic diversity inherent to the native microbiota may play a key role in the increased release of antioxidant compounds, as supported by studies on phenolic transformation during spontaneous fermentation processes [31]. These results agree with those obtained in the study of microbial changes, where it was reported that the initial stage of fermentation was dominated by native microbiota. From day 21 onwards, the differences between groups decreased, although antioxidant activity remained relatively stable, indicating a possible stabilisation phase in microbial activity or a plateau in phenolic extraction. At the end of the fermentation process (day 42), the L. rhamnosus GG treatment showed the highest catechin equivalent content (2.14 mg CE g⁻¹ FW), which was significantly higher than that observed in the L. plantarum 229v treatment (0.17 mg CE g⁻¹ FW). These results suggest strain-dependent variability in phenolic metabolism and indicate the possibility of late-stage enzyme activation by L. rhamnosus GG. This is in agreement with previous studies reporting an increased antioxidant capacity in vegetable juices fermented with L. rhamnosus GG under controlled conditions [32].

These results underline the importance of both microbial composition and fermentation dynamics on the bioavailability of functional compounds. Furthermore, others authors have compared the cabbage fermentation with other cooking methods, observing that this procedure is the best at maintaining, or increasing, the antioxidant activity of this vegetable [33].

3.4. Determination of Folate Results

3.4.1. Determination of Total Folate Content

As shown in Figure 4, fermentation carried out at 18 °C showed that the total folate content (TFC) was initially similar in the three experiments: 1361.8 µg 100 g⁻¹ in spontaneous fermentation, 1242.2 µg 100 g⁻¹ with L. rhamnosus GG and 1300.5 µg 100 g⁻¹ with L. plantarum 229v. A significant reduction in total folate was observed during the first 14 days. However, at 42 days, spontaneous fermentation retained significantly more total folate, with values of 878.9 µg 100 g⁻¹, compared to the values obtained with L. rhamnosus GG (508.5 µg 100 g⁻¹) and fermentation with L. plantarum 229v (519.6 µg 100 g⁻¹). Treatment with L. plantarum 229v showed the greatest reduction in folate, indicating that under these conditions, this strain may be consuming more folate or that its synthesis capacity is limited. L. rhamnosus GG showed intermediate values and a partial recovery towards the end of the process.

At 25 °C, a more pronounced decrease in TFC was observed during the first few days of fermentation, especially in the L. rhamnosus GG treatment. On day 0, all treatments started with values of 1051.840 CFT µg/100 g, 1071.495 CFT µg/100 g and 1330.267 CFT µg/100 g for the spontaneous, L. rhamnosus and L. plantarum fermentation, respectively. All treatments showed an initial decrease, but from day 21 another small increase was observed until reaching almost the initial levels (985,221 TFC µg/100 g, 1,053,726 TFC µg/100 g and 841 TFC µg/100 g for spontaneous, L. rhamnosus and L. plantarum fermentation, respectively). Thereafter, relatively higher folate levels were maintained compared to the other treatments. These results are consistent with the known potential of L. plantarum for folate production, which can synthesise it when supplied with precursors such as para-aminobenzoic acid [34,35]. In contrast, L. rhamnosus lacks folate biosynthetic pathways and depends on exogenous sources, particularly plant substrates [36]. These results confirm that the use of specific microbial strains in SK fermentation can lead to a substantial increase in folate values, consistent with observations made in other fermented food matrices [37] such as wine, beer and bread, where yeast acts as a rich source of folates [38], and in plant products such as fermented soy products, including tempeh, which uses Rhizopus as a fermentation microorganism, producing a higher folate content [39].

3.4.2. Determination of Folate Isoforms

As shown in

Table 1. Evolution of folate isoforms at a temperature of 18 °C in the three sauerkraut (SK) fermentation treatments: spontaneous, L. rhamnosus GG and L. plantarum 229v.

		Day
			0	7	14	21	28	42
18 °C	Spontaneous	FA	37.3 a ± 5.10	17.0 a ± 14.9	24.6 a ± 1.2	26.9 a ± 1.5	17.0 a ± 2.3	20.2 a ± 0.6
		DHF	375.8 a ± 103.6	152.0 a ± 57.5	243.9 a ± 21.2	219.7 a ± 25.1	206.8 a ± 49.0	166.5 a ± 11.7
		THF	490.6 a ± 49.0	340.1 a ± 69.0	311.0 a ± 11.9	340.5 a ± 5.8	183.2 a ± 10.3	277.2 a ± 13.1
		5-MTHF	294.4 a ± 47.6	175.3 a ± 55.5	290.2 a ± 128.3	323.8 a ± 67.8	383.2 a ± 91.3	248.0 a ± 7.3
		5-FTHF	117.9 a ± 24.3	57.3 a ± 26.2	79.0 a ± 22.1	74.4 a ± 16.6	88.7 a ± 8.2	82.9 a ± 4.6
					Day
			0	7	14	21	28	42
	L. rhamnosus GG	FA	37.7 a ± 5.30	20.4 a ± 4.8	20.8 b ± 0.7	21.4 a ± 5.1	11.0 b ± 0.6	18.5 a ± 1.8
		DHF	366.6 a ± 159.3	228.6 a ± 69.4	212.0 a ± 7.6	239.8 a ± 44.0	128.1 b ± 2.8	173.2 a ± 20.8
		THF	476.3 a ± 18.1	362.3 a ± 22.1	258.8 b ± 8.3	357.2 a ± 24.8	117.7 b ± 9.3	195.8 b ± 6.1
		5MTHF	261.1 a ± 17.1	229.9 a ± 15.0	299.8 a ± 12.3	363.0 a ± 100.8	199.0 b ± 8.5	229.7 a ± 23.5
		5FTHF	100.5 a ± 15.8	77.6 a ± 6.3	81.0 a ± 7.3	72.4 a ± 11.5	52.8 b ± 6.1	55.6 b ± 2.4
					Day
			0	7	14	21	28	42
	L. plantarum 229v	FA	39.1 a ± 5.3	26.8 a ± 9.4	25.5 a ± 0.9	28.6 a ± 5.6	10.6 b ± 0.4	14.6 b ± 1.4
		DHF	372.8 a ± 217.3	195.3 a ± 31.5	249.3 a ± 24.7	198.6 a ± 22.1	133.0 b ± 14.5	126.2 b ± 8.2
		THF	492.7 a ± 96.4	285.1 a ± 24.7	259.9 b ± 10.0	324.6 a ± 25.6	169.7 b ± 7.4	187.2 b ± 6.3
		5MTHF	299.7 a ± 81.9	211.8 a ± 4.8	279.1 a ± 7.2	217.3 a ± 3.9	157.3 b ± 7.7	163.3 b ± 10.9
		5FTHF	96.2 a ± 43.3	66.3 a ± 7.9	79.6 a ± 1.5	72.5 a ± 1.7	48.1 a ± 4.6	43.0 c ± 3.1

Letters (a–c) denote statistically significant differences (p < 0.05) between isoform values compared for each temperature, fermentation type and day. FA: folic acid; DHF: dihydrofolate; THF: tetrahydrofolate; 5MTHF: 5-methyl tetrahydrofolate; 5FTHF: 5-formyl tetrahydrofolate.

, at 18 °C, there is a significant reduction in each isoform during the first 14 days, followed by partial recovery or stabilisation around day 28 of fermentation, but without reaching the initial values. In fact, by day 14, FA values showed a significantly higher concentration of L. plantarum (25 µg/100 g) and spontaneous fermentation (24.6 µg 100 g⁻¹) compared to L. rhamnosus GG (20.7 µg 100 g⁻¹) (p < 0.05). These differences were particularly marked on days 28 and 42, when spontaneous fermentation significantly increased its FA levels, exceeding both treatments with starter cultures (p < 0.05). However, at 42 days, spontaneous fermentation retained significantly more total folates, with values of 878.9 µg 100 g⁻¹ compared to the values obtained with L. rhamnosus GG (508.5 µg 100 g⁻¹) and fermentation with L. plantarum 229v (519.6 µg 100 g⁻¹), differences that were statistically significant (p < 0.05). This trend was also reflected in the main isoforms, where spontaneous fermentation showed high values of tetrahydrofolate (THF) (183.2 µg 100 g⁻¹) and 5-methyl tetrahydrofolate (5-MTHF) (383.2 µg 100 g⁻¹) at day 28, being a significantly higher value (p < 0.05) compared to the treatments with both inocula. These behaviours suggest a potential microbial synthesis of 5MTHF during the later stages of fermentation in the spontaneous sample, as well as folate consumption in both probiotics’ inocula. This hypothesis is supported by previous reports indicating that certain indigenous LAB, commonly present in vegetable fermentation, possess the enzymatic capacity to convert folate precursors into 5-MTHF. Such microbial activity typically intensifies as the fermentation progresses and the microbial ecosystem stabilises, potentially leading to an increased accumulation of this bioactive folate isoform [40].

As shown in

Table 2. Evolution of folate isoforms at a temperature of 25 °C in the three sauerkraut (SK) fermentation treatments: spontaneous, L. rhamnosus GG and L. plantarum 229v.

		Day
			0	7	14	21	28	42
25 °C	Spontaneous	FA	28.8 a ± 9.5	20.1 b ± 1.6	13.6 b ± 2.1	25.8 b ± 2.5	10.3 b ± 1.0	18.1 a ± 1.4
		DHF	433.0 a ± 108.3	190.3 a ± 0.1	130.1 ac ± 9.1	233.2 a ± 8.6	100.8 c ± 7.3	140.0 a ± 4.5
		THF	547.9 a ± 125.0	284.3 a ± 35.3	141.8 c ± 7.8	332.0 a ± 15.1	116.8 b ± 6.6	214.3 a ± 12.1
		5MTHF	285.8 a ± 23.0	231.5 a ± 32.5	296.0 a ± 4.4	285.6 a ± 99.6	156.2 b ± 6.9	176.6 a ± 5.1
		5FTHF	86.3 a ± 18.2	62.5 a ± 2.9	73.5 a ± 8.2	74.7 a ± 13.3	48.7 b ± 7.3	40.2 a ± 1.0
					Day
			0	7	14	21	28	42
	L. rhamnosus GG	FA	31.7 a ± 4.3	11.2 c ± 0.1	22.8 a ± 1.8	20.7 b ± 0.9	9.0 b ± 0.1	14.0 b ± 2.7
		DHF	248.3 a ± 20.9	116.4 b ± 14	211.5 a ± 14.5	181.6 b ± 10.7	119.6 b ± 7.1	140.7 a ± 32.0
		THF	453.8 a ± 66.6	115.9 a ± 8.3	236.1 a ± 8.6	263.5 b ± 7.2	121.2 b ± 12.6	157.7 b ± 5.7
		5MTHF	259.3 a ± 25.4	177.0 a ± 21.9	289.3 a ± 10.8	208.4 a ± 3.2	211.8 b ± 42.3	180.3 a ± 22.2
		5FTHF	78.4 a ± 5.0	51.2 a ± 5.2	77.2 a ± 4.6	62.6 a ± 2.8	43.3 b ± 1.4	42.6 a ± 6.3
					Day
			0	7	14	21	28	42
	L. plantarum 229v	FA	38.5 a ± 10.3	24.8 a ± 0.0	19.8 a ± 2.7	22.2 a ± 3.7	16.5 a ± 1.7	8.4 c ± 1.0
		DHF	423.5 a ± 191.1	146.2 ab ± 8.9	166.9 b ± 13.6	219.2 ab ± 33.8	166.8 a ± 5.7	98.0 a ± 15.5
		THF	485.3 a ± 99.4	359.6 a ± 3.9	201.0 b ± 7.8	339.3 a ± 35.3	177.4 a ± 18.2	58.2 c ± 7.3
		5MTHF	282.9 a ± 68.3	149.0 a ± 3.3	280.2 a ± 9.4	275.4 a ± 24.4	292.0 a ± 26.9	135.1 b ± 4.5
		5FTHF	100.1 a ± 31.8	74.9 a ± 8.5	81.5 a ± 8.5	75.2 a ± 17.2	84.4 a ± 11.6	36.8 a ± 2.5

Letters (a–c) denote statistically significant differences (p < 0.05) between isoform values compared for each temperature, fermentation type and day. FA: folic acid; DHF: dihydrofolate; THF: tetrahydrofolate; 5MTHF: 5-methyl tetrahydrofolate; 5FTHF: 5-formyl tetrahydrofolate.

, during SK fermentation at 25 °C, significant differences were observed in the evolution of their main isoforms. At the start, total folate values were high in all treatments (1051.8 µg 100 g⁻¹ for spontaneous, 1071.5 µg 100 g⁻¹ for L. rhamnosus GG and 1330.3 µg 100 g⁻¹ for L. plantarum 229v), with a greater loss in the first 14 days of the experiments. On day 7, L. plantarum 229v showed a significant increase in FA, reaching the highest value observed throughout the experiment, surpassing the other two fermentation treatments (p < 0.05), and although concentrations decreased towards day 28, L. plantarum 229v maintained higher FA levels (p < 0.05). The study carried out by Sybesma et al., 2003 [36], showed that strains such as L. plantarum 229v can synthesise folates. The isoforms with the highest values are THF and DHF, which are predominant at the start of fermentation, but both decrease significantly throughout the process. The 5-MTHF isoform, considered metabolically relevant, remains more stable throughout the experiment, with values around 130–280 µg 100 g⁻¹ at the end of the process, depending on the inoculum, especially in spontaneous fermentation, where it reached final values of 176.6 µg 100 g⁻¹, compared to 180.3 µg 100 g⁻¹ for L. rhamnosus GG and 135.1 µg 100 g⁻¹ for L. plantarum 229v (p < 0.05).

3.5. Sensory Evaluation Results

Table 3. Sensory attributes scores (appearance, overall evaluation, acidity, colour, flavour, texture and aroma) at a temperature of 25 °C and 18 °C in the three sauerkraut (SK) fermentation treatments: spontaneous, L. rhamnosus GG and L. plantarum 229v.

	Parameter	Spontaneous	L. rhamnosus GG	L. plantarum 229v
18 °C	Appearance	6.2 a ± 0.45	4.8 b ± 0.45	5.8 ab ± 0.84
	Aroma	6.2 a ± 0.84	5.4 a ± 0.55	5.0 a ± 0.71
	Texture	5.4 a ± 0.55	5.2 a ± 0.45	5.2 a ± 0.45
	Flavour	5.8 a ± 0.45	5.4 a ± 0.89	4.2 b ± 0.45
	Colour	6.0 a ± 0.71	4.6 b ± 5.8	5.6 ab ± 0.55
	Acidity	6.2 a ± 0.84	5.8 a ± 0.84	4.6 a ± 2.19
	Overall acceptability	6.2 a ± 0.84	5.3 a ± 1.30	5.5 a ± 2.55
	Parameter	Spontaneous	L. rhamnosus GG	L. plantarum 229v
25 °C	Appearance	5.8 a ± 0.45	5.4 a ± 1.14	6 a ± 1.41
	Aroma	5.4 a ± 0.89	4.8 a ± 0.45	4.6 a ± 0.89
	Texture	5.6 a ± 0.55	5.2 a ± 0.45	5.4 a ± 0.55
	Flavour	4.8 a ± 0.45	5 a ± 1.00	3.6 a ± 1.14
	Colour	5.4 a ± 0.55	5.4 a ± 0.55	5.2 a ± 0.84
	Acidity	4.6 a ± 0.55	3.8 a ± 0.84	4.0 a ± 0.71
	Overall acceptability	5.4 a ± 0.55	4.6 a ± 0.55	5.0 a ± 0.71

Letters (a–b) denote statistically significant differences (p < 0.05) in each sensory attributes (appearance, overall evaluation, acidity, colour, flavour, texture and aroma) between different fermentation treatments.

shows the mean scores for seven different sensory attributes (appearance, overall evaluation, acidity, colour, flavour, texture and aroma) at a temperature of 25 °C and 18 °C in the three sauerkraut (SK) fermentation treatments: spontaneous, L. rhamnosus GG and L. plantarum 229v. The radar chart (Figure 5) showed the sensory evaluation of SK fermented at 18 °C, and the highest score (6.3) was for spontaneous fermentation, followed by L. plantarum 229v (5.8) and L. rhamnosus GG (4.8). Differences between spontaneous fermentation and L. rhamnosus GG were significant (p < 0.05), indicating a visual preference for spontaneous fermentation. For aroma, spontaneous fermentation again yielded the highest score (6.2), ahead of L. rhamnosus GG (5.4) and L. plantarum 229v (5.0), with no statistically significant differences among treatments (p > 0.05). Regarding texture, all modalities showed closely similar values (range 5.2–5.4) without significant differences, suggesting that under the conditions evaluated, the plant matrix and/or baseline process dominated textural perception over the inoculation type. For flavour, spontaneous fermentation exhibited a moderate advantage (5.8) over L. rhamnosus GG (5.4) and L. plantarum 229v (4.2). The comparison between spontaneous fermentation and L. plantarum 229v was significant (p < 0.05), indicating an effect of the spontaneous process on gustatory complexity. For colour, spontaneous fermentation, again, received the highest score (6.0), followed by L. plantarum 229v (5.6) and L. rhamnosus GG (4.6), with significant differences between spontaneous fermentation and L. plantarum 229v (p < 0.05). Perceived acidity was higher in spontaneous fermentation (6.2) and L. rhamnosus GG (5.8) than in L. plantarum 229v (4.6), without significant differences (p > 0.05), suggesting statistically comparable acidity profiles despite the observed trend. The overall acceptability score ranked spontaneous fermentation highest (6.2), followed by L. plantarum 229v (5.5) and L. rhamnosus GG (5.3), with no significant differences (p > 0.05). These data indicate that spontaneous fermentation tends to favour several sensory attributes (appearance, aroma, flavour and colour), although only some differences reach statistical significance. This pattern supports the hypothesis that the greater microbial diversity inherent to spontaneous fermentation contributes to enhanced organoleptic complexity, particularly in aroma and flavour. According to Anumudu et al., 2024 [38], the microbial diversity inherent in spontaneous fermentation may contribute to greater organoleptic complexity due to the synergy between lactic acid bacteria and other microorganisms in the vegetable microbiome. However, the tendency towards slightly higher scores at 18 °C could be related to better preservation of aromatic compounds and more balanced acidity, aspects that have been highlighted in the literature as favourable for consumer perception, as it favours the slow development of indigenous lactic acid microorganisms, allowing for a more gradual evolution of volatile and aromatic compounds [41].

The sensory evaluation of SK fermented at 25 °C (Figure 6) in terms of appearance showed a higher value in L. plantarum 229v (6.0) compared to spontaneous fermentation (5.8) and L. rhamnosus GG (5.4). In terms of aroma, spontaneous fermentation showed a slightly higher value (5.4) than fermentation with L. rhamnosus GG (4.8) and L. plantarum 229v (4.6). In terms of texture, the three treatments had similar values between 5.6 and 5.2, with the highest value corresponding to spontaneous fermentation. In terms of flavour, spontaneous fermentation came out on top with a value of 4.8, followed by L. rhamnosus GG (5.0) and L. plantarum 229v with a lower value (3.6). In terms of colour, spontaneous fermentation scored highest with a value of 5.4, followed by L. plantarum 229v (5.2) and L. rhamnosus GG with a score of 5.4. In terms of acidity, spontaneous fermentation was rated highest by the panellists with a value of 4.6, followed by L. rhamnosus GG (3.8) and finally L. plantarum 229v with an average score of 4.0, without significant differences (p > 0.05). These results coincide with Xiong et al., 2012 [42], which shows that both spontaneous fermentation and fermentation directed by selected strains of lactic acid bacteria produce sauerkraut with high sensory acceptability when fermented at 25 °C, due to the greater stability of the medium for LAB to grow in an optimal environment for their growth. The overall acceptance scores were 5.4 for spontaneous fermentation, 4.6 for the treatment with L. rhamnosus GG and 5.0 for L. plantarum 229v. However, comparisons between treatments did not reveal significant differences between them (p > 0.05), suggesting that although spontaneous fermentation tends to receive a slightly higher rating, the sensory acceptance of the three treatments is comparable. These results are consistent with previous studies that have reported good sensory acceptance of cabbage fermented at moderate temperatures, allowing the development of pleasant organoleptic characteristics, as the indigenous microbiota can contribute organoleptic profiles that provide greater sensory complexity, increasing overall acceptance by consumers [39].

4. Conclusions

In the assayed conditions, the L. rhamnosus GG strain proved ineffective as a starter culture at both 18 °C and 25 °C, as it failed to adequately control the growth of competing microorganisms during the initial days of fermentation. In contrast, L. plantarum 229v achieved greater and more rapid acidification of the medium at both temperatures. Counts of moulds, yeasts and Enterobacteriaceae during the critical early phase of fermentation (days 1–7) confirmed the ability of L. plantarum 229v to suppress undesirable microbiota, supporting its suitability and effectiveness as a starter culture for SK.

Regarding total folate content (TFC), spontaneous fermentation was particularly effective in retaining TFC and biologically relevant isoforms such as 5-MTHF. However, L. plantarum 229v also showed high folate retention and is recognised in the literature as a folate-producing strain. At 25 °C, it exhibited superior performance during intermediate stages, with only a slight progressive folate loss towards the end of fermentation. By comparison, L. rhamnosus GG produced the lowest TFC values and isoform levels, particularly at the higher temperature. In terms of antioxidant activity, L. rhamnosus GG at 25 °C showed a significant increase in antioxidant compound concentration, comparable to spontaneous fermentation and higher than the other treatments. As for sensory analysis, spontaneous fermentation at 18 °C tends to maximise organoleptic complexity, especially in aroma and flavour, probably due to the action of a more diverse microbiota and a better aromatic balance. However, the fermentation with L. plantarum 229v at 25 °C achieved comparable overall sensory acceptance, making it a valid alternative when general acceptability is prioritised.

Overall, targeted fermentation with the probiotic strain L. plantarum 229v at 25 °C emerges as a promising strategy for producing SK with high folate content and greater antioxidant activity, maintaining a favourable organoleptic profile and effective microbial control that reduces the risk of proliferation of pathogenic or toxigenic microorganisms, including biogenic amine producers. These findings have relevant implications for the functional food industry.

Although spontaneous fermentation yielded the best results, fermentation with L. plantarum 229v can be a promising approach. The idea of promoting fermentation with the probiotic strain of L. plantarum 229v at 25 °C is interesting as it has proven to be an efficient strain to obtain a product with high folate content, antioxidant activity, organoleptic acceptability and efficient control during fermentation. It minimises the risk of growth of potentially pathogenic or toxigenic microorganisms, such as biogenic amine producers.