Academic Editor: Yasuhito Shimada
Background: Cheese microbiome plays a key role in determining the organoleptic and physico-chemical properties and may be also used as an authenticity tool for distinguishing probiotic cultures. Due to significant reduction of cell viability often witnessed during food production processes and storage, immobilization is proposed to ascertain high probiotic cell loads required to confer the potential health benefits. Hence, the aim of the present study was to investigate the effect of free or immobilized Lactiplantibacillus plantarum T571 on whey protein on feta cheese microbiome. Methods: Next-Generation Sequencing technology was used to investigate cheese microbiome. Cheese samples containing free or immobilized Lactiplantibacillus plantarum T571 (a wild type strain isolated from Feta cheese brine) on whey protein, along with products containing commercial starter culture, were analyzed. Results: The results showed a great diversity of bacteria and fungi genera among the samples. An increased presence of Lactobacillus OTUs in cheese with immobilized cells on whey protein was witnessed, highlighting the survival of the strain in the final product. The immobilized culture had also a significant impact on other genera, such as Lactococcus, Leuconostoc and Debaryomyces, which are associated with improved technological characteristics and health benefits. Conclusions: Enrichment of feta cheese with immobilized potential probiotics to secure cell viability consists of an industrial challenge and leads to distinct microbiome composition that may be used as a valuable food authenticity tool.
Nowadays, a growing interest in developing novel foods enriched with beneficial
microorganisms and proteins that promote human health is witnessed [1, 2, 3].
According to the latest definition of FAO/WHO, probiotics are defined as “viable
microorganisms (bacteria or yeasts) which, when administered in adequate amounts,
confer a health benefit on the host” [4]. Probiotic foods are products that
contain viable probiotic microorganisms in a suitable matrix and in adequate
concentration [5]. To induce the health benefits, probiotic products need to
contain a sufficient number of live cells, able to survive the transit through
the gastrointestinal (GI) tract and subsequently proliferate and colonize in the
gut [6, 7]. The food industry has adopted the recommended levels of 10
Incorporating probiotics into foods and ensuring high cell loads constitutes a real bottleneck for the food industry, not only because of cell interactions with food constituents, but also because of the severe conditions often employed during food processing and storage, as well as during the GI transit until they reach the desired site in the body. Such contingencies often lead to important losses in cell viability [7]. To overcome this deficiency, cell immobilization has been proposed to assure the active and functional form of probiotic cells and cell survival enhancement [8, 9]. In this vein, immobilized probiotics on milk and whey proteins have been successfully tested as adjunct cultures in cheeses [10, 11], as they are efficient natural vehicles for probiotic cells in food products, due to their structural and physicochemical properties [12, 13].
However, knowledge on the effect of the probiotic cultures on the food microbiome is missing. The impact of probiotics colonizing a food matrix depends not only on their ability to graft in the microbiota, but on sharing genes and metabolites, supporting challenged microbiota.
Until present, culture-based microbial methodologies were employed to explore the dynamics of a microbial community in foods, with critical limitations on detecting no-cultivable or no-abundant microorganisms during cheese ripening. Recently developed advanced methodologies, such as next-generation sequencing (NGS), are extremely useful because of the enhanced sequencing depth that can be achieved compared to previous technologies. To date, NGS methods have been applied mostly to describe the human microbiome, but they have also been used to describe a vast array of environmental and agricultural ecologies [14, 15]. This technology has also been applied for the investigation of cheese bacterial diversity [16, 17].
Greek Feta cheeses are white-brined products and their microbiota differ through time and depend on the environmental, processing and ripening conditions. Feta cheese microbiota are originating from starter or adjunct cultures, the milk microbial cultures of indigenous sheep and goat breeds if no pasteurized milk is used, and the secondary microflora. The starter lactic acid bacteria (SLAB) are inoculated cultures, involved in acid production during manufacture and contribute to the complexity of the ripening process. Secondary microorganisms are not responsible for acid production during manufacture, but generally play an important role during ripening, as these bacteria contribute to the lipolysis and proteolysis [18]. The secondary microbial flora consists of non-starter lactic acid bacteria (NSLAB), which grow internally and are often unique to specific cheese varieties. Specifically, the microbial ecology of cheese is based on the complex interaction among SLAB and NSLAB.
Hence, the aim of the present study was to investigate the effect of immobilized Lactiplantibacillus plantarum T571 (formerly known as Lactobacillus plantarum T571) [19] on whey protein on Greek feta cheese microbiome. L. plantarum T571 properties have been previously studied [20] and the immobilized culture has been tested as adjunct culture in pilot-scale feta production [21].
Lactiplantibacillus plantarum T571 was isolated from Feta cheese brine,
as described before [20]. The strain was retrieved from a –80 °C stock
culture in 10 mL of MRS broth (LABM, Lancashire, UK) and incubated overnight at
30 °C. A subculture was made in 10 mL MRS broth and incubated at 30
°C for 24 hours. For the manufacturing of probiotic Feta cheese, the
cells were extracted by centrifugation at 10,000 g for 5 minutes, washed twice
with 1/4 strength Ringer’s solution (LABM, Lancashire, UK), and resuspended in
Ringer’s solution to provide a final population of 10
Feta cheese samples were prepared in a dairy industry in Greece (Ecofarma Peloponnese Dairy Industry). In brief, pasteurized goat and ewe’s milk was inoculated with either the commercial Feta starter culture containing Lactococcus lactis subsp. lactis, Lactococcus lactis subsp. cremoris, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus helveticus, and Streptococcus thermophilus (FTC, control trial), or the aforementioned starter culture and L. plantarum T571 immobilized on whey protein (FTI) [10], or the aforementioned starter culture and L. plantarum T571 free cells (FTF) [21]. The inoculum of the cultures added to pasteurized milk was estimated at 7 log cfu/mL. The curd was sliced and poured into molds. Dry salt was also added, and the molds were turned over at regular intervals for 24 hours. The cheese was placed in 15 kg metal jars with brine (7% w/v NaCl) the next day, and ripening took place in two stages: (a) 12 days at 18 °C (1st ripening), and (b) 76 days at 4 °C (2nd ripening). Then, the cheese was sliced into 400 g pieces and stored in plastic containers with fresh brine (7% w/v NaCl) [21]. Samples were taken at the end of ripening and subjected to NGS analysis.
Total DNA was extracted using the NucleoSpin® Food
(MACHEREY-NAGEL GmbH & Co. KG, Germany), following the manufacturer’s
instructions. NGS was performed using MiSeq sequencing by MR DNA
(http://www.mrdnalab.com, Shallowater, TX, USA). The
V1-V3 region of the bacterial 16S rRNA gene was amplified from cheese genomic DNA
with 27F/519R primers (AGRGTTTGATCMTGGCTCAG/GTNTTACNGCGGCKGCTG) and the highly
variable Internal Transcribed Spacer (ITS) regions of fungi ITS1 and ITS2
sequences surrounding the 5.8S-coding sequence and situated between the Small
Subunit-coding sequence (SSU) and the Large Subunit-coding sequence (LSU) of the
ribosomal operon was amplified with the primers ITS1/ITS4
(CTTGGTCATTTAGAGGAAGTAA/TCCTCCGCTTATTGATATGC). Polymerase Chain Reaction (PCR)
amplification was performed using the HotStarTaq Plus Master Mix Kit (Qiagen,
Germatown, Maryland, USA), consisting of 30 cycles with the following steps: 94
°C for 3 minutes, 30 cycles of 94 °C for 30 seconds, 53
°C for 40 seconds, 72 °C for 1 minute, and the final elongation
step at 72 °C for 5 minutes. PCR products were then subjected to
electrophoresis in 2% agarose gel to confirm the amplification and to determine
the relative intensity of bands. Then, the amplicons were purified using Ampure
XP beads (Beckman Coulter, Brea, California, USA). Samples were subsequently
prepared for the illumina DNA library using MiSeq sequencing, following the
manufacturer’s guidelines. Procession of the sequencing data was held using a
proprietary analysis pipeline by MR DNA. Operational taxonomic units (OTUs) were
defined by clustering at 3% divergence (97% similarity) and the final OTUs were
taxonomically classified using BLASTn against a curated database derived from
RDPII and NCBI (http://www.ncbi.nlm.nih.gov, http://rdp.cme.msu.edu) and compiled into each
taxonomic level into both “counts” and “percentage” files. The analysis
identified 3332 bacterial and 421 fungal OTUs and low-abundance (
All experiments were carried out in duplicate. Significance was established at
p
Principal Component Analysis (PCA) was performed using the FactoMineR (https://doi.org/10.18637/jss.v025.i01) and factoextra (https://CRAN.R-project.org/package=factoextra) R packages, while PCA plots were constructed using tools from the corrplot (https://github.com/taiyun/corrplot) R package.
Shannon and Simpson indices were calculated to measure bacterial diversity
(Table 1). FTI replicates were more diverse compared to the replicates in FTF and
FTC (p
Principal Coordinate Analysis (PCA) constructed using FactoMineR, factoextra and corrplot R packages. Figure is showing clustering of the feta cheese samples based on bacterial diversity (duplicates are included) in genus level. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC, feta-type cheese control.
16S rRNA OTUs | FTI | FTF | FTC |
Shannon’s Index | 2.10 |
1.65 |
1.64 |
Simpson’s Index | 0.28 |
0.41 |
0.47 |
FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey
protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC,
feta-type cheese control. All data represent the mean |
At phyla level, Firmicutes was the most abundant (90–95% presence in
all samples) phylum in all cases (p
Bacterial phyla | FTI | FTF | FTC |
Actinobacteria | 0.18 |
0.18 |
0.22 |
Bacteroidetes | 2.26 |
2.00 |
2.60 |
Firmicutes | 94.55 |
90.18 |
93.72 |
Gemmatimonadetes | 0.01 |
0.02 |
0.03 |
Proteobacteria | 2.78 |
7.40 |
3.11 |
Tenericutes | 0.21 |
0.20 |
0.31 |
Verrucomicrobia | 0.01 |
0.02 |
0.02 |
FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey
protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC,
feta-type cheese control. All data represent the mean |
At genera level, Streptococcus was the most abundant genus in all
samples, with a relative abundance ranging 51–69%, and it was decreased in FTI
group compared to FTF and FTC samples (p
Relative abundances (%) of different bacterial genera, after 16S rRNA NGS sequencing in feta cheese samples. All data represent the mean values from two independent experiments. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC, feta-type cheese control.
Heatmap, constructed using heatmap.2 function from the gplots package, illustrating the normalized (logarithmic) counts of the bacterial OTUs (genus level) of the various samples. X axis presents the various independent samples shown individually, Y axis presents the different identified genera and their phylogenetic relationships and increasing abundances are indicated with yellow to red colors. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC, feta-type cheese control.
FTI samples had a tendency for elevated diversity as shown in Table 2. However,
Shannon and Simpson indices were at similar levels between all samples
(p
Principal Coordinate Analysis (PCA) constructed using FactoMineR, factoextra and corrplot R packages. Figure is showing clustering of the feta cheese samples based on fungi diversity (duplicates are included) in genus level. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with L. plantarum T571 free cells; FTC, feta-type cheese control.
ITS OTUs | FTI | FTF | FTC |
Shannon’s Index | 2.12 |
1.71 |
1.44 |
Simpson’s Index | 0.17 |
0.32 |
0.48 |
FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey
protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC,
feta-type cheese control. All data represent the mean |
Ascomycota was the most abundant (p
Fungal phyla | FTI | FTF | FTC |
Ascomycota | 97.43 |
90.86 |
98.16 |
Basidiomycota | 2.54 |
9.10 |
1.81 |
Chytridiomycota | 0.03 |
0.04 |
0.02 |
FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey
protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC,
feta-type cheese control. All data represent the mean |
At genera level, Cryptococcus was more abundant in FTF group compared
to FTI and FTC (p
Relative abundances (%) of different fungal genera, after ITS NGS sequencing in feta cheese samples. All data represent the mean values from two independent experiments. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC, feta-type cheese control.
Heatmap, constructed using heatmap.2 function from the gplots package, illustrating the normalized (logarithmic) counts of the fungal OTUs (genus level) of the various samples. X axis presents the various independent samples shown individually, Y axis presents the different identified genera and their phylogenetic relationships and increasing abundances are indicated with yellow to red colors. FTI, feta-type cheese with immobilized L. plantarum T571 cells on whey protein; FTF, feta-type cheese with free L. plantarum T571 cells; FTC, feta-type cheese control.
Food carriers are proposed to ensure the survival of probiotic cells during production, ripening, storage and passage through the GI tract, which involves exposure to low pH, due to the presence of acids in the stomach and bile salts in the small intestine [24]. Cheese owed to its high pH (4.4–4.6 for feta cheese), is regarded an excellent delivery product of viable probiotics into the human intestine, acting as a buffer against the acidic stomach environment. As a result of the dense matrix and relatively high fat content, a favorable environment for probiotic cell survival throughout GI tract transit can be achieved, providing probiotic bacteria protection.
Whey proteins have been previously tested as natural carriers for probiotic cells, and due to their structural and physicochemical properties, they can be used as a delivery vehicle [10, 11]. In this vein, immobilized L. plantarum T571 culture immobilized on whey protein was incorporated in feta cheese and the effect of the adjunct culture on the product’s microbiome was investigated. For comparison reasons, feta cheese with free L. plantarum T571 cells and feta cheese with no adjunct culture were also included in the study.
Each type of cheese has its own variety of microbes, making the cheese microbiome a dynamic ecosystem that evolves over time as the cheese ripens [25]. The results of our study revealed that the adjunct potential probiotic culture resulted in significant differences among the genera mapped after 16S rRNA and ITS sequencing for bacterial and fungal microbiota, respectively.
Bacterial and fungal phyla were in similar levels between all groups, with no differences attributed to probiotic supplementation, apart from an elevated abundance of Proteobacteria phylum in feta cheese with L. plantarum T571 free cells as compared to FTI and FTC samples; the relative abundance was in levels also reported elsewhere [26, 27]. Adjunct immobilized probiotic resulted in a significant increase in diversity, as calculated with Shannon and Simpson indices, compared to control samples. High values for diversity indicate more diverse communities [28]. Fungal diversity was not significant different between the different samples, although at genera level, major differences were highlighted.
16S rRNA sequencing revealed distinct alterations between the probiotic (FTI and FTF) and control (FTC) cheese samples. Streptococcus, which was employed as one of the starter cultures, was the most abundant genus in all cases, as noted before [27, 29]. However, in probiotic feta cheese with immobilized L. plantarum T571 cells, its presence was significantly lower compared to FTF and FTC samples, as a result of immobilized probiotic supplementation that enhanced the presence of other genera. Of note, all cheese products were prepared with pasteurized milk and commercial LAB starters and L. plantarum T571 was incorporated either as free or immobilized cells on whey protein in probiotic feta cheeses. As expected and stated elsewhere [27], the three specific species frequently used as SLAB, S. thermophilus, L. delbrueckii and L. lactis, were mapped with the highest abundances in their corresponding genera, although in varying percentages between the different samples.
The adjunct probiotic cultures had a significant effect on the relative LAB abundances. Supplementation of feta cheese with probiotic L. plantarum T571 (both immobilized and free) cells led to similar relative abundance of total Lactobacillus OTUs (approximately 18%). However, their presence was significantly higher compared to the control sample (FTC), underlying that probiotic addition resulted in an enhanced survival of the strain and increased the relative abundance of the total genus. Lactococcus and Lactobacillus spp. are considered SLAB cultures and their relative proportions can vary among the final products, as described previously [27, 30, 31]. Feta cheese with immobilized L. plantarum T571 cells on whey protein (FTI sample) was also characterized by an increased abundance of Lactococcus compared to FTF sample, but it ranged in similar levels with the control feta cheese (FTC sample), preserving a favorable SLAB population in a normal balance. Noticeably, lactococci are often employed in cheese making and are related to aroma development [32]. In a similar experiment, microbiological analysis in feta cheeses manufactured by immobilized L. casei cells on whey protein-revealed an elevated presence of lactococci (approximately 8.6 logcfu/g) and lactobacilli (approximately 8.7 logcfu/g), in accordance to our results [10].
L. plantarum is used extensively in food industry as microbial starter and constitutes a widespread member of the genus Lactobacillus, while specific strains have been employed to produce novel functional foods and beverages with improved nutritional and technological features [33, 34]. The genus has been shown to exert antimicrobial activity against pathogens and thus can be used to enhance safety and shelf-life of fermented foods [33].
Many L. plantarum cultures have been associated to vast a range of medical properties, including antioxidant, anticancer, anti-inflammatory, antiproliferative, anti-obesity, and antidiabetic capabilities [35, 36]. Recently, L. plantarum has been applied in medical fields for the treatment of various chronic and cardiovascular diseases, such as Alzheimer’s, Parkinson’s, diabetes, obesity, cancer, hypertension, urinogenital complications, liver disorders, etc. [35, 37]. Interestingly, several strains have been used in clinical trials; L. plantarum 299v (DSM 9843) provided effective symptom relief in irritable bowel syndrome [38], and eight-week supplementation with L. plantarum HAC01 significantly improved HbA1c levels relative to placebo in prediabetic subjects [39].
Bacteria belonging to Lactococcus genus produce an antagonistic extracellular peptide called nisin, which has wide antimicrobial activity against Gram-positive spoilage and pathogenic bacteria; yet some strains have been characterized as probiotics [40].
Immobilized L. plantarum T571 on whey protein led to increased Leuconostoc genus, considered as NSLAB, that was more abundant in FTI samples (compared to FTF and FTC groups), indicating that it was a result of the immobilized probiotic supplementation. Leuconostoc is associated with milk products, and in brine cheese production, it has been responsible for flavor development [41]. It exhibits antibacterial effect against bacteria that cause cheese spoilage. Additionally, Leuconostoc has been proven to inhibit pathogen growth and can be used as a safe probiotic for further research [42, 43]. Indeed, several strains belonging to Leuconostoc genus are regarded as potential probiotics owing to various health beneficial effects [44].
Carnobacterium, a lactic acid bacterium that is associated with soft
cheese was also elevated in FTI group [45], compared to FTF.
Carnobacterium produce antimicrobial peptides, known as bacteriocins,
and has been thoroughly investigated as a protective culture for preventing
growth of Listeria monocytogenes in fish and meat products [46].
Hafnia, a genus associated with cheese ripening [47] and
Psychrobacter that contributes positively to many technological cheese
characteristics, especially improving the aroma properties [48], were slightly
increased in FTI group compared to the other two groups. Hafnia produces
caseinolytic protease B (ClpB) protein, which was previously identified as a
conformational mimetic of
Mapping the fungal microbiome in different feta cheeses, striking differences were revealed, as well. Probiotic feta cheese with immobilized cells resulted in increased abundance of Debaryomyces, associated with the production of cheese flavor compounds [51]. Debaryomyces are potential biopreservative agents for use in eliminating the growth of the Aspergillus that produce food-contaminating mycotoxins like ochratoxin, which is associated with cancer [52]. The predominant genus in FTI group was Pichia, whereas Galactomyces was the predominant genus in probiotic feta cheese with free L. plantarum T571 cells (FTF sample). Pichia has been previously reported in cheese products [53, 54], while Galactomyces is one of the major genera detected in cheese [51]. Cryptococcus, related to improved organoleptic characteristics in cheese and usually found on mature cheeses [55], was increased in probiotic cheese with free L. plantarum T571 cells (FTF sample). These differences support that probiotic addition led to microbial profiles and genera associated of high quality and with sensory characteristics similar to the typical characteristics of feta cheese, as reported previously [21].
Incorporation of probiotics in cheeses is widely reported in literature [10, 56, 57, 58, 59]. Whey protein can act as a natural immobilization carrier [10, 11, 33] and is expected to create high added value in the food sector, establishing an economically feasible, marketable, and environmentally friendly process in a series of food industrial applications by a concurrent exploitation of whey, a by-product of the dairy industry, which is difficult to dispose or treat [10]. White brined soft feta cheese, is a traditional Greek product that can be enriched with beneficial cultures, considering the increasing industrial demands and consumers’ needs for the development of functional foods. Hence, the proposed technology that employs insertion of immobilized probiotic cultures on natural food carriers, able to produce feta cheese with distinct microbial composition may have a major impact on sustainable development. Immobilization of L. plantarum T571 on whey protein is expected to improve cell survival during digestion and release the probiotic cells in the large intestine, given that whey protein could function as a buffering agent in vivo, and protect ingested bacterial strains at the upper GI tract [9].
Although NGS technology consists a key tool for the investigation of complex microbial systems (i.e., cheese microbial flora) and can be used for the culture-independent study of food microbiota, compared to time-consuming microbiological techniques, there are vast limitations that have to be considered carefully. Microbe-driven changes during food processing involve intricate relationships and especially bacterial and fungal interactions within the life of a food constituent [17]. In our case, samples after milking and before milk pasteurization, as well as in each of the two stages of feta ripening period [30]and during product storage would probably give more insight on microbiota evolvement over time. A major limitation of 16S rRNA (and ITS) NGS method is that its sensitivity is enough to detect potentially significant alterations at genus-level at best [16], thereby variations at species level or changes in gene expression are ignored.
Immobilized L. plantarum T571 cells on whey protein were used as functional ingredient to enrich feta cheese and the effect of probiotic supplementation on food microbiome was investigated. Enrichment with immobilized probiotic cells led to diversity both in bacteria and fungi profiles, as revealed by NGS sequencing. The presence of Lactobacillus, Leuconostoc and Debaryomyces in high proportions in feta cheese with immobilized cells compared to the control samples, may lead to advanced technological characteristics and contribute to possible health effects. Incorporation of viable probiotic strains into traditional cheese products consists an industrial challenge presently under investigation and resulted in distinct microbiome composition that could be a key factor in determining food authenticity. However, more research is still required in the field.
GI, gastrointestinal; IPA, International Probiotics Association; NGS, next-generation sequencing; SLAB, starter lactic acid bacteria; NSLAB, non-starter lactic acid bacteria; ITS, Internal Trancribed Spacer; SSU, Small SubUnit-coding sequence; LSU, Large SubUnit-coding sequence; OTU, Operational taxonomic Units; PCA, Principal Component Analysis.
YK and CT designed the research study. GM, IP, AN, NC and KT performed the research. MEG, PK and TT provided help and advice on data analysis. GM, IP and KT analyzed the data. GM, IP and YK wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.
Not applicable.
Not applicable.
We acknowledge support of this work by the project “Infrastructure of Microbiome Applications in Food Systems-FOODBIOMES” (MIS 5047291), which is implemented under the Action “Regional Excellence in R&D Infrastructures”, funded by the Operational Programme “Competitiveness, Entrepreneurship and Innovation” (NSRF 2014-2020) and co-financed by Greece and the European Union (European Regional Development Fund).
The authors declare no conflict of interest. YK is serving as one of the Editorial Board members and Guest editors of this journal. MEG is serving as one of the Guest editors of this journal. We declare that YK and MEG had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to YS.
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.