Microbial Profiling of Smear-Ripened Cheeses: Identification of Starter Cultures and Environmental Microbiota

Abstract

Smear-ripened cheeses are distinguished by their complex microbiota, which play an important role in ripening, flavour development, texture and microbiological safety. Although commercial production typically relies on defined starter and adjunct cultures, microorganisms from the production environment may also shape the product’s characteristics. This study examined the microbial composition of smear-ripened cheeses from six commercial manufacturers using culture and culture-independent techniques, including 16S rRNA gene sequencing and sequencing of internal transcribed spacers. A limited number of microorganisms was recovered by culture, with 37 different isolates identified across all samples. Sequencing of the 16S rRNA gene and internal transcribed spacers revealed 75 and 7 distinct operational taxonomic units, respectively. The microbiota composition reflected the contribution of both mesophilic and thermophilic starter and adjunct cultures, alongside microorganisms originating from the production environment. These included various psychrotrophic bacteria, marine (i.e., osmotolerant) bacteria, and other halophiles from Proteobacteria (Psychrobacter, Pseudoalteromonas, Marinomonas, and Vibrio), Firmicutes (Vagococcus and Marinilactibacillus), Actinobacteriota (Glutamicibacter), Bacteroidota (Winogradskyella and Brumimicrobium), Campylobacterota (Malaciobacter) and Fusobacteriota (Psychrilyobacter) specific to the environment of particular manufacturers. The results indicate that, although pasteurised milk and defined starter cultures are used in commercial production, microorganisms originating from the cheese factory environment form a substantial part of the microbiota of smear-ripened cheese.

1. Introduction

Microorganisms play a crucial role in the production of various cheeses, significantly influencing the development of their unique flavour, texture, aroma and visual characteristics. In modern commercial cheese production, the use of specific dairy cultures, which facilitate more standardised production processes, has become widespread. Starter cultures (SCs), which are added to the milk at the beginning of the process, are critical for acid production during the early stages of fermentation, establishing an acidic environment that effectively inhibits spoilage flora and gas-producing microorganisms and improves control over cheese quality [1]. SCs that are primarily composed of lactic acid bacteria (LAB) can be categorised into two main groups based on their optimal temperature activity: mesophilic cultures and thermophilic cultures. This classification is important for various types of cheese as temperature changes affect the behaviour of the bacteria and, thereby, the cheesemaking and ripening processes. Mesophilic cultures include, for example, Lactococcus lactis subsp. lactis, Lactococcus lactis subsp. cremoris, some Lactobacillus species and Leuconostoc mesenteroides [1,2] and are intended for basic dairy products, such as cottage cheeses, fresh cheeses, semi-hard cheeses, and smear-ripened cheeses [1]. In contrast, thermophilic cultures include Streptococcus salivarius subsp. thermophilus, Lactobacillus helveticus, and various subspecies of Lactobacillus delbrueckii and are primarily used in the production of high-heat cheeses, pasta filata-type cheeses, mozzarella, and Dutch-type cheeses [1,2].

In the production of smear-ripened cheeses, additional bacteria are deliberately applied to the cheese surface to develop the characteristic soft, moist rind [2]. Traditionally, a technique known as old-young smearing was used, which entails washing aged (old) cheese with a saline solution and subsequently using that solution to wash freshly made (young) cheese [2]. However, for safety reasons, this method has been replaced in commercial cheese production by using defined adjunct cultures of smear-associated bacteria. Commonly used adjunct bacteria include Brevibacterium linens (newly reclassified as Brevibacterium aurantiacum), Brevibacterium casei, Staphylococcus equorum, and occasionally Microbacterium gubbeenense, Glutamicibacter spp. or Corynebacterium casei [2,3]. These bacteria degrade proteins and fats in the cheese, leading to the production of aromatic compounds such as ammonia and sulphur, which contribute significantly to the distinctive aroma of the cheese. In addition to flavour enhancement, these bacteria also help soften the interior of the cheese, creating a creamy texture. Controlled application of adjunct bacterial cultures is crucial for achieving the characteristic appearance, including the reddish or orange rind, and for ensuring the safety and consistency of the cheese during the ripening process.

The development of surface microbiota is further supported by yeasts that are applied by spraying onto the cheese surface. Debaryomyces hansenii and other commonly used yeasts, such as Geotrichum candidum, Kluyveromyces marxianus, and Candida krusei, deacidify the cheese environment and thus promote the growth of smear bacteria [4]. Their proteolytic and lipolytic activity and the production of volatile compounds in yeast cultures also contribute to the organoleptic properties of the cheese [5].

In addition to starter and adjunct cultures, a variety of other microorganisms are present in the microbiota of smear-ripened cheeses [6,7,8]. Excluding pathogenic bacteria that may enter the production environment through raw milk or human carriers, these microorganisms, primarily found on the cheese surface, usually originate from the cheese plant environment, such as the brine bath or ageing surfaces [6,7]. This “house microbiota” is considered plant-specific [9,10] and may include various halophilic and psychrotolerant microorganisms, such as Psychrobacter spp., Vibrio spp., Pseudomonas spp., and Vagococcus spp. While their direct influence has not been fully investigated, it is thought that their metabolic activities may contribute to the distinctive characteristics of products from individual producers [5,8,11].

Smear-ripened cheeses are well-known and produced predominantly in European countries such as Austria, Belgium, Germany, and France. They include varieties such as Munster, Limburger, Tête de Moine, Reblochon, Romadur, and Tilsit. In this study, we focused on smear-ripened cheeses available in the Czech retail market, examining varieties originating from the Czech Republic, Germany, and France. The aim of this study was to explore the microbial composition of different smear-ripened cheeses, comparing the distribution of deliberately added starter and adjunct cultures and the occurrence of the putative house microbiota.

2. Materials and Methods

2.1. Sample Processing and DNA Extraction

Smear-ripened cheeses from six commercial manufacturers originating in the Czech Republic (CZ), Germany (DE), and France (FR) were purchased on the Czech retail market (

Table 1. A list of samples analysed in the study. CZ—Czech Republic; DE—Germany; FR—France. The protein, carbohydrate, fat, and salt contents (g/100 g) are given according to the manufacturer’s specifications.

Cheese	Country of Origin	Cheese Type	No of Samples	Proteins (g/100 g)	Carbohydrates (g/100 g)	Fat (g/100 g)	Salt (g/100 g)
A	CZ	Romadur	3	22	1.5	20	2
B	CZ	Quargel	3	28	2.6	0.5	4.5
C	CZ	Beer cheese	4	22.7	1.4	22	2.8
D	DE	Limburger	3	23	<0.5	19	2.3
E	FR	Munster	2	20	1	27	1.9
F	FR	Petit Tourtain	1	17	<0.5	33	1.2

). Samples from one production batch were analysed in one to four replicates. A total of 25 g of cheese was homogenised with 225 mL of phosphate-buffered saline (PBS) supplemented with 0.2% (v/w) Tween 80. Subsequently, 10 mL of the suspension was spun (12,000× g for 30 min. at 4 °C) and washed with PBS. The pellet was then subjected to DNA extraction. DNA extraction was performed using the DNeasy PowerFood kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. The initial homogenisation step was carried out using a MagNALyser (Roche, Rotkreuz, Switzerland). DNA concentration was determined spectrophotometrically and fluorometrically.

2.2. Microbiota Composition Determined by 16S rRNA Gene Sequencing and Sequencing of Internal Transcribed Spacer Regions

For 16S rRNA gene sequencing, DNA samples were diluted to 5 ng/mL and were used as a template in PCR with the eubacterial primers 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG-MID-GT-CCTACGGGNGGCWGCAG-3′ and 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG-MID-GT-GACTACHVGGGTATCTAATCC-3′, targeting the V3/V4 variable region of the 16S rRNA gene. Following amplification, the products were processed exactly as described previously [12]. Sequencing was performed using the MiSeq Reagent Kit v3 (600 cycles) and the MiSeq sequencing platform according to the manufacturer’s instructions (Illumina, San Diego, CA, USA). Raw reads were processed as previously described [12]. Sequencing of the internal transcribed spacer region ITS2a was carried out externally by Eurofins Genomics Europe Sequencing GmbH service (Konstanz, Germany) using the Illumina MiSeq platform. The raw reads were processed by the same facility.

2.3. Data Analysis and Statistics

Downstream analyses at the OTU (operational taxonomic unit) level were performed with OTUs that accounted for more than 0.1% of the total microbiota in at least one sample. These filtered OTUs covered 99.62–100.00% of the total bacteria identified by 16S rRNA sequencing and 99.95–100.00% of the total microbiota identified by ITS2a sequencing. Principal coordinate analysis (PCoA) followed by RM PERMANOVA (factors: manufacturer, replicate) on the Bray–Curtis dissimilarity matrix of all samples was performed by using function adonis2 (R-project 4.4.0, library vegan 2.6.10).

2.4. Total Bacteria Count Determination and Microbiota Cultivation

To estimate the total bacteria count, 10 g of cheese was homogenised in 90 mL of PBS, and serial dilutions were prepared. A 100 µL aliquot from each dilution was plated on PCA and BHI agar plates (plate count agar, brain-heart agar, Oxoid, Hampshire, UK) supplemented with 5% sodium chloride and incubated aerobically at 30 °C for 72 h. Total bacteria counts were calculated from two consecutive dilutions performed in duplicate. Morphologically distinct colonies were selected from an appropriate dilution of both media and identified using an AutoFlex Speed MALDI-TOF mass spectrometer (Bruker, Bremen, Germany). Isolates representing different species were stored in glycerol stocks at −20 °C for further analysis.

2.5. Sanger Sequencing of Full-Length 16S rRNA PCR Products

Representative bacterial isolates retrieved from the cheeses were characterised by 16S rRNA gene sequencing. Full-length 16S rRNA gene sequences were amplified from the extracted DNA using the primers UNCB1-5′-TGAAGAGTTTGATCATGGCTCAG-3′ and UNC2b-5′-AGGAGGTGATCCAGCCGCA-3′, followed by Sanger sequencing. The resulting sequences were compared to GenBank entries using BLAST+ 2.16.0 for taxonomic classification. Bacterial isolates were clustered based on their taxonomic relatedness using Clustal Omega 1.2.0 [13], and a final phylogenetic tree was constructed using iTOL v5.5.1 [14].

3. Results

3.1. Microbiota Composition of Smear-Ripened Cheeses Determined by 16S rRNA Gene Sequencing

A total of 75 OTUs were identified across all samples, with cheese B exhibiting the least complex microbiota (11 OTUs) and cheese D exhibiting the most complex microbiota (40 OTUs) (

Table 2. Number of OTUs obtained by 16S rRNA gene sequencing and the number of distinct bacterial isolates cultured from smear-ripened cheeses. OTU—operational taxonomic unit.

Cheese	No of OTUs	No of Isolates
A	32	20
B	11	7
C	23	10
D	40	6
E	23	7
F	19	4

). There were just 2 OTUs (Lactococcus lactis subsp. cremoris and Streptococcus thermophilus) shared by the microbiota of cheeses from all six producers. The microflora of cheese from six manufacturers was statistically significantly different (p < 0.01; RM PERMANOVA). The low similarity in microbiota composition was also confirmed by PCoA analysis, which clearly differentiated cheeses from different producers (Figure 1).

3.2. Bacterial Starter and Adjunct Cultures in Smear-Ripened Cheeses

The sequencing data recorded the presence of both mesophilic and thermophilic SCs in the analysed cheeses (Figure 2A). In cheese A, only the mesophilic L. lactis subsp. cremoris was detected as an SC, comprising 9.2% of the total microbiota. The microbiota of cheese B contained mesophilic SCs at abundances below 0.1%, while the thermophilic SCs consisting of S. thermophilus and L. delbrueckii were present at 18.1% and 2.6%, respectively. Additionally, thermophilic SCs were supplemented with Staphylococcus xylosus, which accounted for 15.6%. In cheese C, the predominant SCs were L. lactis subsp. cremoris and Lacticaseibacillus casei, with relative abundances of 15.9% and 20.5%, respectively. Minor contributions were made by Lactococcus laudensis, Leuconostoc, and S. xylosus, which were detected at 1.5%, 2.5%, and 4.4%, respectively. Cheese D exclusively contained the thermophilic S. thermophilus, which represented 21.6% of the total microbiota. In cheese E, L. lactis subsp. cremoris was the predominant SC, comprising 26.3%, while S. thermophilus and S. xylosus were present at low abundances of 1.4% and 1.3%, respectively. Finally, only mesophilic SCs were detected in cheese F, and these comprised L. lactis subsp. cremoris, L. laudensis and Leuconostoc, present at relative abundances of 15.7%, 8.8%, and 20.2%, respectively.

The relative abundance of the adjunct smear SC Brevibacterium aurantiacum varied widely among the cheese microbiota, ranging from 0.1% to 62.8% across five of the six producers of smear-ripened cheeses.

3.3. Bacteria of the Production Environment in Smear-Ripened Cheeses

Bacteria presumed to originate from the production environment accounted for varying proportions of the total microbiota, ranging from 0.3% to 88.9% (Figure 2B). These bacteria were represented by six phyla: Actinobacteriota, Proteobacteria, Firmicutes, Bacteroidota, Campylobacterota, and Fusobacteriota (Figure 2A).

The microbiota of cheese A exhibited the highest proportion of microorganisms originating from the environment (88.9%) of all the cheeses analysed. These bacteria were composed predominantly of Proteobacteria, of which the three genera, Psychrobacter, Pseudoalteromonas, and Vibrio, comprised 52.0%, 6.5%, and 8.2% of the total microbiota, respectively. The phylum Firmicutes was represented by two genera, Vagococcus and Marinilactibacillus, which accounted for 6.9% and 3.5% of the total microbiota, respectively. Other abundant taxa included Psychroflexus (Bacteroidota), Malaciobacter marinus (Campylobacterota), and Psychrilyobacter (Fusobacteriota), contributing 1.5%, 7.0%, and 1.8%, respectively, to the total microbiota. Cheese B exhibited almost no contribution of microbiota from the production environment, consisting exclusively of SC bacteria and the smear-associated bacterium B. aurantiacum. In cheese C, the microbiota from the production environment were again dominated by Proteobacteria, particularly Psychrobacter, which comprised 43.4%, and Celerinatantimonas, which accounted for 3.8%. Of the Gram-positive bacteria, Marinilactibacillus (Firmicutes) and smear-associated Corynebacterium variabile (Actinobacteriota) were present at 3.7% and 3.2%, respectively. In cheese D, Proteobacteria were represented by five genera, including Psychrobacter, Pseudoalteromonas, Marinomonas, Vibrio, and Cobetia, exhibiting relative abundances of 10.8%, 9.0%, 5.6%, 2.3%, and 5.8%, respectively. The Firmicutes were not represented by any bacterial taxon. The Actinobacteriota were represented by Glutamicibacter arilaitensis, which comprised 10.4% of the total microbiota. Bacteria from the phylum Bacteroidota, specifically Winogradskyella vidalii and Brumimicrobium, were substantially represented, contributing 11.8% and 8.1%, respectively. In cheese E, the microbiota from the production environment was composed of Proteobacteria, with Psychrobacter contributing 18.6% and Pseudoalteromonas accounting for 9.8%. The phylum Firmicutes was represented by Vagococcus, which accounted for 18.3%, while the phylum Actinobacteriota was predominantly represented by the smear-associated bacterium G. arilaitensis, which comprised 20.3% of the total microbiota. Finally, in cheese F, the microbiota of the production environment was almost exclusively composed of Proteobacteria, of which various species of Pseudoalteromonas, Marinomonas, and Vibrio were present at abundances of 36.8%, 14.7% and 2.8%.

3.4. Microbiota Composition Determined by Sequencing of Internal Transcribed Spacer Regions

A total of 7 OTUs were identified by sequencing of the ITS2 region across all samples—Debaryomyces hansenii represented by three different OTUs, Kluyveromyces marxianus, Yamadazyma triangularis, Pichia kudriavzevii, and Saccharomyces cerevisiae. Of these, two OTUs of D. hansenii and K. marxianus clearly dominated in individual cheese samples (Figure 3). Other yeast species, including Y. triangularis, P. kudriavzevii, and S. cerevisiae, were detected at an abundance lower than 1% and were only present in cheeses A and B. D. hansenii comprised 99.63–100% of the yeast population in samples from five of the six producers. Only the cheese from manufacturer B exhibited similar proportions of the two OTUs of D. hansenii and K. marxianus (Figure 3).

3.5. Microbiota Composition Determined by Cultivation

By cultivation, the total number of bacteria in the cheese samples ranged between 10⁷ and 10⁹ CFUs per gram of the cheese, with the lowest number in cheese D and the highest in cheese B (

Table 3. Total bacteria counts determined in the samples of smear-ripened cheese. CFU—colony forming unit.

Cheese	CFU/g
A	1.3–2.3 × 108
B	1.2–2.8 × 109
C	2.9–4.0 × 108
D	1.7–3.9 × 107
E	6.9–8.4 × 108
F	2.7 × 108

).

A total of 72 isolates were obtained in pure culture from all samples that were identified as distinct using MALDI-TOF mass spectrometry. Approximately half of the isolates exhibited identical 16S rRNA sequences, reducing the final number of unique isolates to 37 (Figure 4). The isolates were classified into four phyla—Firmicutes (15 isolates), Actinobacteriota (7 isolates), Proteobacteria (14 isolates), and Pseudomonadota (1 isolate). Mesophilic SCs of L. lactis subsp. lactis, L. lactis subsp. cremoris, L. laudensis, and Leuconostoc were cultured from cheeses A, C, E, and F, but not from producers B and D. Other representatives of the phylum Firmicutes included various species of the genera Staphylococcus and Vagococcus, along with one isolate of Marinilactibacillus found in cheeses from all producers except F. Brevibacterium, Glutamicibacter, Micrococcus, and Corynebacterium, all from the phylum Actinobacteriota, were also successfully cultured. The phylum Proteobacteria was represented by various species of the genera Vibrio and Psychrobacter. In contrast to the 16S rRNA gene amplicon sequencing results, species such as Morganella, Shewanella, Enterobacter, and Citrobacter were additionally cultured from cheeses A and E. Sphingomonas paucimobilis (the phylum Pseudomonadota), not detected by amplicon sequencing, was cultured from cheese C. Not a single isolate of the phylum Bacteroidota, which was abundant in cheese D, was captured by culture.

4. Discussion

The microbiota composition of smear-ripened cheeses from six commercial producers from the Czech Republic, Germany, and France was investigated in detail. 16S rRNA gene sequencing results indicated that each cheese exhibited a unique microbiota composition. Notably, only two bacterial species, likely originating from the starter culture, were shared across all six producers. A similar observation was reported by Mounier et al. [6], who analysed the microbial composition of surface microbiota cultivated from four Irish farmhouse smear-ripened cheeses. Despite analysing the same types of cheese, only two bacterial species—C. casei and G. arilaitensis—were consistently found across all samples.

The presence of various starter cultures was confirmed, with individual producers showing a distinct preference for specific strains. To illustrate, mesophilic LABs were predominantly used by producers A, C, E and F and thermophilic LABs were predominantly employed by producers B and D. Mesophilic SC are used to obtain soft or semi-hard smear-ripened cheese varieties, which are characterised by high moisture content and a soft or semi-hard texture. In contrast, thermophilic SCs are employed for harder smear-ripened cheese varieties, which also require heating the milk to higher temperatures and pressing the curd at higher pressure after syneresis, which results in a lower moisture content and a firmer texture [1,2]. Although we did not measure the moisture content, cheese B is classified as a harder smear-ripened cheese variety.

The term smear-ripened cheese refers to the layer of smear that develops on the cheese surface due to the activity of smear-associated bacteria when the young cheese is exposed to air at high relative humidity and suitable temperatures [6]. Nowadays, commercially available smear culture B. aurantiacum is the most widely employed [3,4,15]. The relative abundance of B. aurantiacum varied significantly across the cheeses from the six producers. This finding may suggest differences in production practices, for example, removing the smear or the possibility that individual strains of deliberately added bacteria may not always successfully colonise the cheese surface during the ripening process. Several findings were reported for B. linens on French soft red-smear cheeses and B. aurantiacum strains on Époisses cheeses, where they were rather outcompeted by adventitious strains [15,16], similarly as we observed with the cheeses A, C and F.

As part of the adjunct cultures, yeasts are applied to the surface of the smear-ripened cheese to ensure deamination and production of growth factors such as vitamins, pantothenic acid, niacin, or riboflavin to promote smear-associated bacteria [17,18,19]. The results of ITS sequencing indicated that D. hansenii was the most frequently used yeast culture in the production of smear-ripened cheese by all six producers. In cheese B, thermotolerant K. marxianus was further employed. Debaryomyces hansenii was described as being associated with the production of alcohol and carboxylic acids [20], while K. marxianus was linked to the production of fruity ester flavours [3], thus influencing the cheese flavour.

In addition to the defined starter and adjunct microbial cultures, microorganisms derived from the production environment can also make a substantial part of cheese microbiota and thus contribute to the organoleptic characteristics of products [2,8,11,21]. These bacteria naturally appear during the ripening and are derived from the brine, as well as from the wooden shelves, the air, and the environment of the ripening cellars [6,7,8]. We have shown that the house microbiota predominated over the deliberately added starter and adjunct cultures in the cheese microbiota of five producers. The only exception was cheese B, whose microbiota consisted exclusively of bacteria of starter and adjunct cultures. This finding is noteworthy; however, the question remains as to whether this observation can be attributed to effective hygiene practices or to the significantly higher salinity of the product that may affect microbial growth.

Proteobacteria were the most abundant bacteria detected by 16S rRNA gene sequencing in the total microbiota in cheeses from five of the six producers. The main representatives included psychrotrophic bacteria such as Psychrobacter and Pseudoalteromonas, “marine bacteria” such as Marinomonas, various species of the genus Vibrio, as well as other Proteobacteria such as Cobetia and Celerinatantimonas. It has been shown that Psychrobacter can significantly reduce the bacterial community and therefore, successfully implant itself into cheese, regardless of its inoculation level [7,22]. This bacterium may contribute positively to flavour development by producing volatile aromatic compounds such as aldehydes, ketones and sulphur, thereby affecting the aromatic properties of the cheese [8,22,23]. Various Pseudoalteromonas species have been identified in soft and semi-hard pasteurised and unpasteurised cheeses, as well as on the surface of smear-ripened cheeses [24,25,26,27]. This bacterium has been found to possess cold-adapted enzymes that contribute to the flavour development of cheeses during ripening and storage at low temperatures [7,23]. Generally, managing the presence of psychrotrophic bacteria is essential in cheese production to ensure desirable sensory attributes, as some species may lead to spoilage by producing off-flavours [28]. Bacteria of the genus Vibrio have been reported to be among the most abundant species in Époisses cheese at the end of the ripening [16]. Similarly, Celerinatantimonas spp. and Cobetia marina were found in traditional cheese from south Brazil; however, their effect on cheese characteristics has not been described [29].

Proteobacteria were also among the bacteria that were captured with the highest frequency by culture methods. Consistent with the results of 16S rRNA gene sequencing, distinct isolates of the genus Vibrio and Psychrobacter were obtained. However, other isolated bacteria from the phylum Proteobacteria, such as Morganella, Shewanella, Enterobacter, and Citrobacter, were hardly detected by 16S rRNA gene sequencing. These bacteria are frequently isolated from the surfaces of food processing areas in the food industry, where they can easily survive in the form of biofilms, and their presence should be monitored [8,30].

Apart from starter cultures, the only representatives of the phylum Firmicutes identified by 16S rRNA gene sequencing were bacteria of the genera Vagococcus and Marinilactibacillus. Correspondingly, distinct isolates of Vagococcus and Marinilactibacillus were retrieved by cultivation. The presence of these bacteria in smear-ripened cheeses has already been reported [31]. It has also been shown experimentally that the defined isolate of Marinilactibacillus produced acetate from lactate during ripening in Brie-type model cheeses, and the Vagococcus lutrae isolate produced acetate from lactose under aerobic conditions, thereby influencing the properties of the model product [32,33].

Different strains of staphylococci were obtained by cultivation in our study. In addition to S. xylosus and S. equorum, which can be considered technologically utilised bacterial cultures, isolates of S. epidermidis, S. capitis, S. hominis, and S. saprophyticus were also recovered. Staphylococci are known to thrive in low pH environments and high salt concentrations, and the presence of certain strains in food products may indicate secondary contamination originating from human carriers [34,35]. Nevertheless, various strains of staphylococci are frequently isolated from ripened cheeses, which contribute to aroma and flavour development through the production of extracellular proteolytic and lipolytic enzymes [36,37,38]. Staphylococci also contribute to the establishment of other smear-associated microbiota, such as coryneform bacteria, by producing essential growth factors [37].

The term coryneform bacteria refers to a group of genera, including Corynebacterium, Brevibacterium, Microbacterium, and Glutamicibacter, which share a characteristic coryneform cell morphology [2]. These bacteria preferentially inhabit pH-neutral surface environments and contribute significantly to the flavour and colour of cheese, the latter through the production of carotenoids [3,39,40]. G. arilaitensis was found in high abundance in cheeses D and E, and C. variabile in cheese C. In agreement, different coryneform isolates were obtained by cultivation, further indicating their accessibility in the cheese microbiota. Recent sequencing of G. arilaitensis revealed a plethora of genes associated with lipase activities and pigment production [41]. In another study, the presence of G. arilaitensis on the cheese surface was linked to the production of alcohols, carboxylic acids and ketones [20]. The contribution of Corynebacterium to the cheese aroma appears to be minimal, as it shows only weak esterase and lipase activities [2,3].

Members of the phylum Bacteroidota were identified predominantly in cheese D (Winogradskyella vidalii and Brumimicrobium) and A (Psychroflexus), as revealed by 16S rRNA gene sequencing. No isolates of the Bacteroidota were retrieved by cultivation, reflecting the fact that Bacteroidota are typically anaerobic bacteria. Bacteroidota are not typically part of the microbiota of cheese, although increased counts of Bacteroidota have been identified as a cause of defective smear in samples of prepacked smear cheeses [8].

Representatives of other phyla included Malaciobacter marinus of the Campylobacterota and Psychrilyobacter of the Fusobacteriota, which constituted substantial portions of the total microbiota in cheeses from producer A. No isolates of these phyla were retrieved by cultivation. One isolate from the phylum Pseudomonadota, Sphingomonas paucimobilis, was cultured from cheese from producer C despite not being detected by amplicon sequencing. Strains of Malaciobacter spp. (previously known as Arcobacter) have been isolated from cow and water buffalo milk [42], sheep ricotta cheeses [43], and the dairy plant environment [44]. Psychrilyobacter spp. are commonly found in marine environments [45], and their presence has rarely been recorded in fermented foods [46] or the cheese production environment [47]. So far, the effect of their metabolism on the properties of cheese is not yet known.

The microbial composition of smear-ripened cheese analysed in this study generally showed consistent results between the cultivation-based and sequencing methods. Culture methods play an indispensable role in the detection and isolation of pathogenic microorganisms in food, serving as a baseline for microbiological analysis. However, they may not be an appropriate choice for the monitoring of the dynamics of complex microbial communities. It is estimated that approximately 25–50% of microbial species may remain uncultivable despite the extensive use of special cultivation media [8,48]. For this reason, sequencing techniques are preferentially used as they provide a more time-efficient, sensitive and specific method in microbiota profiling and monitoring the technological processes [8,49,50]. However, the combination of culture and sequencing methods is advantageous in addressing technological failures in cheese production by identifying the key microbial players involved and offering potential solutions for improving the production process.

5. Conclusions

Cheese can be regarded as a complex multi-microbial ecosystem in which microorganisms from starter cultures, adjunct cultures and the production environment interact to define the final characteristics of the product. This study revealed that each of the six producers employed distinct starter cultures, comprising either mesophilic, thermophilic, or a combination of both types of LABs. In contrast, the employment of Debaryomyces hansenii as an adjunct yeast culture was nearly ubiquitous, with only one exception among the producers. We have further shown that the proportion of environmental microbiota varied considerably between producers, with representatives of six phyla identified, including Gram-negative psychrotrophic and marine bacteria of the Proteobacteria, Gram-positive halophilic lactic acid bacteria of the Firmicutes, as well as various coryneform bacteria of the Actinobacteriota. Applying the sequencing of cheese microbiota to the dairy industry can bring substantial benefits in terms of maintaining consistent product quality and improving production efficiency. Additionally, it can allow producers to optimise the use of starter and adjunct cultures, promoting more efficient and tailored fermentation processes. The ability to create “microbial fingerprints” for different cheese varieties can also offer an innovative approach to ensure product authenticity and support the development of unique cheeses.