3.1. Raw Materials
The characteristics of milk and buttermilk used for Camembert-type cheese production and the results of one-factor ANOVA analysis are presented in
Table 2. Buttermilk had a significantly higher pH than milk, but the materials did not differ in terms of titratable acidity. Buttermilk was characterized by lower fat and total solid content but higher total protein content than milk. Higher protein content indicates that it is reasonable to use buttermilk in cheese production. Analyzing the color of raw materials, we found that buttermilk had significantly lower values of all parameters. This material was less light and had a more distinct green-yellow shade than milk. According to Gassi et al. [
45], the composition and characteristics of sweet BM depend on the origin of the milk used for cream production, the parameters of cream heat treatment, and the technology of butter production, especially whether it is a slow or fast process. This is the reason for the differences in sweet buttermilk properties found in the literature. Govindasamy-Lucey et al. [
34] used sweet BM with 8.26% total solids, 2.66% protein, and 0.75% fat, which is distinctly lower than in our study. Morin et al. [
13] produced rennet gels from BM with 9.33% total solids, 3.11% protein, and 0.48% fat, which is also lower than our findings. Additionally, El-Dardiry et al. [
46] obtained sweet BM with lower total solids, fat, and protein content (9.6%, 0.67%, and 3.56%, respectively). Sweet BM with similar to our results’ total solids, fat, and lactic acid content (10.25, 1.33, and 0.14, respectively) was tested by Bahrami et al. [
31]; however, the protein content (3.05%) and pH (6.63) in their study were lower than our results.
3.3. Physicochemical Properties
During the first three weeks of ripening, the control cheese had a significantly higher total solid content than cheese samples with BM (
Figure 2). After 4 weeks of ripening, cheese BM20 had the highest total solid content (45.64%). The lower total solid content of cheese samples with BM that was observed during most of the study period indicates the water-holding capacity of buttermilk ingredients. The presence of MFGM compounds, especially PL, provides the capacity to retain and absorb whey in the cheese structure [
31]. PL and other MFGM fragments form complexes with casein and alter rennet-induced gel formation, resulting in a less dense gel structure and higher moisture content [
48]. Moreover, the moisture increase in samples with BM addition could be due to the presence of denatured whey proteins. The denaturation of whey proteins may occur during cream pasteurization, and these proteins are able to form cross-links with casein micelles, which strengthens the gel and reduces the extent of syneresis [
34,
48]. A higher moisture content in cheese with BM addition compared to cheese without BM was also observed by Govindasamy-Lucey et al. [
34], Morin et al. [
13], Bahrami et al. [
31], and Krebs et al. [
48].
A gradual increase in the total solid content was observed during the first two weeks in both products C and BM10 when the cheese ripened without packaging. Wrapping the cheese discs with polyethylene-coated paper inhibited water loss, leading to a decrease in the total solid content during the last two weeks. However, in the case of cheese BM20, total solids continued to increase even after packaging. The total solid content in the control sample is typical for Camembert made from milk and similar to the results obtained by Batty et al. [
44,
47], Bensmail et al. [
49], and Los et al. [
50].
Similarly to the total solids, the lowest fat content in all cheese variants was observed on the 1st day (
Figure 3). During the first two weeks, when the cheese ripened without packaging, the fat content increased. This increase is associated with the moisture loss of Camembert-type cheese [
50]. Comparing the samples, it can be observed that products BM10 and BM20 had significantly lower (
p < 0.05) fat content throughout the ripening period compared to the control, except on the 21st day, when cheese BM20 did not differ significantly from the control. This is presumably the result of lower fat content in buttermilk compared to milk (
Table 2). Lower fat content in cheese with BM addition was also observed by Bahrami et al. [
31]. Low levels of fat in buttermilk-enriched cheese products may be perceived as an advantage in terms of their lower calorie content. In the case of fat content in the control cheese, the results are consistent with those reported by other authors on Camembert cheese [
47,
49,
50].
The total protein content was measured at the beginning and end of the ripening period. Directly after production, the protein content ranged from 11.09% to 13.33% and was significantly higher in the control cheese compared to cheese with buttermilk (
Figure 4). This may be associated with lower total solid content in samples with BM (
Figure 2). During ripening, a notable increase (
p < 0.05) in the total protein content was observed, and at the end of ripening, it reached values between 16.16% and 19.85%. The increase in protein content, similar to fat, is related to moisture loss in Camembert-type cheese during ripening [
50]. Statistical analysis indicated that protein content in BM20 at day 28 did not differ from the control, whereas BM10 had significantly lower (
p < 0.05) protein content. The protein levels both at the beginning and end of ripening are lower than the results obtained by Los et al. [
50], who found 19–22% of protein directly after production and 21–23% of protein after 28 days of ripening Camembert-type cheese samples with different starter cultures. In the study conducted by Bahrami et al. [
31] on cheese with sweet BM, samples with 10% and 20% obtained after production did not differ in protein content, which was lower than that of the control, consistent with our findings. In the cited work, the cheese samples were not ripened.
The changes in acidity during the ripening of Camembert-type cheese are presented in
Figure 5 and
Figure 6 (pH and titratable acidity expressed as % of lactic acid). The results of the statistical analysis of acidity and water activity are presented in
Appendix A (
Table A1). Throughout the ripening, a gradual decrease in acidity was observed, which was reflected in increases in pH and a decline in the lactic acid content. The initial pH was 4.45–4.48 and did not differ among cheese variants. During the first two weeks, the pH increased slightly to 4.61, 4.62, and 4.65 on day 7 and 5.18, 5.26, and 5.26 on day 14 for the control, BM10, and BM20, respectively. Subsequently, the increase in pH was more rapid, especially for cheese with buttermilk, reaching 6.01, 6.16, and 7.45 on day 21 and 6.65, 7.81, and 7.79 on day 28, respectively. More distinct pH changes in Camembert-type cheese with BM are probably connected with its higher moisture content and less dense casein gel structure, which facilitate proteolysis and other changes occurring in the cheese during maturation. At the end of ripening, samples BM10 and BM20 did not differ (
Table A1). The lack of impact of 10% and 20% buttermilk addition on cheese pH immediately after production was also reported by Bahrami et al. [
31].
The initial lactic acid content was 1.23, 1.32, and 1.46, respectively, for the control, BM10, and BM20 cheese. An increase in the lactic acid content was observed between days 1 and 7, which was associated with lactose fermentation by starter bacteria. On day 7, the lactic acid content was 2.10%, 1.64%, and 1.71%, respectively. During the remaining period, lactic acid content decreased dynamically, reaching the lowest values on day 28 (0.71%, 0.26%, and 0.21%, respectively). Similar changes in acidity during the ripening of Camembert-type cheese have also been observed by other authors [
7,
8,
40,
47,
50].
The reduction in lactic acid is caused by its consumption by mold
P. candidum developed on the cheese surface during the first two weeks. During weeks 3 and 4, as the mold was already established, the transformation carried out by its enzymes accelerated, resulting in a more dynamic drop in acidity.
P. candidum metabolizes lactate and results in the proteolysis of cheese proteins, producing ammonia, water, and oxygen [
2]. The proteolysis resulting from molds causes a decrease in acidity on the surface of the cheese, and the pH gradient from the surface to the core diminishes over time. The decrease in acidity affects the texture and is responsible for the softening of Camembert-type cheese [
8]. As a result of high pH, calcium phosphate precipitates as a layer on the surface of the cheese. This leads to a gradient of calcium phosphate from the cheese core to the surface, with this compound migrating toward the surface. This phenomenon, along with casein proteolysis, contributes to the softening of the cheese interior, which is characteristic of ripened Camembert-type cheese [
40]. The high proteolytic activity of molds used in Camembert production results in a cheese with a relatively short ripening period [
51].
The water activity of Camembert-type cheese samples with BM, as well as control Camembert, did not change during the ripening period (
Figure 7). Differences among cheese variants were found only on day 1, when the control cheese had significantly higher (
p < 0.05) a
w than BM10 and BM20 (
Table A1), with values of 0.959, 0.928, and 0.923, respectively. Los et al. [
50] recorded higher values of water activity in Camembert cheese and noted a significant decrease during ripening (from 0.973–0.976 on day 0 to 0.962–0.965 on day 28). High water activity, along with the decline in acidity during ripening, makes Camembert a type of cheese prone to spoilage [
2].
3.4. Microstructure and Texture
The microstructure of cheese is a crucial factor influencing its quality characteristics and functional properties. Thus, the knowledge about the structure of newly developed products is of high importance for producers. Our study on cheese microstructure enables an understanding of the association between cheese components each other and how they change during ripening [
52,
53]. Modern methods of microscopy, such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), and confocal laser scanning microscopy (CLSM), provide detailed insights into cheese structure at the microscopic level. Electron microscopy provides images with much higher resolution compared to traditional light microscopy [
53,
54]. This technique has been applied to analyze the microstructure of various types of cheese, including full-fat Iranian White Cheese [
55], Halloumi [
56], Cheddar [
16], Tallaga cheese [
57], and Kashar cheese [
58]. SEM has also been used to visualize the microstructure of Camembert cheese. Michalski et al. [
59] analyzed the SEM images of Camembert with different fat globule sizes. In the case of cheese, the SEM method clearly illustrates how cheese components like protein, fat, water channels, microorganisms, and precipitated minerals interact [
52,
53]. The SEM method was also used in studies on other dairy products like buttermilk [
60,
61], yogurt, and other fermented milk and powdered milk products [
62]. Besides dairy products, SEM was used to observe the microstructure of meat and soup [
63].
In terms of microstructure, cheese is a complex material composed of a protein matrix made from coagulated casein micelles to which other components such as fat, water, minerals, and microorganisms are bound. The arrangement of the microstructure is affected by several factors, including the type of raw material, cheese type, processing parameters, the addition of calcium chloride, and the type of microbiological culture used for ripening. The microstructure impacts the cheese’s texture and quality characteristics [
53,
64,
65]. Each type of cheese has characteristic microstructural properties, which are related to chemical and biological transformations [
55]. The microstructure of Camembert-type cheese samples observed on the 21st day of ripening is presented in
Figure 8.
The images of the control cheese are coarser and more porous compared to cheese samples with 10% and 20% buttermilk. However, the images of BM10, BM20, and the control cheese show considerable similarity, indicating that the level of buttermilk addition does not notably alter the structural characteristics of the products. The microstructure of all samples is characterized by a compact and dense structure. The surface of the protein matrix in BM10 and BM20 samples is very even, homogenous, and free from ripples, which may be connected with the smooth and soft texture of Camembert-type cheese at this stage of ripening. Krebs et al. [
48] observed the microstructure of rennet cheese produced from ultrafiltered buttermilk or skim milk. The results of SEM visualization showed a more irregular cheese matrix microstructure for buttermilk cheese compared to milk cheese, which contrasts with our findings and may be related to differences in cheese type and maturity level. The compact structure of the protein matrix on the 21st day of ripening was also observed by Rahimi et al. [
55] in full-fat Iranian White Cheese. Romeih et al. [
16] analyzed the microstructure of low-fat Cheddar cheese with the addition of buttermilk powder or skim milk powder. The author revealed that buttermilk addition resulted in a more homogenous and denser microstructure observed with the SEM technique, which is in line with our findings. This is explained by the ability of MFGM present in buttermilk to form cross-links with the casein matrix. This interaction strongly affects the microstructure, texture, and water-holding capacity of cheese [
13,
16,
48]. The fat globules in the cheese microstructure are sparse in the field of view. This indicates that fat is present not only in the form of emulsified fat globules but also largely in the form of free oil pools, which are caused by the coalescence of fat globules, probably linked to the high degree of cheese maturity. During cheese ripening, fat globules aggregate and coalesce, forming large fat particles and pools of free oil. This phenomenon is accelerated by the changes in the structure of the casein matrix, which stabilizes fat globules. During ripening, the degradation of the casein matrix, in which fat globules are embedded, occurs. Residual chymosin, indigenous milk enzymes, and proteases from starter culture microorganisms facilitate the proteolysis of casein. Moreover, the calcium bound to the casein solubilizes over time. As a result, the cheese softens during the ripening process [
52,
55,
66].
The instrumental texture analysis of cheese enables the characterization of various mechanical properties that reflect the product’s structure, mouthfeel, and acceptability. Several factors affect cheese texture, including moisture; pH; fat content; protein content; and notably, the structure of the protein matrix [
23]. In this study, the analysis of ripened (day 28) Camembert-type cheese allowed for the determination of the following texture parameters: hardness, adhesiveness, springiness, cohesiveness, gumminess, chewiness, and resilience. Both cheese samples with BM addition exhibited significantly lower (
p < 0.05) hardness, adhesiveness, and gumminess than the control. No significant differences were observed for the other parameters. Additionally, variants BM10 and BM20 differed only in terms of adhesiveness, with the cheese containing a higher BM content and therefore being more adhesive (
Table 3).
Camembert is a type of soft cheese, and its hardness decreases during ripening. This is due to the intense proteolysis resulting from molds. The hydrolysis of the cheese protein network into small peptides and free amino acids leads to the soft texture of the cheese. Additionally, the high water and fat contents contribute to the softness of Camembert [
2,
40,
55]. In the study by Los et al. [
50], the hardness value measured on day 28 ranged from 1 to 14 N; our results fall within this range. Low hardness indicates that the ripening process has proceeded properly.
According to Bielska and Cais-Sokolińska [
39], the texture as well as the sensory properties can be improved by BM addition. Bahrami et al. [
31], who produced cream cheese from milk combined with sweet BM, found that the cheese with BM had a softer body, which is consistent with our study. The authors suggested that this effect is connected to changes in the electric charge of casein during cream churning and the presence of milk fat globule membrane (MFGM) components, mainly phospholipids (PLs) and free fat in the BM. The lower hardness of cheese from ultrafiltered BM compared with cheese from skim milk was also stated by Krebs et al. [
48]. Romeih et al. [
16] found lower hardness in Cheddar cheese with MFGM components.
3.5. Color and Sensory Properties
The color and appearance of cheese are important factors in consumer acceptability and directly influence the quality of the product. Color is the first property of cheese noticed by consumers and may determine whether it will be accepted or rejected [
23,
43]. The color values of the tested cheese variants were significantly affected by the ripening time and BM addition (
Table 4). During ripening, L* and WI values both in the paste and on the surface of Camembert-type cheese samples with BM decreased, with the exception of WI on the surface of BM10, in which these parameters remained stable. However, at the end of ripening, BM cheese samples did not differ from the control with respect to these parameters. In regard to all cheese variants, values of a* and b* parameters increased in the interior and were stable in the exterior. BM cheese samples were characterized by lower a* values in the interior and higher a* values in the exterior compared to the control. Additionally, BM10 and BM20 had lower b* values in the cheese paste. The stability of color parameters on the cheese rind indicates a full and even coating with mycelium. The similarity of both BM Camembert-type cheese samples to the control was particularly evident at the end of the ripening period with regard to lightness and the whiteness index. This similarity suggests that the color of the developed cheese products will be appealing to consumers.
Camembert-type cheese made without or with 10% and 20% of BM addition underwent sensory assessment throughout the entire ripening period (
Table 5). For appearance, consistency, and aroma, the grades were high (from rather good to very good) during the entire ripening process, and there were no statistically significant differences among cheese samples. Also, the overall acceptability did not differ. However, the taste of Camembert-type cheese samples with BM was significantly lower (
p < 0.05) than the control at the end of the ripening period. Presumably due to the higher moisture content in cheese samples with BM, the process of ripening was faster, and the most attractive properties were reached already after 2 weeks of ripening.
The changes in the flavor of Camembert during the ripening period result from the enzymatic activity of rennet, lactic acid bacteria (LAB) present in the starter culture, and the mycelium. The aroma of ripened Camembert cheese is shaped by ammonia and other compounds produced from amino acids through processes such as deamination, decarboxylation, and further transformation. Volatile sulfur compounds also play a role in forming the characteristic Camembert flavor. Dimethyl sulfide and dimethyl trisulfide contribute to the garlicky flavor of a matured Camembert [
2,
67]. Moreover, at the later stages of ripening, the increasing concentration of bitter compounds—specifically hydrophobic peptides released during proteolysis—may negatively impact the taste of the cheese [
50]. Consequently, an extended ripening period for Camembert-type cheese may diminish its appeal. It is advisable to shorten the ripening period of Camembert-type cheese with BM addition, which is economically beneficial for the cheesemaking industry. For better insights into the sensory quality of BM Camembert-type cheeses, and to predict their commercial success, it is reasonable to perform a consumer acceptability test.