Cheese samples

Samples of PC made in an artisanal way in the municipality of Cuajinicuilapa, Guerrero, Mexico (16°28′18′ N, 99°24′55′ W), were analyzed in July 2018. Based on a targeted sampling, four cheese factories were selected as sampling units, which will henceforth be named A, B, C and D. Four freshly made cheeses, weighing 1 kg, were purchased from each cheese factory. The samples were moved to the municipality of San Marcos, Guerrero (16°47′46′ N, 99°23′05′ W) to a space with characteristics similar to those of the cheese factories of Cuajinicuilapa. In this place, the samples of the cheeses from the cheese factories A, B, C and D (four of each cheese factory) were left to mature for 5, 30, 60 and 90 d. After the ripening time, each batch was transferred in polyethylene bags, inside coolers with refrigerant, to the laboratory. The samples were kept in refrigeration at 4 °C until analysis. The maximum ripening time was 90 d because after that period the flavors intensify, and local consumers avoid it because they prefer softer flavors.

Sample preparation

Each of the cheese samples (25 g) was mixed with 225 mL of peptone diluent, the mixture was homogenized for 2 min (VWR^® symphony D S41 Vortex, VWR International) and dilutions from 10^-1 to 10^-6 were made by transferring 1 mL of the sample to vials containing 9 mL of peptone diluent¹⁷.

Microorganism count

The following culture media (3M Petrifilm^TM plates) were used: aerobe count (AC No. of catalog 6400), coliform count (TC No. of catalog 6410), lactic acid bacteria (No. of catalog 6461) and staph express for coagulase-positive staphylococci (No. of catalog 6493); 1 mL of the corresponding dilution was placed in each of the plates^18-21.

All counts were done in duplicate. For AMB, dilutions 10^-3 and 10^-4 were sown and the medium was incubated at 35 ± 2 °C for 48 ± 3 h¹⁸. TC microorganisms were studied based on dilutions 10^-2 and 10^-3, being incubated at 35 ± 1 °C for 24 ± 2 h¹⁹. The determination of LAB was made by inoculating the media with dilutions from 10^-3 to 10^-6 and incubating at 35 ± 2 °C for 48 ± 3 h²⁰. The CPS study was conducted from dilutions 10^-2 to 10^-4 and an incubation at 37 ± 1 °C for 24 ± 3 h²¹.

Once the incubation time was completed, the growth was reviewed and the plates containing between 15 and 300 colonies were counted, the mean of the two repetitions was obtained and this average was multiplied by the inverse of the dilution with which the plate was inoculated²². The result of the count was reported as log₁₀ of the number of colony-forming units per gram (log₁₀ CFU g^-1)²³.

Statistical analysis

The statistical analysis was based on a design with repeated means and completely random distribution of treatments, tested over time, whose probabilistic model corresponds to:

Where:

μ+ai+yk+ayik is the mean of treatment i at time k, which contains the effects of treatment, time, and the time × treatment interaction;

eijk is the random error associated with the measurement at time k on j assigned to treatment i.

The effect of the treatments (cheese factories A, B, C and D) was evaluated through the ripening time (5, 30, 60 and 90 d) with four repetitions, generating 64 experimental units, each consisting of 25 g of cheese.

The response variables evaluated were: total count of aerobic mesophilic bacteria (AMB), total coliforms (TCs), lactic acid bacteria (LAB) and coagulase-positive staphylococci (CPS). The data were analyzed using a mixed model^24,25 whose random effect corresponds to the maturation time and the fixed effect to the cheese factories. The Tukey-Kramer method (P<0.05) was applied to identify the effect of the treatments. Analyses were performed in the SAS package version 9.1 (SAS Institute, Inc., Cary, NC, USA).

Aerobic mesophilic bacteria

In the AMB counts, there were significant differences (P<0.05) between the cheese factories (Table 1). The gradual decrease in AMB from day 5 to 90 was close to 2 log₁₀ CFU g^-1, for cheese factories A, B and D; while for cheese factory C, it was 3.49 log₁₀ CFU g^-1.

Most of the AMB values found in the PC are within the range of 4 to 9 logarithms CFU g^-1, reported for cheeses made from raw milk and matured for 60 or more days²⁶. Since the maturation process of cheeses involves the multiplication of the microorganisms present, AMB concentrations of 4 to 9 logarithms (10⁴ to 10⁹ CFU g^-1) are expected in this type of products, without this implying a deterioration of the food or suggesting that non-sanitary conditions occurred during its preparation or storage^26,27,28. In the region of Ojos Negros in the state of Baja California, based on a study that included matured cheeses from 22 cheese factories, made with raw milk, it is reported that AMB were found in a range of 4.6 to 7.2 log₁₀ CFU g^-1(14.

In studies on the ripening of artisanal Cotija cheese, salted and matured at temperatures of 14 °C to 32 °C, Chombo¹¹ reports the following variations in AMB: 8.3, 7.0, 3.5 and 4.7 log₁₀ CFU g^-1, on d 8, 30, 60 and 90, while Magallón²⁹ found 5.3 and 1.8 log₁₀ on d 30 and 90, respectively. The counts found in the PC at 30 and 90 d are very close to those reported for Cotija cheese that is ripened in temperature ranges similar to those of PC.

The decrease in AMB was common in the cheeses from the four cheese factories (Table 1), reflecting a certain homogeneity in the preparation processes and the ripening conditions of the cheeses. The statistical difference (P<0.05) between cheese factories suggests quantitative or qualitative variations in the microbiota of cheese generated by milk and the microenvironments of each cheese factory. It should be noted that, between d 5 and 60, it is observed that the AMB of cheeses from cheese factory A descend 2.31 log₁₀ CFU g^-1, however, between d 60 and 90 they remain unchanged; while the AMB of cheeses from cheese factory C, from d 60 to 90, show a reduction of 2 log₁₀ CFU g^-1.

The development of AMB in cheeses from cheese factory A shows the ability of bacteria to adapt and survive, while the decrease in AMB in the cheeses from cheese factory C exhibits a loss of viability with the release of enzymes that contribute to the generation of flavors and textures⁵. This suggests that it is appropriate to study the relationship of the organoleptic characteristics of prensa cheese, between cheese factories and between different concentrations of AMB over maturation time, as well as the relationship of AMB with the shelf life of cheese at room temperature. On the other hand, AMB studies are useful as an initial stage in the search for starter cultures from artisanal cheeses. Variations in salt concentration and moisture may have influenced the survival of AMB. The results suggest that, in cheeses from cheese factories A and D, there are bacteria adapted for long-term survival. While in cheeses from cheese factories B and C, bacteria of short survival may be present, and therefore can contribute more quickly to the production of tastes, smells and textures (pleasant or unpleasant)⁵. Another possibility is that the reduction of AMB in cheeses from cheese factories B and C is due to the presence of substances with antimicrobial action, caused by the metabolism of microorganisms, or by the biochemical changes that occur in the cheese, derived from the proteolysis of casein to give rise to peptides with antimicrobial activity^5,30.

Total coliform bacteria

Total coliform counts in the cheeses showed significant differences (P<0.05) between cheese factories during the ripening period (Table 1). The highest counts were found on d 5 (4.72 log₁₀ CFU g^-1) and the lowest on d 90 (<1 log₁₀). Cheese, being a solid sample, hinders a direct count, so it is necessary to make an initial dilution that leads to the minimum detection level being 10 CFU g^-1, so the absence of growth was reported as < 1 log₁₀.

Two different dynamics were observed, cheeses from cheese factories A and D had the highest initial TC loads, which decreased at 30 d of maturation to reach 3.72 to 3.10 log₁₀ CFU g^-1, respectively (Table 1). In cheeses from cheese factories B and C, the TCs showed initial levels of 2 logarithms, which rose during the first month and remained with slight variations to coincide on d 90 with values very close to each other.

The dynamics of TCs in cheeses from cheese factories A and D have been reported in semi-hard cheeses and are characterized by a progressive decrease in coliforms as ripened progresses³¹, which is attributed to the decrease in pH due to the fermentation of lactose³². In ripened cheeses such as Cheddar, coliforms die at a rate of 0.3 log₁₀ CFU g^-1 per week and in Gouda cheese at 0.7 log₁₀ per week³³. Therefore, the dynamics observed in cheeses from cheese factories B and C is atypical, because between d 5 to 30, there is an increase of 0.73 and 1.37 log₁₀, respectively, followed on d 60 to 90 by very small decreases, 0.03 and 0.19 log₁₀, respectively (Table 1). This suggests that there are common aspects between the two cheese factories that favor the selection of bacteria that persist during ripening, such as water and milk quality, personnel or variations in the preparation or cleaning processes.

The accepted levels of total coliforms in matured cheeses are less than 100 CFU/g (< 2 log₁₀ CFU g^-1)³⁴, values that were reached between d 60 and 90 (Table 1) in cheese factories A and D. According to Metz³², cheeses made with good quality raw milk, under hygienic sanitary conditions, applying good manufacturing practices and properly ripened, will have low levels of total coliform bacteria, fecal coliforms, enterococci, Enterobacteriaceae and Escherichia coli.

The activity of coliform organisms in cheeses can adversely affect their sensory characteristics^28,35. However, it has been observed that certain genera of coliforms contribute positively to the texture and sensory characteristics of the product; in addition to the fact that some strains of Hafnia contribute to the accumulation of aromas, and generation of flavors^32,35,36. The persistence of coliforms suggests the possibility of the participation of these microorganisms in the organoleptic characteristics of cheeses from cheese factories B and C, an aspect that has not been addressed in studies of Mexican cheeses.

Lactic acid bacteria

During the ripening period, the LAB counts of the cheeses from the cheese factories studied showed significant differences (P<0.05) (Table 1). This microbial group had the highest counts of the entire study. On d 5 the counts ranged between 5.50 and 6.73 log₁₀ CFU g^-1, these values decreased from day 30. From day 5 to 60, the reductions were 2.99, 2.15, 2.31 and 1.77 log₁₀, for cheeses from cheese factories A, B, C and D, respectively.

During the ripening of the Spanish artisanal cheeses Casar de Cáceres, Afuega’l Pitu and Cabrales, decreases in lactococcal counts of 2 to 3 log₁₀ CFU g^-1 were reported between d 0 and 60, while in “La Serena” cheese there was only a reduction of less than one logarithm, this is attributed to the fact that this cheese had a low salt content during the first weeks of ripening⁵. This suggests that the reductions observed in the PC from d 5 to 60 are consistent with what happens with homofermentative lactic acid bacteria in cheeses made in an artisanal way with native microbiota⁵. The concentrations of LAB found in the PC from d 60 to 90 are lower than those reported in ripened cheeses from Europe, which have values of 7 to 9 log₁₀ CFU g^-1(5,37,38.

With differences in the type of cattle and geographical areas, Cotija cheese and PC share temperatures, preparation and ripening processes. In Cotija cheese, small increases and decreases have been found in LAB counts; one study reports 2.6 log₁₀ CFU g^-1 on d 30 with an increase to 2.9 log₁₀ on d 90²⁹, another study indicates 5.9 log₁₀ on d 60, which decreases to 5.0 log₁₀ on d 90¹¹. The above data suggest that, in both PC and Cotija cheese, LAB counts tend to be lower than in other ripened cheeses; this could be explained by the temperatures at which they are ripened, which favors a greater loss of moisture that generates values of water activity and moisture/salt ratio that are inhibitory for LAB⁵.

Coagulase-positive staphylococci

The CPS counts of the cheeses from the cheese factories studied showed significant differences (P<0.05) during the ripening period (Table 1). From d 5 to 60, a continuous decrease in CPS was observed in cheeses from cheese factories A, C and D, followed by a stabilization from d 60 to 90. In cheese factory B, there was a decrease from d 5 to 30 followed by an increase from d 30 to 60 and a stabilization from d 60 to 90 (Figure 2). The death rates of CPS (average decrease log₁₀ CFU g^-1 divided between week of ripening)³⁹ in cheeses from cheese factories A and C were 0.30 log₁₀ and 0.23 and 0.31 log₁₀ in cheese factories B and D, respectively.

Figure 2 Antagonism and change in lactic acid bacteria (LAB) concentration with respect to coagulase-positive staphylococci (CPS) 

For cheeses made from raw milk, the accepted limits of coagulase-positive staphylococci are 10⁴ to 10⁵ CFU g^-1, which is equivalent to 4-5 log₁₀ CFU g^-140, limits that were exceeded on day 5 in cheese factories A, C and D but that were reached again on day 30 and remained until day 90 (Table 1). CPS counts greater than 4 log₁₀ show the need to apply corrective measures in the hygiene of the processes of milk collection, cheese making and the selection of raw materials⁴⁰. Values of 10⁵ CFU g^-1 or higher lead to study the presence of staphylococcal toxin in cheeses⁴⁰, since being thermostable, it can persist even when staphylococci have died. Concentrations of 10⁶ CFUs g^-1 are usually needed to produce enough toxin (one nanogram per gram of cheese) to cause a disease outbreak⁵.

Reductions of Staphyloccocus aureus of 1 to 3 logarithms have been reported in different cheeses^41,42,43, figures that coincide with reductions in CPS during PC ripening.

The death rate of S. aureus in Manchego cheese made with raw milk from d 1 to 60 is 0.404 log₁₀ CFU g^-139, the death rates of CPS in cheese factories A, C and D (0.49, 0.50 and 0.52) were close to this value, suggesting that the decrease was as expected for this type of cheese (matured Prensa paste cheeses). No factors that explain the lower death rate observed in cheese factory B (0.36) were identified.

The development and survival of S. aureus are affected by factors such as: physicochemical changes that occurred during the ripening process, secondary metabolites generated by LAB, as well as the composition of the product, storage period and temperature^44,45. Staphyloccocus aureus is inhibited by LAB through nutrient competition, production of lactic acid, hydrogen peroxide and production of antimicrobial substances⁴⁶, which may explain the decrease in CPS in the first 30 d of ripening, a period in which LAB levels were higher (Figure 2). Between d 30 and 60, storage conditions at room temperature could increase moisture loss, which changes the moisture/salt ratio, being inhibitory for LAB⁵, this favors S. aureus, which could explain its slight increase in cheese factory B and the suspension of its decrease in cheese factories A, C and D.