Introduction

The safety of any food is an essential characteristic that every product should have when consuming it, that is why standards have been designed for this to be carried out, in compliance with a mandatory regulatory framework. However, the rules established on some occasions have been elaborated without taking into account the characteristics of real productive activities (^{Arispe and Tapia, 2007}).

In the specific case of genuine Mexican cheeses, for example, the standard NOM-243-SSA1-2010 (^{Ministry of Health, 2010}), makes obligatory several operations that are not carried out many times in the elaboration processes of cheeses of this type; this places them in a situation of non-compliance with the rules and generates various complications to the producers. The use of pasteurized milk to attend safety is one of the main operations that the standard marks, however many of the genuine cheeses could not be elaborated under this process, since many of the characteristics that identify them and make them genuine are generated by native microorganisms from unpasteurized milk. Non-pasteurizing does not necessarily result in a lack of safety, but it is an easy and practical way that health authorities use to force small dairies, usually artisanals to comply with regulations.

This discrimination towards this production way, directly affects the competitiveness of the artisanal cheese-making sector by encouraging not only the economic but also the social depreciation of its products and obliges producers to abandon these production systems or to replace them with others that are more profitable, thus generating the loss of genuine processes and products.

The fact that in México there is a regulation that, focused on safety, generates the loss of various artisan processes and does not allow another alternative that could contribute to their maintenance, is reason to carry out this research, thus helping not only to the preservation of such processes. This might help the preservation of artisanal foods that are an important part not only of our culinary culture but also part of the traits that identify us as a nation; specifically this paper might contribute to solve the regulatory situation that affects producers of genuine Mexican artisanal cheeses (^{Cervantes et al., 2008}).

In the country, the Ministry of Health, through the official standard mentioned above, considers that in dairy products, such as cheeses, there should be a limit of indicator microorganisms and also that there should not be any pathogens, making pasteurisation mandatory (^{NOM, 2010}). However, this official position should be re-evaluated in the light of a better understanding of the characteristics of artisanal cheeses, in which pulps are carried out complex fermentations that contribute to its innocuousness.

According to ^{Jay et al. (2006)}, the most important pathogens in dairy products are: Streptococcus mastiditis, which causes contagious mastitis, Mycobacterium tuberculosis, Brucella abortus and Salmonella spp.

In particular Salmonella spp., is an important pathogen that has been found in fresh cheeses of raw and pasteurized milk, getting there trough several ways and it constitutes a serious problem in the quality of this product. In this sense, through official regulations, it has been emphasized that the problem in artisanal production concerns the processes of cheese processing, without focusing on the complete production system and the different aspects that may exist in production and that can lead to lack of safety in the product such as bad agricultural practices, poor manufacturing practices, bad practices in product distribution and even sales, as well as other existing operations in the distribution chain that can generate contamination of artisanal cheeses (^{Scaramelli et al., 1999}).

The World Health Organization (OMS, 2008) and ^{Longo (2013)} agree that salmonellosis, as well as most gastrointestinal diseases, is one of the main conditions for which people require medical care, not only in México but all over the world. For example, for the specific case of typhoid, there is a worldwide tendency to rise almost exponentially from 2002, with 22 million cases, with a stabilization in 2007; however, it continues to rise.

In developed countries, the incidence of cases has been reduced by sanitary improvements, however, it is estimated that approximately 17 million cases per year occur in the world with 300 000 deaths (^{INS, 2014}).

The disease is caused by Salmonella typhimurium and in other less frequent occasions, by some of the varieties of Salmonella paratyphi, that specifically attack humans. The microorganism has an oral-faecal transmission route, so it can be found in foods handled by carriers, which have had contact with drainage water, such as vegetables irrigated with it, some foods from the sea, among others (^{Díaz, 2005}). One of these foods are artisanal cheeses.

Due to the above, there is a need to find an alternative to milk’s pasteurization with which cheeses are made, one of them could be the acidolactic fermentation generated by acidolactic bacteria (BAL) present in the native milk microflora that perform fermentation processes in cheeses. The fact that fermentation, generated by acidolactic bacteria (BAL), can inhibit a pathogen like salmonella, provides evidence that the problem may be of a different nature. As part of an investigation that addresses this problem, the present paper shows some of the results obtained in this research.

Fortunately, a variety of investigations have shown that native strains of acidolactic bacteria (BAL), isolated from Mexican and other artisanal cheeses, have had inhibitory effects against various pathogens including Salmonella spp. (^{Del Campo et al., 2008}).

This paper aims to demonstrate that the lactic fermentation developed by these BAL is able to inhibit the growth of this pathogen, so the results of this research could contribute to demonstrate that the innocuity in artisanal cheeses of raw milk could be achieved by this biological pathway, without resorting pasteurization.

Acidolactic bacteria (BAL) are a very heterogeneous group of microorganisms from a taxonomic point of view. They are bacilli or cocci Gram(+) anaerobic or facultative aerobic bacilli that respond negatively to catalase and oxidase tests. In common all BALs produce lactic acid as the principal or only metabolite in sugars fermentation. They are classified into two groups, the homofermentative ones that only produce lactic acid from glucose and the heterofermentative ones that produce CO₂ in addition to other metabolites (^{Parra, 2010}).

The fermentation carried out by BAL is one of the techniques used in the industry to preserve and improve the sensory properties of foods; this bioprocess generates metabolites that gives the products their desired flavors and aromas. The main functions of fermentation are the acidification, the texture improvement and flavor development, among others. Acidolactic bacteria are naturally present in milk, and have been found to have the ability to inhibit the growth of pathogens such as Salmonella spp., Listeria monocytogenes, Escherichia coli and Staphylococcus aureus, through different compounds generated in fermentation, with lactic acid being the main agent and bacteriocins being the most effective (^{Piña et al., 2011}).

^{Mataragas (2003)}; ^{Peláez (2005)}; ^{Palomino (2012)}; ^{Parra (2010)}, explain that acidic bacteria, which are not only responsible for providing desirable sensory properties to dairy products, but the fermentation byproducts that they perform, for example lactic and acetic acid, hydrogen peroxide, in addition to others, affect the development of various pathogens, even inhibiting them, giving some harmlessness to the product not only in dairy products; this has been demonstrated in each of those investigations.

At present, both the pure cultures added to food and the byproducts of acid-lactic bacteria fermentation are used for commercial food preservation (Vásquez, 2009).

The most effective antimicrobial substances produced by BALs are the “bacteriocins”, which are natural peptidic compounds, which have an effect against Gram(+) bacteria, but also affect the Gram(-). For example, nisin, a bacteriosin produced by Lactococcus lactis, is approved and considered safe by the FDA as an additive for food preservation and is already used commercially (^{Piña et al., 2011}). There are also antimicrobial products of a non-peptidic nature such as organic acids, diacetyl, acetoin, hydrogen peroxide, reuterine, reuterocycline, antifungal peptides and bacteriosins. Some lactic acid bacteria that have been found to secrete bacteriocins belong to the genera Lactococcus, Lactobacillus, Carnobacterium, Enterococcus and Pediococcus. There have already been made preparations of pure and semipure bacteriocins for the inhibition of pathogens in food (^{Saidi, 2011}).

The investigations made show that native BALs of genuine cheeses have this pathogen inhibiting effect, so that Salmonella spp., can also be inhibited by the action of BALs (^{Del Campo, 2008}). The same result has also been found in several tests of antagonism of BAL strains with other pathogens such as Escherichia coli and Staphylococcus aureus as reported by ^{Larios (2007)} and ^{Aguilar and Klotz (2008)}.

For all of the above, the following research question arises: in an artisanal panela cheese produced in the state of Jalisco, México can be distinguished BAL with an antagonistic effect against Salmonella enterica var. Typhimurium? In order to answer the question, an investigation was proposed that aimed to identify the genus and species of some acidolactic bacteria native to a genuine Mexican cheese that had an antagonistic capacity against Salmonella enterica var. Typhimurium and to study its technological characteristics. The hypothesis was that in a genuine Mexican cheese of unpasteurised milk there are BAL species with antagonistic effect against Salmonella enterica var. Typhimurium.

Materials and methods

For the isolation, two samples of 25 g of genuine Mexican handmade panela cheese from a dairy, located in Soyatlan del Oro, Jalisco, obtained after its elaboration, were transported at refrigeration temperature (below 8 °C) in plastic covers and removed from refrigeration, the first at one hour and the second at 24 h, then they were diluted in 225 mL of peptone distilled water (0.001%, casein peptone) sterile and homogenized for one minute. 10 mL of this dilution was transferred to a 100 mL capacity vial containing 90 mL of sterile peptone water. This same process was repeated until a 10^-6 dilution was obtained. One milliliter of the 10^-4, 10^-5 and 10^-6 dilutions were seeded in triplicate and placed in Petri dishes with ManRogosa and Sharpe (MRSA) agar medium.

Seed boxes were incubated in anaerobic jars at 37 ±1 °C for three days. Subsequently the colonial diversity and the population of BAL (UFC g^-1 of fresh cheese) were quantified. A strain of each type of colony was isolated with bacteriological handle and re-seeded on MRS agar and incubated at 37 ±1 °C for three days. Purity was confirmed by colonial and microscopic morphology by Gram staining. Gram(+) and catalase(-) strains were selected, as presumably BAL. These strains were conserved by re-planting on lean MRSA medium.

The selected BAL strains were reactivated in MRS broth at 37 °C for three days in anaerobiosis. At the end of the incubation they were centrifuged at 10 000 rpm for 10 min and the supernatant from each strain was recovered. Subsequently, 5 mm diameter disks were impregnated with 20 μl of supernatant in three replicates per supernatant, which were placed on the surface of trypsin soy agar (TSA), previously seeded massively with a liquid medium culture of Salmonella enterica var. Typhimurium of 1*10⁸ UFC mL^-1. The boxes were incubated at 42 °C in aerobiosis for 48 h, after which the supernatants of the BALs that formed inhibition halo of the salmonella strain were identified and the diameter of the halo was measured.

The selected strains were identified by biochemical tests of the “API step 20” system (bioMérieux) and the catalog of biochemical profiles provided by the system.

Growth kinetics were then studied where the reactivated culture of each selected strain was seeded, in triplicate, in 25 mL sterile MRS broth. The media were incubated in anaerobiosis at 37 °C, the absorbance was measured in a spectrophotometer at 540 nm every 30 min. The absorbance at each measurement time was used to determine the growth rate (S; seconds), the end of the lag or adaptation phase (L; in seconds) and maximum growth (DOm; absorbance) of the 𝐷𝑂𝑜= 𝐷𝑂𝑚 ∗ 1+ 𝐸𝑥𝑝 2−4∗𝑆∗ 𝑇−𝐿 −1 model, for each selected strain. The data obtained from each of the parameters were tested for normality and homogeneity of variances to later apply variance analysis (ANOVA) with a significance level of (α= 0.05) and Tukey mean comparison tests (α= 0.05) to find which strain had the best developmental characteristics. The analyzes were supported by the SAS V9.4 package.

Results and discussion

Genuine cheese used in this experiment, with one hour of thawing, had a lower BAL population (1*10⁶ UFC g^-1) than the 24h thaw at room temperature (9.7*10⁶ UFC g^-1). This coincides with ^{Del Campo et al. (2008)}, who found a concentration between 6 and 8 log cycles of UFC g^-1 of panela cheese, whereas ^{Ruth et al. (2003)} quantified an average population equivalent to five log cycles in fresh cheeses, obtaining a slightly lower concentration.

During the isolation and selection procedure, 55 presumptively BAL strains were obtained. In the subsequent transfers (six) for purification, 24 (43.6%) were removed that did not meet the morphological characteristics and that were positive to the catalase test. The remaining isolates 31 (56.4%) were considered as BAL and were used for the antagonism test. Eleven of the 31 supernatants (35.5%) formed inhibition halos against Salmonella enterica var. Typhimurium. This inhibition was attributed to the presence of some BAL metabolite (Figure 1).

Figure 1 Low inhibition halo of Salmonella enterica var. Typhimurium generated by the metabolite effect of a BAL strain isolated from artisanal cheese. 

However, the effect of eight of the eleven supernatants was low (Figure 1) and it was remarkable only in three. The latter indicates that 9.6% of the BAL that colonize from the milk with which the genuine artisanal panela cheese was made has a considerable inhibitory effect against the growth of Salmonella enterica var. Typhimurium. ^{Ramírez (2008)} found that 26% of the asylum strains showed an inhibitory effect against various Gram(+) or Gram(-) pathogens, which is higher than that obtained in this research (20%), considering 55 isolates (11/55), but lower if we only consider the presumed BALs (35.5%, 11/31).

^{Parras (2010)} considers that the low inhibitory effect could be due to a low concentration of the fermentation products (among them lactic acid) of the BAL since for example the amount of lactic acid produced depends on the species and even on the strain.

The pH in the supernatants was measured giving a value of 4.9. The above agrees with that published by ^{Makras et al. (2005)} who demonstrated that the main cause of Salmonella enterica var. Typhimurium by BAL strains in a synthetic medium, is the drop in pH to values between 5 and 4.5. On the other hand ^{Aguilar et al. (2008)} perceived that a pH decrease to a value lower than 5 is related to the inhibition effect of BAL against Salmonella sp., whose population decreased in three log cycles. ^{Piña et al. (2011)} mentions the possibility of other metabolites with the same function.

After the tests of the “API step 20” system (bioMérieux) the biochemical profiles of the three strains studied were obtained, after which it was searched and compared with the profiles presented in the catalog, provided by the same system, to know the Genus and species of each microorganism. The biochemical profiles of the strains are shown in Table 1, comparing them with the biochemical profiles of the species to which they correspond, where it is represented by the symbol “+” if they metabolized the corresponding substance and with the symbol “-” in case they did not, thus generating the profiles of each strain.

Table 1 Identification of three BAL strains isolated from genuine Mexican artisanal panela cheese from unpasteurized milk, according to the biochemical tests of the “API step 20” system. 

+= el microorganismo respondió positivamente a la prueba; -= el microorganismo respondió negativamente.

According to the biochemical reactions of the “API step 20” system (bioMérieux), strain “1” was identified as Leuconostoc spp. The “API 20 step” system does not identify species of the genus Leuconostoc; however, ^{García et al. (2004)} shows that the species of the genus that metabolize the lactose found in dairy products and its derivatives that have been found in raw milk are L. mesenteroides, L. lactis and L. argentinum; therefore the strain must belong to one of these three species.

Strains “2” and “3” were adjusted to the profile of Lactococcus lactis subspecie cremoris. ^{Koneman (2006)}; ^{Alais (1985)} explain that this homofermentative mesophilic species is the most widely used in dairy products, mainly in cheeses. Also ^{Del Castillo et al. (2004)} report that Lactococcus lactis subspecie cremoris is a desirable species as it gives organoleptic properties to the product; in addition to providing a presumptive protective effect against pathogens.

Rodríguez (2007) found that 22% of isolated BAL strains from milk and from different artisanal and semi-industrial cheeses belong to the genus Leuconostoc, most of which were found in unripened cheeses or 10 days of maximum ripening and with pH between 4.6 and 6.3, similar conditions to the cheese used for the BAL isolation in this study.

The values of the growth kinetics parameters showed that the three strains (1, 2 and 3) grow very similarly (Figure 2). Without a different L phase (p> 0.05). The adaptation phase of the three strains is found in approximately 0 to 168 min after being inoculated, the third strain being selected (identified as Lactococcus lactis subspecie cremoris) which showed a higher adaptation rate; its exponential phase is approximately between this point and up to 360 min, from that point to 615 is the stationary phase (Figure 2).

Figure 2 Adjusted BAL growth curve in MRS broth measured in optical density. 

As shown in Table 2, strain 1 has a faster overall development on average (S= 5.4*10^-3); however, it reaches lower concentrations than the others. The strain that reaches a highest DO is strin three (DO m= 1.14) and is also the one that fastest reaches the exponential phase (L= 113). Likewise, the ranges to which each of the strains in their different phases are found are practically the same.

Table 2 Parameters of the growth kinetics of three BALs isolated from Mexican genuine artisanal panela cheese from unpasteurized milk from Soyatlan del Oro, Jalisco, with antagonistic capacity against Salmonella enterica var. Typhimurium. 

DO m= crecimiento máximo; S= tasa de crecimiento; L= adaptación. Medias con diferentes letras minúsculas indican diferencias significativas α= 0.05.

The strains show a rapid development and a fairly high growth, a growth that is explicable since it was performed in conditions very close to the ideal ones for the development of this species, these are a temperature of 37 ±2 °C, in anaerobiosis and in a synthetic medium enriched with the ideal nutrients for its metabolism. In a medium with such complex features as cheese in which the conditions are not as close to those suitable for BALs, the growth rate may be much less rapid.

Conclusions

The Panela cheese from Atengo, Jalisco, contains a high BAL population greater than 1*10⁶ UFC g^-1, which in addition to conferring sensory, conservation and technological properties to the product, can participate in the control of pathogens such as Salmonella enterica var. Typhimurium. Of a total of 55 bacterial isolates, 31 were found to be BAL of which eight showed mild inhibitory effect against Salmonella enterica var. Typhimurium and three a more severe effect. The latter were Leuconostoc spp., and Lactococcus lactis subspecies cremoris, with growth characteristics that allow it to reach its maximum growth between 8 and 9 h of incubation that makes them competitive against Salmonella enterica var. Typhimurium.

The role of the inhibitory BAL consortium as a whole and not only in isolation should be investigated to determine if there is a synergistic effect that potentiates the inhibition of pathogens. Also, when there was BAL present in artisanal cheese with antagonistic activity against a pathogen such as salmonella, evidence was provided that the acidolactic fermentation, generated by these, is an alternative to the pasteurization of milk for the cheese processing and that should be considered within the standards in México.