Classification of Latilactobacillus sakei subspecies based on MALDI-TOF MS protein profiles using machine learning models.

Latilactobacillus sakei is an important bacterial species used as a starter culture for fermented foods; however, two subspecies within this species exhibit different properties in the foods. Matrix-assisted laser desorption/ionization-time of flight mass spectrometer (MALDI-TOF MS) is the gold standard for microbial fingerprinting. However, the resolution power is down to the species level. This study was to combine MALDI-TOF mass spectra and machine learning to develop a new method to identify two L. sakei subspecies (L. sakei subsp. sakei and L. sakei subsp. carnosus) and non-L. sakei species. Totally, 227 strains were collected, with 908 spectra obtained via on- and off-plate protein extraction. Only 68.7% of strains were correctly identified at the subspecies level in the Biotyper database; however, a high level of performance was observed from the machine learning models. Partial least squares-discriminant analysis (PLS-DA), principal component analysis-K-nearest neighbor (PCA-KNN), and support vector machine (SVM) demonstrated 0.823, 0.914, and 0.903 accuracies, respectively, whereas the random forest (RF) achieved an accuracy of 0.954, with an area under the receiver operating characteristic (AUROC) curve of 0.99, outperforming the other algorithms in distinguishing the subspecies. The machine learning proved to be a promising technique for the rapid and high-resolution classification of L. sakei subspecies using MALDI-TOF MS. IMPORTANCE: Latilactobacillus sakei plays a significant role in the realm of food bacteria. One particular subspecies of L. sakei is employed as a protective agent during food fermentation, whereas another strain is responsible for food spoilage. Hence, it is crucial to precisely differentiate between the two subspecies of L. sakei. In this study, machine learning models based on protein mass peaks were developed for the first time to distinguish L. sakei subspecies. Furthermore, the efficacy of three commonly used machine learning algorithms for microbial classification was evaluated. Our results provide the foundation for future research on developing machine learning models for the classification of microbial species or subspecies. In addition, the developed model can be used in the food industry to monitor L. sakei subspecies in fermented foods in a time- and cost-effective method for food quality and safety.

Evaluation of the effect of Lactobacillus sakei strain L115 on Listeria monocytogenes at different conditions of temperature by using predictive interaction models.

In this study, the inhibitory capacity of Lactobacillus sakei strain L115 against Listeria monocytogenes has been assayed at 4, 8, 11, 15 and 20 °C in broth culture. Besides, the use of predictive microbiology models for describing growth of both microorganisms in monoculture and coculture has been proposed. A preliminary inhibitory test confirmed the ability of Lb. sakei strain L115 to prevent the growth of a five-strain cocktail of L. monocytogenes. Next, the growth of microorganisms in isolation, i.e. in monoculture, was monitored and kinetic parameters maximum specific growth rate (µsp;max) and maximum population density (Nmax) were estimated by fitting the Baranyi model to recorded data. Inhibition coefficients (α) were calculated for the two kinetic parameters tested (µsp:max and Nmax) to quantify the percentage of reduction of growth when the microorganisms were in coculture in comparison with monoculture. The kinetic parameters were input into three interaction models, developed based on modifications of the Baranyi growth model, namely Jameson effect, new modified version of the Jameson effect and Lotka-Volterra models. Two approaches were utilized for simulation, one using the monoculture µsp;max, under the hypothesis that the growth potential is similar under monoculture and coculture conditions provided the environmental conditions are not modified, and the other one, based on adjusting the monoculture kinetic parameter by applying the corresponding α to reproduce the observed µsp;max under coculture conditions, assuming, in this approach, that the existence of a heterogeneous population can change the growth potential of each microbial population. It was observed that in coculture, µsp;max of L. monocytogenes decreased (e.g., α = 31% at 4 °C) and the Nmax was much lower than that of monoculture (e.g., α = 36% at 4 °C). The best simulation performance was achieved applying α to adjust the estimated monoculture growth rate, with the modified Jameson and Lotka-Volterra models showing better fit to the observed microbial interaction data as demonstrated by the fact that 100% data points fell within the acceptable simulation zone (±0.5 log CFU/mL from the simulated data). More research is needed to clarify the mechanisms of interaction between the microorganisms as well as the role of temperature.

Developing novel species-specific DNA markers for PCR-based species identification of the Lactobacillus sakei group.

Identification of members of the Lactobacillus sakei group (LSG) by common phenotypic and genotypic methods is generally inadequate and time-consuming. The objective of this study was to develop novel species-specific primers based on sequence-characterized amplified region (SCAR) markers using random amplified polymorphic DNA polymerase chain reaction (RAPD-PCR) analysis. Three species-specific fragments were gel-purified, cloned and sequenced after preliminary screening of 80 random primers. Accordingly, three pairs of primers Lcur-F/R, Lgram-F/R and Lsakei-F/R were designed based on single species-specific bands (281, 278 and 472 bp) that were obtained from Lactobacillus curvatus, Lactobacillus graminis and L. sakei, respectively. The specificities of these primer pairs were confirmed in 21 LSG strains and 31 nontarget Lactobacillus strains. In addition, the detection limits for each primer pair were approx. 105 , 104 and 106 cells per gram of meat samples spiked with L. curvatus, L. graminis and L. sakei, respectively. In conclusion, we have successfully developed a rapid, accurate and effective PCR-based method for identification of species in the LSG. SIGNIFICANCE AND IMPACT OF THE STUDY: Neither phenotypic nor the most commonly used genotypic method (16S rRNA gene sequencing) provides sufficient resolution for accurate identification of the Lactobacillus sakei group. A sequence-characterized amplified region method developed in this study provides a rapid, cost-effective way to detect the member of the L. sakei group in meat sample.

Detection of different microenvironments and Lactobacillus sakei biotypes in Ventricina, a traditional fermented sausage from central Italy.

The present study evaluated the physico-chemical and microbiological features of Ventricina, considering for the first time the presence of different compartments deriving from the technology of production. In fact meat pieces (pork muscle and fat cut into cubes of about 10-20cm3), mixed with other ingredients and then stuffed into pig bladder, are still distinguishable at the end of the ripening. They appear delimited on the outside by the casing and inside by thin layers consisting of spices (mainly red pepper powder), salt and meat juices. Our results showed that the exterior (portion of the product in contact with the casing), the interstice (area between the different cubes of meat or fat) and the heart (the inner portion of meat cubes) had distinctive values of pH and aw, and a typical microbial progression, so that they can be considered as different ecological niches, here called microenvironments. The study of lactic acid bacteria population, performed with PCR-DGGE and sequence analysis targeting the V1-V3 region of the 16S rRNA gene (rDNA), highlighted the presence of a few species, including Lactobacillus sakei, Lb. plantarum, Weissella hellenica and Leuconostoc mesenteroides. The RAPD-PCR analysis performed on Lb. sakei, recognised as the predominant species, allowed the differentiation into three biotypes, with that characterised by the highest acidifying and proteolytic activities and the highest ability to grow in the presence of sodium chloride prevailing. This leading biotype, detectable in the interstice during the entire ripening period, was isolated in the microenvironments exterior and heart starting from the 30th d of ripening, and it was the sole biotype present at the end of the ripening. The analysis of microenvironments through the scanning electron microscopy (SEM) evidenced the presence of micro-channels, which could favour the microbial flow from the interstice to the exterior and the heart. Moreover, the SEM analysis allowed the detection of biofilms, recognised as responsible for the correct colonisation of the different meat niches.