Selection of commercial protective cultures to be added in Sardinian fermented sausage to control Listeria monocytogenes.

Sardinian fermented sausage "Salsiccia Sarda" is a Mediterranean-style, semi-dry, fermented, RTE product, representing the main pork meat product in Sardinia (Italy). The high variability that characterizes the technological processes applied in different production plants results in sausages with different chemico-physical features sometimes permissive for the growth of Listeria monocytogenes. In order to guarantee the hygienic-sanitary quality of the final product and to innovate the manufacturing process, the main objective of this study was to evaluate the use of different commercial protective cultures to control L. monocytogenes growth in the Sardinian fermented sausage. In the first step, in vitro tests were carried out to evaluate the effectiveness of five freeze-dried bioprotective cultures availabe on the market in limiting the growth of L. monocytogenes. The two protective cultures that showed the best in vitro results were selected for a challenge test on artificially contaminated Sardinian fermented sausages. Moreover, the protective culture that showed the best results in inhibiting the growth of L. monocytogenes according to in vitro and challenge test experiments, was included into real production settings and validated in three producing plants. As a result, it was observed that protective cultures represent an important technological innovation for the Sardinian fermented sausage processing plants as they allow to control L. monocytogenes growth without altering the composition, the microflora and the chemical-physical characteristics of the product, thus ensuring safety and quality. Protective cultures also showed to reduce Enterobacteriaceae mean levels at the end of ripening and not to affect the natural concentration of lactic acid bacteria and coagulase-negative staphylococci. ©Copyright: the Author(s).

Introduction

Listeriosis caused by Listeria monocytogenes (L. monocytogenes) is one of the most severe food-borne diseases under EU surveillance. According to the report published in 2021 by the European Food Safety Authority and the European Center for Disease Prevention and Control, referring to 2020 data, L. monocytogenes was ranked the fifth most commonly reported zoonosis agent, with 1,876 confirmed invasive human cases of listeriosis and 16 outbreaks. A decrease in cases and outbreaks was observed (2,621 confirmed cases and 21 outbreaks in 2019), but the overall growing trend for listeriosis in 2016–2020 did not show any statistically significant modification (EFSA & ECDC, 2021). The fatality rate of L. monocytogenes infections in the EU was 13.0% in 2020, showing the highest number of fatal cases among foodborne infections. The pathogen is therefore a significant burden for public health, causing hospitalization, high morbidity and mortality, notably among the elderly (EFSA & ECDC, 2021).

The most significant infection source of L. monocytogenes for humans is represented by foods, especially ready-to-eat (RTE) that can be contaminated during or after processing and do not undergo any treatment that ensures their safety before consumption (Neri et al., 2019). In the EU during 2020, L. monocytogenes was detected in 3.0% of RTE pig meat products, with a 0.9% increase respect to 2019 (EFSA & ECDC, 2021). L. monocytogenes has been identified at every point of the pig-meat supply chain (Kanuganti et al., 2002; Thèveneot et al., 2006; Meloni et al., 2013). Contamination is often due to the presence of L. monocytogenes in raw materials (Thevenot et al., 2006): pigs can be carriers of L. monocytogenes in their intestines most often as asymptomatic shedders (Esteban et al., 2009, Boscher et al., 2012) and not being identified neither in the farm nor in the slaughterhouse during ante and post mortem inspections, can potentially contaminate the carcasses (Kanuganti et al., 2002; Fosse et al., 2009). The incidence of the microorganism increases when going further in the pork processing industry from the slaughterhouse to the following steps (e.g. cutting, mincing), also due to crosscontamination which occurs by the environment and equipment of the processing plants, in consideration of L. monocytogenes capability of contaminating surfaces with bacterial attachment and biofilm formation (Autio et al., 2000; Peccio et al., 2003).

Salsiccia Sarda or Sardinian fermented sausage (SFS) is the main pork meat product in Sardinia (Italy): it is a Mediterraneanstyle, semi-dry, fermented, RTE product included in the National List of traditional food products (Twentieth revision of the list of traditional agri-food products, Italian Republic, 2020). The SFS production sector includes small businesses, with distinctly artisanal productions, and larger plants with standardized industrial processing. For this reason, the manufacturing process is characterized by a great variability, most of all in artisanal plants that are strongly influenced by customs and family recipes (Meloni, 2015). Fermented sausages are usually made using lactic acid bacteria (LAB) and nitrate-reducing coagulase-negative staphylococci (CNS), which are often naturally present in the meat or added by inoculation of starter cultures during the mixing step (Greco et al., 2005). The safety of the SFS depends on the application of several sequential “hurdles” at different stages of the fermentation and ripening process (Mangia et al., 2007; Meloni, 2015; Piras et al., 2019). Safety of the final product is therefore ensured by the presence of multiple factors and specific physico-chemical conditions, such as pH, water activity (aw), sodium chloride, nitrates and nitrites, which interact in limiting microbial growth (Piras et al., 2019). In SFSs, pH values comprised between 5.3-5.5 and aw values ≤ 0.920 indicate correct acidification drying processes (Greco et al., 2005): products reporting these values at the end of ripening can be included in the category of RTE products unable to support the growth of L. monocytogenes (Regulation EC No. 2073/2005). However, the aforementioned high variability that characterizes the technological processes, especially in small artisanal production plants, results in products having different chemical and physical features, which can sometimes be permissive for L. monocytogenes growth. In fact, either strongly contaminated raw materials or inadequately applied production process steps (e.g., insufficient fermentation or inappropriate dry-curing) can determine a deficiency in the development of the hurdles and the creation of favorable conditions for L. monocytogenes growth (Mureddu et al., 2014).

Therefore, it is necessary to implement technological innovations into the SFS production process in order to guarantee the hygienic-sanitary quality of the final product and to ensure that the level of contamination by L. monocytogenes does not exceed 100 CFU/g up to the end of the shelf life as established by the EU legislation (Regulation EC No. 2073/2005) for the whole commercial life.

The use of bioprotective cultures has been gaining interest recently, especially in the dairy and meat industries, because they are safe for consumption and naturally dominate the microbiota of many foods. Protective cultures consist of bacteria that are specifically selected for their ability to inhibit the growth of pathogens or microbiological spoilage agents (Young and O’Sullivan, 2011). Protective cultures can exert a bioprotective or inhibitory effect against other microorganisms, due to their competition for nutrients and the production of bacteriocins and other antagonistic compounds such as organic acids, hydrogen peroxide and enzymes (Davidson et al., 2015). Furthermore, protective cultures can delay the development of spoilage microorganisms, extending the shelf-life. The faster metabolism of protective cultures outcompetes pathogens for the available nutrients and thus offers protection against L. monocytogenes, as a further barrier in the hurdles technique of protection (Youg and O’Sullivan, 2011). To the best of our knowledge, no available studies investigated the use of bioprotective cultures against L. monocytogenes in SFS. In this framework, the main objective of this study was to evaluate the effectiveness of different commercial protective cultures in inhibiting the development and the growth of L. monocytogenes in the SFS.

Materials and methods

Study set-up

The project was divided into different steps: 1) in the first phase, the in vitro efficacy of 5 freeze-dried protective cultures against L. monocytogenes was assessed; 2) in the second phase, the protective cultures that showed the best in vitro results against the pathogen were used during a challenge test for the production of the SFS; 3) in the third step, the protective culture that showed the best results during the challenge test was used during the manufacturing process of three SFS processing plants in Sardinia, representative of the entire regional sector.

Bioprotective cultures

Five lyophilized bioprotective cultures (A, B, C, D, E) available on the market, consisting of a mix of lactic acid bacteria (LAB) and micrococci, were selected. The selection criteria were represented by their proven activity against L. monocytogenes and their adaptability to both meat substrate and temperatures typically used in the production process of SFS. In particular, the species included in the composition of bioprotective culture A were Lactobacillus sakei, Pediococcus acidilactici, Staphylococcus carnosus and Staphylococcus carnosus subsp. utilis. Bioprotective culture B consisted of a mixture of Pediococcus acidilactici. The composition of bioprotective culture C included Lactobacillus plantarum. Bioprotective culture D included bacteriocin-producing strains of Carnobacterium ssp. Species included in bioprotective culture E were a mixture of different Carnobacterium spp. strains. According to the manufacturer’s instruction, the cultures were individually rehydrated by dilution in 0.85% sterile NaCl solution immediately before their use.

Samples

SFSs were manufactured according to the technological process applied by producing plants representative of the sector. Briefly, the production process involved selection, chopping and mincing of pork meat and fat, followed by mixing with curing ingredients, spices and authorized additives, including nitrates and nitrites at maximum concentrations of 150 mg/kg each (Reg. EC 1333/2008). Starter cultures consisting of lactic acid bacteria (LAB) and nitrate-reducing coagulase-negative staphylococci (CNS) were added during the mixing step. After overnight refrigerated storage, the mixture was stuffed in natural bowel (mutton or beef). The fermentation stage continued during the next steps of initial dipping (20-22°C for 24h, 70-80% humidity) and drying (2-3 days with progressive decrease of temperature and humidity). Ripening was carried out for about 20 days in storerooms at 15°C and 70–75% humidity. The production process is summarized in Figure 1. The finished products were cylindrical in shape, with a length of 40-45 cm and a diameter of 3-4 cm, folded with the characteristic horseshoe shape; the weight was between 300 and 600 grams. Each SFS was regarded as a sample.

Figure 1.

Process flow diagram of SFS.

Step one: in vitro assessment

Two L. monocytogenes reference strains (American Type Culture Collection, ATCC 19111 and National Collection of Type Cultures, NCTC 10887) and three L. monocytogenes wild type strains, were used for the challenge tests. L. monocytogenes wild type strains were isolated from naturally contaminated SFSs samples and producing plants environment from previous investigations and identified by PCR according to the protocol by Lyu et al. (2013). The wildtype strains were selected in order to evaluate the effectiveness of the bioprotective cultures on isolates that were already adapted to plants’ environments. L. monocytogenes strains were stored at -80°C and revitalized after incubation at 37°C for 18-24 h.

The well diffusion assay method was used according to the protocol defined by Cosentino et al. (2012), with modifications. In a first phase the selected protective cultures was tested as it is in order to test its anti-Listeria activity. Afterwards the selected protective cultures was tested as a “cell free” supernatant. The antimicrobial activity was expressed as the diameter of the inhibition zones around the wells. The bioprotective cultures were considered effective against L. monocytogenes if an inhibition zone greater than 15 mm was measured (Maragkoudakis et al., 2009).

Step two: challenge test

The definition of the protocol for the challenge test was conducted according to the “EURL Lm Technical guidance document on challenge tests and durability studies for assessing shelf-life of ready-to-eat foods related to L. monocytogenes” (version 4 July 2021). An experimental inoculum was set up consisting of a mixture of n. 5 strains of L. monocytogenes (n. 2 reference strains + n. 3 wild type strains) to challenge SFSs. The experimental inoculum was added in the mixing machine, together with the other ingredients. In order to account for natural contamination levels and, at the same time, to obtain a level of inactivation of L. monocytogenes estimated around 1-2 log at the end of the ripening, the inoculum level was set between 10 and 100 CFU/gr.

The bioprotective cultures that had showed the best in vitro results were selected for the challenge test of SFSs. According to the manufacturer’s instruction, the freeze-dried cultures were resuspended in sterile 0.85% NaCl to obtain a final concentration of ca. 107 CFU/mL.

In a pilot producing plant located at the Department of Veterinary Medicine of Sassari University, n. 3 batches of SFSs were made with ingredients obtained by 3 representative production plans were prepared (total of 9 batches). Four types of samples for each batch were produced: 1) negative control samples (C, n. 36 samples), 2) positive control samples added with L. monocytogenes broth culture (CL, n. 36 samples), 3) samples added with protective culture A and L. monocytogenes (CLA, n. 36 samples) and 4) samples added with protective culture B and L. monocytogenes (CLB, n.36 samples). All the samples were produced according to the process shown in figure n. 1. L. monocytogenes broth cultures and protective culture were added during the mixing step. Triplicate samples of each of the nine batches of SFS were analyzed at four analysis times: after stuffing (T0), 24h after stuffing (T1), 6 days after stuffing (end of drying phase, T6) and 20 days after stuffing (end of ripening, T20). A summary of the experimental design with the test units, sampling point and analysis conducted during the challenge study is reported in Table 1.

Table 1.

Type of analysis, testing time and relative minimum number of test units performed per batch during the challenge test.

Analysis	Samples	Analysis time				Total
		T0	T1	T6	T20
Detection and enumeration of L. monocytogenes
Intrinsic properties: pH and aW; composition (%): moisture; fat; protein	C	9	9	9	9	36
	CL	9	9	9	9	36
	CLA	9	9	9	9	36
	CLB	9	9	9	9	36
Total						144

C: negative control samples; CL: positive control samples; CLA: samples added with protective culture A and L. monocytogenes broth culture of; CLB: samples added with protective culture B and L. monocytogenes broth culture; T0: after stuffing; T1: 24h after stuffing; T6: 6 days after stuffing; T20: 20 days after stuffing (end of ripening).

Microbiological profile

At each time-point, L. monocytogenes quantitative and qualitative detection (UNI EN ISO 11290-1/2:2017) was carried out on both control and treated samples. Mesophilic lactic acid bacteria (LAB) were investigated according to ISO 15214:1998, using De Man, Rogosa and Sharpe agar (MRS agar, Biolife, Milan, Italy). For the enumeration of micrococci, Coagulase- Positive Staphylococci (CPS) and Coagulase-Negative Staphylococci (CNS), Mannitol Salt Agar (MSA, Biolife, Milan, Italy) was used. Finally, for Enterobacteriaceae enumeration the EN ISO 21528-2:2017 was applied.

Physico-chemical and composition analysis

On each sample, pH and aw were determined using pH meter GLP 22 (Crison Instruments SA, Barcelona, Spain) and Aqualab CX3 (Decagon, Pullman, Washington, USA). Moisture, fat and protein (expressed as %) were determined by the FoodScanLab (FOSS, Analytic, Hillerød, Denmark) using the Near-Infrared Transmittance (NIT) technology and a previously set calibration curve. Analyzes were performed in triplicate on a homogenized sample representative of the product.

Step three: application of the protective culture during the manufacturing processes of three SFS producing plants

In the third step, the protective culture that showed the best results in inhibiting the growth of L. monocytogenes according to in vitro and challenge test experiments, was included into the normal production process of three SFS producing plants (P1, P2, P3). The aim of this experimental application was to identify any changes in physicochemical and composition characteristics, as well as microbiological profile, of the SFSs added with the protective culture.

Three batches of SFSs were produced at each plant. Two types of samples were manufactured: 1) negative control samples without protective culture (n. 27 samples) and 2) samples inoculated with protective culture during the mixing step of the production process (n. 27 samples). Triplicate samples of each batch of SFS were analyzed at the end of ripening for a total of n. 54 samples.

On each sample, L. monocytogenes quantitative and qualitative detection was carried out according to UNI EN ISO 11290-1/2:2017. Moreover, from an initial suspension and decimal dilution, mesophilic lactic acid bacteria (LAB), micrococci, Coagulase-Positive Staphylococci (CPS), Coagulase-Negative Staphylococci (CNS) and Enterobacteriaceae enumeration was conducted as described before. Salmonella spp. presence was determined according to ISO 6579-1:2020.

Physico-chemical and composition analysi

On each sample at the end of ripening, physico-chemical and composition analyses were conducted, as described before.

Statistical analysis

Differences among average microbiological group counts (log10 cfu/g) and pH, over time (T0, T1, T6 and T20) and among treatments were compared using Fisher’s least significant difference test. Statistical analyses were performed with Statgraphics Centurion XIX software (Stat Point Technologies, Warrenton, VA, USA).

Results

In vitro assessment

The results showed that the protective cultures with the most efficient anti-Listeria effect were cultures A and B. Both cultures showed a clear inhibition zone around the well greater than 15 mm on all tested Listeria strains. On the other hand, bioprotective cultures C, D and E showed an inhibition zone less clear and smaller than 15 mm. Based on in vitro tests results the cultures A and B were used for the challenge test.

Challenge test

Both protective cultures A and B, selected during the in vitro test, showed in vivo efficacy against L. monocytogenes. However, protective culture A showed the strongest anti-Listeria effect: in samples added with protective culture A (CLA samples), a reduction in the number of L. monocytogenes was detected in the product starting from T1 analysis time, with a significant decrease (P<0.01) between T0 and T1 when a reduction > 1 log was observed. Afterwards, mean L. monocytogenes levels remained stable until T20, without significant differences. The anti-Listeria efficacy detected for protective culture A was consistent in all three experimental production batches. L. monocytogenes was not detected with the qualitative method. Regarding samples added with protective culture B (CLB samples), even though an anti- Listeria activity was detected, it was lower in all batches representative of the three producing plants compared to that observed in CLA samples (<1 log if compared to the experimental inoculation). In these types of samples, the pathogen was also detected at T20 with the qualitative method.

In positive control samples (CL samples) L. monocytogenes was detected using both the quantitative method (T0, T1 and T6 analysis times) and the qualitative method (T0, T1, T6 and T20 analysis time).

As expected, LAB mean levels were significantly higher (P<0.01) in samples added with the protective cultures A and B respect to control samples at T0, T1 and T6, but comparable at T20. Also micrococci and CNS showed higher mean levels in samples added with the protective culture at T0 and T1, while at the following analysis time, the trend was more irregular. These differences in LAB, micrococci and CNS levels are probably due to the use of the protective cultures that, as said, consisted of microorganisms belonging to the genera Lactobacillus and Staphylococcus. Enterobacteriaceae mean counts in control samples were consistent, with values between 3 and 4 log10 CFU/g in all analysis times; on the other hand, counts were lower in samples added with the protective culture A respect to control samples with significant differences (P<0.05) starting from T1, at T6 and T20. Results regarding L. monocytogenes, LAB, micrococci, CNS and Enterobacteriaceae counts in challenge test are reported in Table 2.

Table 2.

L. monocytogenes, LAB, micrococci, Coagulase Negative Staphylococci and Enterobacteriaceae mean values (log10 CFU/g; x ®± S.D.) in SFS samples during the challenge test.

Parameters	Samples	Analysis time
		T0	T1	T6	T20
Listeria monocytogenes	C	0 3	0 4	0 3	0 3
	CL	1.04 ± 0.62 b2	1.81 ± 0.78 a1	1.50 ± 0.92 ab1	1.01 ± 1.22 b1
	CLA	1.36 ± 0.48 a1	0.33 ± 0.78 b3	0.34 ± 0.70 b3	0.16 ± 0.43 b23
	CLB	1.13 ± 0.64 a12	1.31 ± 0.63 a2	0.75 ± 0.74 b2	0.46 ± 0.63 b2
LAB	C	3.80 ± 1.14 c4	7.17 ± 0.59 b2	7.83 ± 0.42 a3	7.75± 0.46 a1
	CL	4.48 ± 0.93 c3	7.23 ± 0.47 b2	7.84 ± 0.50 a3	7.78 ± 0.36 a1
	CLA	6.56 ± 0.35 b1	8.16 ± 1.88 a1	8.24 ± 0.26 a2	7.70 ± 0.57 a1
	CLB	6.17 ± 0.47 b2	8.21 ± 0.34 b1	8.46 ± 0.34 a1	7.83 ± 0.32 c1
Micrococci and CNS	C	3.04 ± 0.74 c3	5.53 ± 0.50 bc12	5.99 ± 0.65 ab1	6.03 ± 0.73 a1
	CL	3.22 ± 0.70 c3	5.06 ± 1.30 b3	5.69 ± 0.46 a1	5.37 ± 0.83 ab2
	CLA	4.75 ± 0.77 b1	5.72 ± 1.30 a1	5.84 ± 0.66 a1	5.86 ± 0.64 a1
	CLB	4.07 ± 0.65 b2	5.49 ± 0.52 a12	5.73 ± 0.85a1	5.80 ± 0.73 a12
Enterobacteriaceae	C	2.44 ± 0.83 b1	3.55 ± 0.50 a12	2.96 ± 1.50 ab12	3.79 ± 1.80 a1
	CL	2.58 ± 0.96 c1	3.47 ± 1.21 ab12	4.15 ± 0.96 a1	2.99 ± 1.94 c12
	CLA	2.78 ± 1.02 a1	3.01 ± 1.31 a2	2.78 ± 1.69 a23	1.87± 1.93 b3
	CLB	2.83 ± 0.57 b1	3.81 ± 0.75 a1	2.21 ± 1.77 b3	2.56 ± 1.70 b23

C: negative control samples; CL: positive control samples; CLA: samples added with protective culture A and L. monocytogenes broth culture of; CLB: samples added with protective culture B and L. monocytogenes broth culture; T0: after stuffing; T1: 24h after stuffing; T6: 6 days after stuffing; T20: 20 days after stuffing (end of ripening). Means in the same row with different superscript letter were significantly different (P< .05); means in the same column among biopreservative treatments with different superscript number were significantly different (P<0.05).

pH values (± SD) were 5.48±0.09 for the final products without protective culture (C and CL) and 5.30 ± 0.01 for those added with the culture (CLA and CLB). The aw (± SD) was 0.83±0.01 for C and CL samples and 0.84±0.01 for CLA and CLB samples. The composition analysis of control samples (C and CL) showed, in the final products, average percentage (%) values of 33.06±2.25 for fat, 23.53±1.16 for protein, 38.51±4.24 for moisture. Samples added with protective cultures had mean percentage values (% ± SD) of 31.03±0.65 for fat, 37.74±1.05 for protein and 27.10±1.65 for moisture. pH, aw and composition results (± SD) in SFS samples during the challenge test are reported in Table 3.

Table 3.

pH, aw and physico-chemical mean values (x ®± S.D.) in SFS samples during the challenge test.

Parameters	Samples	Analysis time
		T0	T1	T6	T20
pH	C	5.72 ± 0.15	5.63 ± 0.15	5.41± 0.35	5.54 ± 0.30
	CL	5.70 ± 0.14	5.62 ± 0.13	5.30 ± 0.32	5.43 ± 0.34
	CLA	5.69 ±0.13	5.41 ± 0.23	5.19 ± 0.31	5.31 ± 0.27
	CLB	5.69 ± 0.14	5.41 ± 0.12	5.19 ± 0.27	5.31 ± 0.27
aw	C	0.975 ± 0.005	0.977 ± 0.004	0.946 ± 0.012	0.830 ± 0.040
	CL	0.979 ± 0.006	0.979 ± 0.004	0.948 ± 0.010	0.846 ± 0.047
	CLA	0.979 ± 0.003	0.979 ± 0.005	0.945 ± 0.009	0.846 ± 0.047
	CLB	0.978 ± 0.004	0.980 ± 0.004	0.945 ± 0.009	0.838 ± 0.047
Fats (%)	C	13.95 ± 2.21	14.22 ± 2.78	20.87 ± 2.59	34.65 ± 3.95
	CL	14.42 ± 2.65	13.94 ± 2.27	20.85 ± 1.30	31.47 ± 1.95
	CLA	14.49 ± 2.60	13.58 ± 1.76	20.66 ± 1.20	30.58 ± 1.42
	CLB	14.07 ± 2.06	13.31 ± 1.94	20.68 ± 1.81	31.49 ± 1.14
Moisture (%)	C	65.30 ± 1.62	64.76 ± 2.21	46.10 ± 0.82	20.53 ± 1.00
	CL	65.00 ± 1.84	65.05 ± 1.89	46.52 ± 0.15	26.54 ± 4.91
	CLA	65.01 ± 1.80	65.60 ± 1.57	46.71 ± 0.88	28.27 ± 4.63
	CLB	65.35 ± 35	65.70 ±1.69	46.45 ± 0.11	25.93 ± 5.23
Proteins (%)	C	19.44 ± 2.74	17.93 ± 1.40	27.30 ± 1.69	39.34 ± 6.43
	CL	18.17 ± 1.08	17.91 ± 1.14	26.82 ± 2.38	37.69 ± 7.10
	CLA	18.11 ± 1.08	18.15 ± 1.01	27.49 ± 1.62	37.00 ± 6.70
	CLB	18.22 ± 1.08	18.15 ± 1.01	27.46 ± 2.01	38.49 ± 7.89

C: negative control samples; CL: positive control samples; CLA: samples added with protective culture A and L. monocytogenes broth culture of; CLB: samples added with protective culture B and L. monocytogenes broth culture; T0: after stuffing; T1: 24h after stuffing; T6: 6 days after stuffing; T20: 20 days after stuffing (end of ripening).

Use of protective culture into the production process

L. monocytogenes was never detected neither in samples with the addition of protective culture A nor in control samples analyzed at the end of ripening (qualitative and quantitative method). Analyses showed the presence of Listeria spp. (qualitative method) in control samples of n. 2 producing plants, in all production batches; however, in samples added with the protective culture A, Listeria spp. was never detected. Enterobacteriaceae mean levels were 2.31±1.64 log10 CFU/g in control samples and 2.14±1.62 in samples added with the protective culture. LAB, micrococci and CNS, showed higher mean levels in samples added with the protective culture, with values between 7-8 log10 CFU/g for LAB and 4-5 log10 CFU/g for micrococci and staphylococci. However, no significant differences (P>0.05) for LAB, micrococci and CNS mean levels were found between control samples and samples added with protective cultures. Results regarding L. monocytogenes, LAB, micrococci, CNS and Enterobacteriaceae counts are reported in Table 4. pH showed mean values (± SD) of 5.67±0.33 for samples without protective culture and of 5.46±0.19 for samples added with the culture. The aw (± SD) was 0.891±0.03 for samples without protective culture and 0.897±0.02 for samples added with the protective culture. The composition analysis showed average percentage values (% ± SD) of 35.62±3.55 for moisture, 25.68±7.09 for fat, 32.01±6.04 for protein in samples with the protective culture. Similar results were found as regards the composition of samples without the protective culture: mean values of 35.95±4.50 for moisture, 26.63±7.03 for fat, and 31.89±6.66 for protein.

Table 4.

L. monocytogenes, LAB, micrococci, Coagulase Negative Staphylococci and Enterobacteriaceae mean values log10 CFU/g; x ® ± S.D. (positive samples/total) in SFS samples produced without and with the addition of protective culture A.

Producing plants	Samples			Parameters
		Listeria monocytogenes	LAB	Micrococci and CNS	Enterobacteriaceae
P1	C	0	7.69 ± 0.17 (9/9)1	5.39 ± 0.37 (9/9)2	2.58 ± 1.14 (7/9)1
	A	0	7.53 ± 0.24 (9/9)1	5.93 ± 0.41 (9/9)1	2.35 ± 1.37 (7/9)1
P2	C	0	7.89 ± 0.09 (9/9)2	5.22 ± 0.28 (9/9)2	3.91 ± 0.43 (9/9)1
	A	0	8.29 ± 0.29 (9/9)1	5.55 ± 0.17 (9/9)1	3.06 ±0.25 (9/9)1
P3	C	0	7.89 ± 0.21 (9/9)1	5.77 ± 0.53 (9/9)1	0.67 ± 1 (3/9)1
	A	0	7.72 ± 0.42 (9/9)1	5.56 ± 0.35 (9/9)1	0.23 ± 0.67 (1/9)1

C: control samples; A: samples added with protective culture A. Means in the same column among biopreservative treatments with different superscript number were significantly different (P<0.05).

pH, aw and composition results (± SD) in SFS samples are reported in Table 5.

Table 5.

pH, aw and physico-chemical mean values (x ®± S.D.) in SFS samples produced without and with the addition of protective culture A.

Producing plants	Samples	Parameters
		pH ± SD	Aw ± SD	Ashes (%)	Fats (%)	Proteins (%)	NaCl (%)	Moisture (%)
P1	C	5.76 ± 0.20	0.867 ± 0.03	4.74 ± 0.62	20.50 ± 4.60	39.43 ± 5.62	5.26 ± 0.38	34.63 ± 6.01
	A	5.67 ± 0.22	0.877 ± 0.03	4.63 ± 0.51	18.40 ± 1.64	38.69 ± 4.48	5.06 ± 0.36	37.70 ± 3.45
P2	C	5.70 ± 0.18	0.921 ± 0.006	5.30 ± 0.31	25.09 ± 0.91	26.85 ± 0.84	4.05 ± 0.10	40.53 ± 1.57
	A	5.43 ± 0.15	0.926 ± 0.004	5.28 ± 0.29	26.08 ± 0.87	26.92 ± 0.65	4.04 ± 0.10	40.07 ± 1.45
P3	C	5.32 ± 0.08	0.885 ± 0.032	4.27 ± 0.70	34.31 ± 6.34	29.38 ± 1.33	4.57 ± 0.48	31.69 ± 6.72
	A	5.30 ± 0.07	0.889 ± 0.040	4.30 ± 0.85	32.56 ± 6.18	30.44 ± 2.53	4.50 ± 0.42	33.08 ± 7.42

C: control samples; A: samples added with protective culture A.

Discussion

There are several sources of L. monocytogenes contamination during the manufacturing process of fermented sausages. Raw meat may be contaminated at the slaughterhouse and during the production process of sausage in particular when good manufacturing and hygienic practices (GMP and GHP) are not correctly applied. Afterwards, the pathogen may survive during the shelflife of the product due to its high tolerance to low pH conditions and high salt concentrations, especially if the standard hurdle technologies are not correctly applied, in particular when short ripening times are used (Meloni et al., 2014; Mureddu et al., 2014; Mataragas et al., 2015). Therefore, it is necessary to provide the SFS with protection against the growth of pathogens. A possible strategy is the use of bioprotective cultures aimed to control L. monocytogenes multiplication.

In our study, the in vitro analysis allowed to test the efficacy of different protective cultures and identify the ones with a stronger anti-Listeria activity.

The challenge test results showed that the use of protective culture A allowed to control L. monocytogenes growth in SFS samples: L. monocytogenes showed a reduction of 1 log approximately in the first 24 h after production. At the end of ripening, L. monocytogenes was not detectable using the qualitative method. The anti- Listeria efficacy was uniformly detected in all three production batches.

The species included in the composition of the bioprotective culture A were Lactobacillus sakei, Pediococcus acidilactici, Staphylococcus carnosus and Staphylococcus carnosus subsp. utilis. These microorganisms have previously shown to be able of inhibiting L. monocytogenes growth in meat products; in particular, Lactobacillus sakei demonstrated an antimicrobial effect due to its capacity to produce organic acids, hydrogen peroxide and bacteriocins (Zagorec et al., 2017; Pedonese et al., 2020). Hugas et al. (1995) demonstrated the ability of L. sakei to inhibit the growth of L. monocytogenes in a model sausage system and in dry fermented sausages. Similar results have been observed by other Authors with regard to the bacteriocins produced by Pediococcus acidilactici (Nielsen et al., 1990; Nieto- Lozano et al., 2006). S. carnosus is not usually considered able to produce specific growth inhibitors against pathogens and its addition to meat products is usually aimed at improving the flavor and color (Tjener et al., 2004; Janssens et al., 2013). However, co-cultures of S. carnosus and L. sakei can assure the quality of raw-cured meat products based on metabolic functions of both species. Tjener et al. (2004) and Leroy et al. (2005) reported that L. sakei co-cultured with S. carnosus in meat products generated an effective fermentation process in sausage manufacturing in which S. carnosus plays a fundamental role as a competitor, especially in the first 48 h of fermentation, while L. sakei maintains its effectiveness during the following stages of ripening. Therefore, in this study, the observed reduction in cell concentration of L. monocytogenes is most likely due to the mutualistic relationship between the microbial components of the protective culture used, the production of organic compounds and bacteriocins. This finding is concordant with what was reported by Blanco-Lizarazo et al. (2016), who performed an in vitro test to evaluate the microbial interactions between L. monocytogenes, L. sakei and S. carnosus in conditions that mimic fermentation in meat; the Authors concluded that L. sakei and S. carnosus can successfully be used in protective cultures thanks to their ability to dominate the competition for nutrients and control L. monocytogenes growth.

As regards pH and aw evaluated during the challenge test, values reported were consistent with what was found by other Authors and typical of the product (Meloni et al., 2013; Piras et al., 2019). Significant differences were found in samples added with the protective cultures compared to control samples (negative control) starting from T1 (P<0.01), at T6 (P<0.05) and also at T20 (P<0.05). Moreover, in samples added with protective cultures A and B, a slight decrease of pH mean values was observed at analysis times T6 and T20, which is compatible with an increase in lactic acid production by the cultures’ microorganisms (LAB and Coagulase Negative Staphylococci). This decrease was statistically significant (P<0.01) in CLA samples between T0 and T1, and in CLB samples between T0 and T1 and T1 and T6 (P<0.01). In this regard, Leroy and De Vuyst (2005) indicated that L. sakei had a higher production rate of lactic acid, compared with other LABs with a bioprotection potential, such as L. curvatus and L. amylovorus. At the end of ripening time (T20) a slight increase in pH mean values,, was detected in all kind of samples, and was significant in CLB samples (P<0.01). Such increase is most likely due to the proteolytic effect of yeasts and molds, which play an important role in proteolysis and lipolysis to develop aroma during the manufacturing process and use lactic acid as substrate at the end of the drying step (Thévenot et al., 2005). However, the composition analyses have shown that the use of protective cultures in the production process of SFS does not affect the typical composition characteristics of the product.

Following the results obtained in the challenge test, the protective culture identified as the most effective against the growth and persistence of L. monocytogenes was supplied to three producing plants in order to include the selected culture into their standard production process and evaluate its effect on the chemical, physical and microbiological characteristics of the SFS at the end of ripening. The use of protective cultures has proved to be easy and quick (freeze-dried mixture of selected microorganisms) to use; furthermore, it does not require modifications of the regular production process steps, as the culture can simply be added into the mixture before the stuffing phase. The economic commitment for the producing plants is therefore limited and adequate to what is required by the Food Sector Operators. The microbiological analyses at the end of ripening (ca. 20 days) showed that, in the samples added with the protective culture A, neither L. monocytogenes nor Listeria spp. were detected (quantitative and qualitative method). This data confirms the results obtained by the challenge test, demonstrating that the use of protective cultures represents a technological innovation capable of eliminating L. monocytogenes or reducing its levels within the limits set by European Regulations (Reg. CE n.2073/2005) in meat products. Microbiological analyses aimed at determining the prevalence of non-pathogenic microorganisms showed that the addition of the protective culture A in the production process of the three processing plants, influenced Enterobacteriaceae counts levels that were always lower in samples added with the protective cultures, although this difference was never significant. Moreover, Enterobacteriaceae levels were reduced during ripening until non-detectable levels in some samples at the end of processing after the addiction of the protective culture. These results are consistent with those reported by Martìn et al. (2021), who demonstrated the decrease in Enterobacteriaceae counts until nondetectable levels at the end of ripening in a dry-cured fermented sausage inoculated with L. sakei.

As regard physicochemical- parameters evaluated after application in the production process, it was shown that while the presence of the protective culture does not affect final aw values, it does influence, although not significantly (P>0.05), the final pH values with an acidifying effect, compatible with an increase in the production of lactic acid by the protective cultures’ microorganisms. The technological innovation has therefore allowed stronger acidification of the product, which represents a further obstacle for the growth of pathogens, improving the product from a safety point of view. The composition analysis, carried out using NIT technology, showed similar values in samples with and without protective culture, highlighting how the inclusion of this technological innovation in the SFS production process does not affect the typical composition characteristics of the finished product.

The results of this project validated the efficacy, both in vitro and in the product, of some protective cultures to reduce or inhibit the growth of L. monocytogenes. Furthermore, the data obtained will allow SFS producing plants to define, with scientific evidence, the healthiness and hygiene of the product added with specific protective cultures.

Conclusions

The use of protective cultures within the standard production process of the SFS represents a significant technological innovation for the producing plants that allows controlling, for the whole commercial life, the possible growth of L. monocytogenes up to values that are not compatible with the current legislation (Reg. 2073/2005), ensuring a high hygienic and sanitary quality of the product. It has been demonstrated that protective cultures control the growth of L. monocytogenes without altering the typical composition, microflora and physico-chemical characteristics of the SFS. Finally, this techological innovation is simply and ready to use, it is not expensive for the Food Business Operators and does not require modification and/or implementation of the production process, since protective cultures can be directly added during the normal mixing step of fat, meat, spices and other ingredients.