Evaluation of microbiological, chemical, and sensory properties of cooked probiotic sausages containing different concentrations of astaxanthin, thymol, and nitrite

Abstract In this study, the effects of different concentrations of thymol and astaxanthin on control of Clostridium perfringenes and also microbial, chemical, and organoleptic properties of common and probiotic beef cooked sausages containing two levels of nitrite during storage at refrigerated condition during 45 days were evaluated. Based on findings, control group had significantly higher total volatile base nitrogen (TVB‐N) than nitrite‐, thymol‐, and astaxanthin‐treated samples. At the end of the storage time in control, thiobarbituric acid reactive substances (TBARS) value reached 1.96 mg/kg, while the values for treated samples remained lower than 1.63 mg/kg. Final count of lactic acid bacteria decreased approximately 1.67–3.79 log CFU/g in treated samples compared with the control group (p < .05). A reduction between 1.46 and 2.46 log CFU/g in C. perfringenes count was recorded for the treated samples in comparison with control group after 45 days of storage.

heat. Unfortunately, it is difficult to inactivate spores of C. perfringens, because of their resistance to heat (survive at 100°C for ≤1 hr).
Astaxanthin, as a keto-carotenoid, seems to be a very promising novel natural additive with a high potential application in food processing. Recent data reveal that the addition of astaxanthin to food, directly and indirectly, improves its oxidative stability . Furthermore, it has been declared that astaxanthin has beneficial effects in the prevention and treatment of many diseases, such as chronic inflammatory diseases, metabolic syndromes, and certain cardiovascular, gastrointestinal, and liver diseases (Leite et al., 2010;Naito et al., 2004;Uchiyama et al., 2002).
Astaxanthin is categorized as GRAS (Generally Recognized As Safe) by the US Food and Drug Administration (USFDA), and it has been considered as safe additive by the European Union Food Safety Authority (EFSA). Applying this compound as an additive or a packaging component can be introduced as a novel constitutes in meat products (Yang et al., 2013).
Thymol, also recognized as 2-isopropyl-5-methylphenol, a dietary monoterpene phenol, is one of the dominant components in thyme species. For centuries, plants containing thymol in their constituents have been used in traditional medicine (Meeran et al., 2017).
Thymol exhibits potent antimicrobial and antioxidant activities that is frequently used in food industry as a natural preservative (Karam et al., 2019;Meeran et al., 2017).
Nitrite is an additive commonly utilized in the preparation of sausages; this compound prevent the lipid oxidation, impeding the development of rancid off-flavors, producing pink color, and inhibiting the growth of spoilage and pathogenic bacteria, especially Clostridium spp. (Choi et al., 2017;Lee et al., 2019;Xiang et al., 2019). However, it has been demonstrated that high intake of nitrite results in a risk to human health as this compound is a well-known precursor of N-nitroso compounds classified as potent human carcinogens (Ma et al., 2018;Šojić et al., 2019). For this reason, the reduction or elimination of the use of nitrite in meat products is highly desirable. Therefore, research and evaluation of natural alternatives have been taken for the reduction of nitrite addition in meat products.
Probiotics are live microorganisms that is its sufficient administration has beneficial health effects such as increasing nutritional value of food, and improving immune system and digestive system function of the host. Although probiotics are mostly administered through dairy products, meat products can also be used as a probiotic carrier (Chugh & Kamal-Eldin, 2020;Khaledabad et al., 2020;Ryan et al., 2020;Zendeboodi et al., 2020;Zhu et al., 2020). In heattreated food products, probiotics are generally not used because of negative effect of thermal treatment on their viability and stability.
Integration of thermophilic probiotic microorganisms into meat products is one of the novel techniques. Hence, Bacillus coagulans can be used as probiotic due to their heat-resistant spore forms (Konuray & Erginkaya, 2018).
Clostridium perfringenes is one of the most important foodborne pathogens that lead to severe infection and even death. It was documented that vegetative cells of C. perfringens are susceptible to heat. Unfortunately, it is difficult to inactivate spores of C. perfringens, because of their resistance to heat (survive at 100°C for ≤1 hr).
Therefore, it is necessary to find new methods to control C. perfringens in meat products (Juneja et al., 2013;Lee et al., 2019;Limbo et al., 2010).
Taking into account the above explanations, the aim of the present work was to study the effect of the use of astaxanthin and thymol in relation to the reduction of nitrite in cooked probiotic sausages and to evaluate their influence on chemical, microbial, and sensory quality of the product and also the growth of C. perfringenes during the storage time.

| Microbiological analysis
Samples (25 g) were aseptically transferred to a stomacher bag and homogenized with 225 ml sterile peptone water (0.1% w/v) for 3 min.  (Lee et al., 2019). Total Enterobacteriaceae count were determined using Violet Red Bile Glucose Agar (VRBG Agar) which were incubated at 37°C for 24-48 hr. Data were recorded and then converted as log (CFU/g). In order to enumerate B. coagulans spore count, appropriate serial dilutions were exposed to heat shock at 80°C for 15 min. Afterward, cultivated on Tryptic Soy Agar (TSA) which was incubated at 37°C for 48 hr according to the method recommended by Somavat et al. (2013) (Somavat et al., 2013). Staphylococcu saureus detection was done, using International Organization for Standardization (ISO) 6888-3:2003 method.

| Chemical analysis
Moisture, protein, and fat were determined at the last day of storage according to the Official Methods of Analysis of AOAC International (Latimer & George Jr, 2016). TBARS (2-thiobarbituric acid reactive substances) and TVN (total volatile nitrogen) content of sausage samples were determined according to the method described by Alirezalu et al. (Alirezalu et al., 2019). For the pH measurements, the sample was homogenized and diluted with distilled ultrapure water at 1:10 ratio, and then, pH was measured using a HI221 Calibration Check Microprocessor pH meter (Hanna Instruments S.L., Spain).

| Sensorial evaluation
For assessing sensorial characteristics, sausage samples were evaluated at the end of the storage by an organoleptic panel conformed of trained members. Briefly, the panelists (5 people) were asked to score color, odor, flavor, and overall acceptance and desirability of sausage samples. Samples were evaluated according to the nine-point hedonic scale, from 0 (dislike extremely) to 9 (like extremely), and considering 5 as stands for limit of acceptance. Samples were labeled with 3-digit random numbers to avoid identification, and they were presented to the panelist in random order.

| Statistical and chemometrics procedures
Data were analyzed through one-way analysis of variance (ANOVA) In order to study the latent relationships (between samples and between variables) residing in the data for the different sausage formulations two display chemometric techniques that were applied to the obtained microbiological and chemical data include principal components analysis (PCA) and hierarchical cluster analysis (HCA). PCA is a chemometric procedure frequently applied for obtaining a primary multidimensional evaluation of the data set and for reducing the data dimension with minimal loss of useful information. PCA was done by decomposing the original data matrix as a product of two other matrices. The first one (score matrix) contains information about the samples, while the second matrix (loading matrix) includes information related to the variables. When the number of principal components considered to analyze the problem is smaller than the number of original variables, PCA simplifies the dimension of the problem and this allows the appropriate study of the original data matrix X in a reduced space (Jolliffe, 1986). HCA was used as a second display chemometric technique. This technique (often used jointly or complementarily with PCA to study the internal structure of a data set) is generally applied to the data matrix to search for natural groups of samples (or variables) on the basis of their distance in the multidimensional space. In the present study, sausage samples of the different formulations were hierarchically clustered according to the squared Euclidean distances between them. Clusters were calculated on the basis of agglomerative Ward's method (Massart, 1983

| RE SULTS AND D ISCUSS I ON
Beef cooked sausages as proteinaceous products are one of the popular meat products all over the world. Natural green preservatives (thymol and astaxanthin) could be used to enhance the shelf life and safety of these products.
Cooked probiotic meat products can be used to transmit probiotic bacteria. In addition, some researchers believe that the sausage matrix protects the survival of probiotic through the gastrointestinal tract. B. coagulans partially, and sodium nitrite substitutes in probiotic sausages, can produce approximately the same amount of myoglobin nitrosyl which may be present due to nitrite reductase activity in the microbial strain, which can reduce nitrite (De Vuyst et al., 2008;Zhu et al., 2020).  Figure 1a, 1b, 1c, and 1d, respectively.

| Microbiological analysis
Lactic acid bacteria (LAB) are one of the main spoilage microorganisms in the refrigerated proteinaceous products such as cooked sausage (Korkeala & Björkroth, 1997). The initial LAB value of control sample was found to be 1.17 log CFU/g. According to Feng et al. (2013), upper microbiological limit of LAB for acceptable quality cooked sausage is 7 log CFU/g (Feng et al., 2013). As shown in Figure  There was a significant decrease in C. prefringenes population from 2.78 to 2.01 log CFU/g at the 120 ppm nitrite, 250 ppm thymol, and 450 ppm astaxanthin (Figure 1c). use of thymol resulted in a significant reduction in C. perfringenes spore growth during exponential cooling of contaminated cooked turkey up to 21 hr (Juneja & Friedman, 2007).
Finally, the results of the counts of inoculated B. coagulans spore-forming bacteria in sausage samples during sampling days are summarized in Figure 1d. The obtained results showed a reduction in the spore count in the inoculated sausages. The lowest decrease rate was observed in control samples. This reduction in the count of B. coagulans spores was related to the effect of sausage ingredients especially those having impact on spore germination and outgrowth (such as the existence of sodium nitrite, salt, sodium ascorbate, thymol, and astaxanthin). However, the counts of the spores inoculated in samples were still above the recommended minimum daily therapeutic dose of spore probiotics (i.e., 10 6 CFU/g) during the entire refrigerated storage (Jafari et al., 2017).

| Chemical analysis
The results of the chemical analysis carried out in the samples of the different formulations for moisture, protein, and fat content at the end of the 45-day storage period are summarized in Table 2. When an analysis of variance was carried out, for the three cases of moisture, protein, and fat, the results of the P-value for the F test was less than 0.05, showing that there is a statistically significant difference between the averages of chemical variables among some of the different formulas, with a 95.0% confidence level. This is an expected result because, although the same raw materials were used for all formulations, these compositional parameters (moisture, protein, and fat) were slightly but clearly affected by the addition of natural antioxidants, antimicrobial compounds producing different microbial activity in the product, or in other words, the different quantities of astaxanthin, thymol, and nitrite, as well as the potential addition or not of B. coagulans leads to products with certain differences in chemical composition, and probably, as consequence, also in organoleptic properties.
This fact has been illustrated in the box-whisker plots for these variables presented in Figure 2a  oxidation under the current experimental conditions. In general, the lower and acceptable level of TBARS in the treated sausages could be related to higher levels of nitrite (120 ppm) combined with high levels of thymol and astaxanthin. This fact could be explained due to radical scavenging activities of these substances reducing the lipid oxidation (Cheng & Wu, 2019;Luna et al., 2017;Weintraub et al., 2017). These results of TBARS in sausage samples during storage at refrigerated condition consistent with the findings reported in previous studies (Lee et al., 2019;Šojić et al., 2015).
Finally, the pH changes in the sausages with different formulations are displayed in Figure 1g. As seen, pH value of formulation 13 decreased to 6.59 at the end of the storage period which is in good agreement with previous study (Šojić et al., 2015). Also, in this present study the highest decrease rate was found (p < .05) in the control samples, in comparison with the other formulations. The growth of lactic acid bacteria (LAB) is responsible for changes in pH during storage of vacuum-packaged sausages. Since the addition of thymol, nitrite, and astaxanthin may have affected the growth of LAB and reduced the production of lactic acid during storage (see Figure 1a), it seems reasonable to assume that this reduction can explain the slight pH decay during storage.

| Chemometric analysis
In order to evaluate the joint influence of microbiological and chemical variables, the two chemometric display procedures described in section 2.4 were applied to a X 76x9 data matrix in which the rows are In order to avoid the influence of the different size of the variables in the chemometric analysis, all variables in the X 76x9 data matrix were autoscaled, by subtracting each value from the mean of the variable and dividing it for the standard deviation of the variable. The result of X AUT78x9 matrix is a new matrix preserving the same chemical information than in the original one, but in this case all variables are of the same scale with 0 mean and 1 deviation standard (Deming et al., 1988). Therefore, latent relationships (between samples and between variables) and the internal structure residing in the X AUT78x9 data set were studied using two chemometric approaches: principal components analysis (PCA) and hierarchical cluster analysis (HCA).
When PCA was applied to the autoscaled data matrix, the first two principal components were attained for the 91.24% of the total data variance. This means that the data can be studied in a 2-dimensional space preserving more than the ninety percent of the total information contained in the data matrix. When the samples are evaluated in the 2D-score plot of principal component 1, PCOMP 1 (representing 81.65% of the total data variance), versus. principal component 2, PCOMP 2 (accounting for the 9.59 of data variance), three groups of formulations were specified (see Figure 3a). The first one, named A, in the negative part of PCOMP 1, included samples from formulations 7-13. The second one, in the positive part of both principal components, is formed by the formulations 2-6, coded as B, and finally based on the formula 1 (as control sample), a single group C appeared which clearly is separated from the other formulations.

| Sensorial evaluation results
As indicated in section 2.3, the sausage samples prepared according to the different formulations subjected to a sensory evaluation carried out by an expert panel composed of 5 members. Results of this sensorial analysis on the basis of color, odor, flavor, and overall acceptance are presented in Figure 5a. The appropriate reproducibility of the results between replicates indicated the satisfactory work of panelists. As it can be seen in Figure 5, formula 5 exhibited sensorial scores lower than the acceptance limit for all the parameters evaluated. For formulas 6 and 7, in spite of the fact that their overall acceptance is slightly higher than 5, several sensorial parameters are below the acceptance limit. On the other hand, formulations 11, 12, and 13 also obtained low scores in the range 5 to 6. PCA, HCA, and TCATA are techniques used to analyze sensory evaluation (de Souza Paglarini et al., 2020;Vidal et al., 2019). According to PCA, treatments with thymol concentration (more than 250 mg/ kg) have a low sensory score because thymol has a strong taste at high concentrations. On the other hand, formulations of 8, 9, and 10 which had obtained highest scores by panelist, all had low level of thymol (125 ppm), high level of Nitrite (120 ppm), and different quantities of astaxanthin between 0 and 450 ppm. In addition, as seen in Figure 5b, the overall acceptance score is also studied in relation to the protein and fat content. As it can be seen, the most valuable formulations in terms of sensorial score (formulas 8, 9, and 10) are those corresponding to slightly high-protein and slightly low-fat levels in comparison with the remaining formulas. According to the findings reported in previous studies, the use of probiotics in food can reduce the deterioration of sensory properties of food products (Karimi et al., 2012).

F I G U R E 4
Dendrograms obtained by HCA (squared Euclidean distance, and Ward's agglomerative method) for samples (a) and variables (b) with different formulations

| Sausage formulation selection
The selection of the better formulation was carried out by taking into account both chemical and microbiological analyses, as well as the results from sensory data. On the basis of microbiological and chemical determinations, among the formulations of cluster A, the formulation 13 (with maximum levels assayed for astaxanthin, thymol, and nitrate) should be selected because it allows for better results as indicated in the text and can easily be checked in Figure 3a.
However, when the results of sensory data were also considered, formulations 8, 9, and 10 were preferred (Figure 5a). By combining the two criteria, formulation 10 must be selected as optimum because (i) relation to the chemical and microbiological characteristics is highly similar, in practice, to the best one to be selected following this criterium, formula 13. In fact, the near position of the samples of both formulations 10 and 13 in the PCA score plot in Figure 3a demonstrate this statement; and (ii) on the basis of sensory data, formula 10 also belongs to the set of formulations (8, 9, and 10) clearly preferred by tasters. Thus, formula 10 (astaxanthin 450 ppm, thymol 125 ppm, and nitrite 120 ppm) was selected as optimum for future developments.

| CON CLUS ION
In summary, among the different treatments in this study, and among sausages formulated containing different concentrations of nitrite, thymol, and astaxanthin, a promising approach was suggested for limiting microbial growth and decreasing TBARS values in the sausages. Furthermore, a synergistic antibacterial effect observed between thymol and nitrite, with the highest concentrations of thymol, astaxanthin, and nitrite showed the best antibacterial activity. However, high concentration of thymol leads to adverse organoleptic properties. Based on our findings, it can be argued that integration of thymol and astaxanthin into the sausage formulation successfully improved the quality of beef cooked sausage during storage in refrigerator condition. The evidence from this study and similar studies suggests that thymol and astaxanthin can play a favorable role in enhancing cooked sausage quality.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The researchers declare that there are no conflicts of interest.

ACK N OWLED G M ENTS
The researchers gratefully acknowledge the support of University of