Portella, A. C. et al. 10
Vol. 3, N. 1: pp. 10-17, February 2012 ISSN: 2179-4804
Journal of Biotechnology and Biodiversity
Studies of antagonistic effect between Lactobacillus sakei on Escherichia coli, Listeria monocytogenes and Staphylococcus aureus
Augustus Caeser Franke Portella1*, Raimundo Wagner de Souza Aguiar1, Jefferson da Luz Costa1,2, André Luís Lopes da Silva2 and Gessiel Newton Scheidt 1
1Departamento de Ciências Agrárias e Tecnológicas; Universidade Federal do Tocantins; 77402-970; Gurupi - TO - Brasil .
2Departamento de Engenharia de Bioprocessos e Biotecnologia; Universidade Federal do Paraná; 81531-970; Curitiba - PR - Brasil.
ABSTRACT
The antimicrobial activity of a commercial probiotic culture, Lactobacillus sakei(ATCC 1521) at following concentrations (30, 42, 60, 78 and 90 µg/ml, 105 - 107UFC/mL), temperatures of interaction (4, 10, 20, 30, 37 °C)
and initial pH (3.5, 4.0, 4.5, 5.0 and 5.5) were tested against three foodborne pathogens, Escherichia coli, Listeria monocytogenesand Staphylococcus aureus. The antagonistic effect of the probiotic culture in vitro was performed by liquid microdilution method. The results indicated that the inhibitory substance present on a 24 hours culture broth could be an advantage when keeping the culture dominant during longer fermentations. For the highest lactic acid production, the incubation period of lactic acid bacteria (1.04 %v/v) was on MRS Broth in aerobic conditions, at 37 ºC/24 hours, which gave a minimum pH value of the supernatant (3.5). The data suggest that supernatant can have significant bacteriostatic activity against E. coli, L. monocytogenesand S. aureus, and may provide curedmeats with a degree of protection against this microorganism, particularly if employed with a combination of acid pH, and adequate refrigeration.
Key words: Antimicrobial activity, complex substrates, surface response
INTRODUCTION
In studies carried out on the effect and the way of action of substances with antimicrobial activity present on foods or added properties during the manufacture process, can be perceived that the
behavior of microorganism are determined by the specific characteristic of the culture conditions (i.e. water activity, temperature and pH) and, also,
for those inherent ones to the effect of the added cultures (Hurst, 1983).
The lactic acid bacteria (LAB) are examples of these cultures that not only have sensorial characteristics, but also allow a better conservation of foods being able to be used as probiotics also - alive bacteria in foods that after the ingestion
exerts beneficial effects to the host. The way of
action of probiotics is still unknown, however, some researches (Parker, 1974; Shah, 2000;Cebeci and Guarakan, 2003) are suggesting some
processes that can act independently or associated.
A previous study showed the competitive exclusion, in which the probiotic would compete with pathogens for nutritional sources, hindering
its action. The microorganisms most commonly
found in foods with probiotic action are the lactobacilli and the bifidobacteria.
There is a recommendation regarding human consumption food, indicating approximately 10 6
living microorganism per gram or mL of the product, at the moment of the consumption, for the
probiotic to produce therapeutically benefits
(Klaenhammer, 2001). LAB are characterized as Gram positive, not sporulant, normally not-mobile and they produce lactic acid during the
fermentation processes using glucose. They grow anaerobically, however most of them are not
sensible to O2 presence and can grow under aerobic conditions also. Thus, they are called anaerobic aerotolerant (Brock et al., 1994).The monitoring of the microbial grow is the aim of
______________________________________________________
Author for correspondence: caesar_portella@hotmail.com
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
https://doi.org/10.20873/jbb.uft.cemaf.v3n1.portella
Portella, A. C. et al. 11
many researchers in the area of biotechnology being the majority, based on statistical models - laws of probabilities, looking for parametrizing the
sequence of identification, isolation and quantification of bacteriocin inhibiting activity in different physical-chemical and environmental
conditions (Vandenbergh, 1993; Koutsoumanis and Sofos, 2005). The adjust method commonly
used the univaried method, in which a factor varies during evaluation time. However, this method
presents disadvantages and is being gradually substituted by multivaried methods, in which diverse factors are evaluated simultaneously.
To combine statistical techniques to a biological interpretation is essential to describe and understand biological systems. A clear example is
the kinetic of growth and the inhibition of microorganism in a culture medium (Jarvis, 1989). There are situations that the use of multivaried methods becomes particularly important in
Broth), with the aid of a platinum holder, and incubated at 30 - 37 °C for 24 hours. After the incubation period, serial dilutions were performed
in which the inoculum was diluted in peptoned water 0.1%, until the desired concentration ofeach microorganism .
Effect of pH, heat treatment and cooling on antagonistic activity
To evaluate the thermal stability at different pH values (3.5, 4.0, 4.5, 5.0 and 5.5), supernatant cultures of L. sakeiwere submitted to thermal treatments (20 °C/24 h, 30 °C/24 h, 37 °C/24 h) and refrigeration (3 °C/24 h, 10 °C/24 h). The pH was adjusted to their respective values using solutions of NaOHorHCl 1N.
Inhibitory activity
The antagonistic effect against the target microorganisms was determined, initially, in 1.5
industrial conditions and routine analysis, mL cuvette at O.D.660nm using
therefore they imply in time and costs, so such
method is suitable for these applications only. The methodology of replying surfaces involves the use
of factorial planning that can be repeated some times, drawing the response surface to the direction of the region desired point of excellence .
MATERIAL AND METHODS
Microorganisms strains
The strainsused in this study were obtained from
American Type Culture Collection (ATCC): Lactobacillussakei (ATCC 1521),
Listeriamonocytogenes (ATCC 15313), Escherichia coli (ATCC 25922) and Staphylococcusaureus (ATCC 25923).
Media
The putative strain able to inhibit the pathogens,
Lactobacillus sakei, was cultivated in MRS agar with 0.3% ofaniline blue solution. L. sakei
colonies have the ability of metabolize aniline blue, being this the way to recognizing their
morphologic characteristics (Silva et al.,
2001).The plates were incubated at 37 ºC for 24 h for forward studies of its morphologic characteristics.
In order to activate the pathogens( L. monocytogenes, E. coli and S. aureus), they were
sown from the agar nutrient into test tubes containing 5mL ofMueller Hinton broth(M H
spectrophotometer (Spectrumlab 22PC). The
supernatant was added in equal volume of Mueller-Hinton broth.
Then, they were inoculated with 2% (v/v) of the indicator microorganism previously reactivated in MH broth and incubated at 37 ºC overnight. The
control sample was composed by MH:MRS ( Man et al., 1961) broth in 1:1 ratio inoculated with 2% of indicator microorganism culture, incubated at
37 ºC during 24 hours.
A second evaluation was performed on micro -
plates by microdilution broth technique on a wavelength of 660 nm using BioTek equipment,
model "Power Wave XS" and the KC Junior program.
The supernatant was added in equal volume of MH broth and inoculated with 10% (v/v) of
microorganism previously reactivated in MH
broth, incubated at 37 ºC, during overnight and diluted 1:10 (MH broth: distilled water). MH:MRS
broth in ratio 1:1 was used as blank and a control
was made with MH:MRS (1:1) inoculated with 10% (v/v) of each target microorganism.The temperature range of incubation experiments were
of two types: those conducted at a single cooling rate and those conducted at two cooling rates. The
dual-rate experiments followed one cooling rate from 3 °C to 20 °C and a second rate from 20 °C to 37 °C.
The inhibition of the growth of the target microorganism occurred when the absorbance was
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 12
lower than the control absorbance, indicating that the microorganism had a reduction of its growth in relation to the control, after 24 hours of i ncubation
(Chang et al., 2001).
Statistical assessment
To maximize the antagonism effect of Lactobacillus sakei supernatants a Full Factorial
Design for three independent variables was adopted. The Experimental Design was based on
Statistica5.0 (StatSofts Inc., Tulsa, OK, USA). Full Factorial Design was used to obtain the combination of values that can optimize the response within the region of the three dimensional observation spaces, which allows one to design a minimal number of experimental runs
(Box et al., 1978). The variables were supernatant concentration of Lactobacillus sakei, the temperature and pH, were submitted for the
analysis in the design.
The variable of each constituent at levels –1.68, –
1, 0 and +1, +1.68 is given in Table 1. The selection of low, middle and high levels for all these variables were based on a prior screening
done in our laboratory (unpublished data). A 2 3
full factorial design with using one central point and two axial points for each variable totalizing 15 experiments. The behavior of the present system described by the following equation 1, which includes all interaction terms regardless of their significance:
ŷ=bo +b1 x1 +b2 x 2 +b3 x3 +b1 2 x1 x 2 +b1 3 x1 x3 +b2 3 x 2 x3 (1)
Where ŷ is predicted response, i.e.absorbance values; x1, x2 and x3 are independent variables; b0 is
coefficient constant for offset term; b1, b2 and b3 are coefficient constant for linear effects and b12, b 13,
b23are coefficient constant for interactions ef fects. The model evaluates the effect of each independent variable to a response. The variables studied were supernatant (30, 42, 60, 78, 90 µ g/mL), temperature (3, 10, 20, 30, 37 °C) and pH (5.0, 5.5, 6.0, 6.5, 7.0).
The analysis of the factorial planning consists of quantifying the effect of the factors on one
determined reply. The studied variables were
supernatant concentration of Lactobacillus sakei , the temperature and pH, in its levels minimum and
maximum (Table 1) correspond to the variable x1 , x2 and x3 respectively in the analysis of time of life
of microorganism L. monocytogenes, E. coli and S. aureus .
Table 1. Optimization of physical condition for the antagonic effect of supernatants Lactobacillus sakei: independent variables in a 23 full factorial experiment desing. The parameters used in this
experiment are supernatant, temperature and pH.
Actual factor level at coded factor level of
Factor Symbol -1.68 -1 0 +1 +1.68
Supernatant (µg/ml) Temperature (°C)
X 1
X 2
30
3
42
10
60
20
78
30
90
37
pH X3 5.0 5.5 6.0 6.5 7.0
Surface response and optimal bacterial growth
Surface response represents three- dimensional graphic which show the variation of the experimental reply in function of alterations in the
levels of two selected variable, as for example, the temperature and pH. There will have other
variable and it will be fixed in one determined level and the equation will have to be rearranged
to get it respective reply surface. In the case of
equations 1, the variable x3 was fixed in level zero (in spaced out terms), correspondent to the central
point, but it could have been fixed in any another level of interest. The use of level zero does not mean, however, that pH was adjusted for this
value, on the other hand, in the level zero pH of the supernatant was adjusted for was of 5.5, thus
the equation is simplified and allows the representation in three dimensions.
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 13
RESULTS AND DISCUSSION
The supernatant of L. sakeipresented d ifferent sensitivities to temperature and pH, this trial
evaluated the effects of different values of pH
(from 3.5 to 5.5), cooling temperatures (4 °C during 24 h, 10 ºC during 24 h) and heat treatment (20 °C during 24h, 30 °C during 24 h, 37 °C
during 24 h) (Table 2).
Table 2. Effect of pH and temperature on the absorbance of supernatantofLactobacillus sakei . Absorbance results
pH Control 4°C/24h 10°C/24h 20°C/24h 30°C/24h 37°C/24h
5.5 0.350 0.356 0.527 0.543 0.623 0.560
5.0 0.389 0.402 0.508 0.604 0.645 0.545
4.5 0.540 0.518 0.538 0.558 0.548 0.560
4.0 0.640 0.494 0.623 0.521 0.622 0.538
3.5 0.639 0.582 0.601 0.595 0.597 0.585
The number of cells of L. monocytogenes, E. coli and S. aureus presented in sample test (inoculated with the crop of L. sakei), as well in the sample
control (not inoculated with the culture L. sakei ), showed little variation until the third hour (Figure
1). From the sixth hour of inoculation was
observed that the population of pathogenic microorganism showed an increase in the counting
of control samples for the test samples.
During lactic acid bacteria growth, it was observed a pH value drop, turning the environment very
acid, probably due to production lactic acid. It can be determined the inhibition of other microorganism. The results of the growth of L.
sakei in MRS broth, to 37 °C in aerobic, had a minimum value of the pH supernatant, around
3.90, after 24 hours of incubation. The maximum
production of lactic acid verified, came after the same period of growing, reaching 1.04 g% (120
mmol) of acid lactic in the supernatant. Figure 1
show values of pH during the period of lactic bacteria incubation.
0,700 0,650 0,600 0,550
6,0
5,5
5,0
4,5
0,500
4,0
0,450
0,400 0,350 0,300
3,5
3,0
2,5
0 3 6 9 12 15 18 21 24 Time (hours)
Figure 1- Kinetics of pathogenic growth.
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 14
Fractionary matrix planning (Table 3) was constructed from a complete factorial planning
from three variables x1, x2 and x3 .
Table 3. Experimental design with experimental and predict values of pathogenics growth.
Run Supernatant X 1
Temperature X 2
pH
X 3
X 1
Factors X 2
X 3
O.D.66 0 n n
24h
1 42 10 5.5 - 1 - 1 - 1 0.506
2 78 10 5.5 +1 - 1 - 1 0.519
3 42 30 5.5 - 1 +1 - 1 0.509
4 78 30 5.5 +1 +1 - 1 0.531
5 42 10 6.5 - 1 - 1 +1 0.506
6 78 10 6.5 +1 - 1 +1 0.513
7 42 30 6.5 - 1 +1 +1 0.525
8 78 30 6.5 +1 +1 +1 0.625
9 30 20 6.0 - 1.68 0 0 0.513
10 90 20 6.0 1.68 0 0 0.619
11 60 3 6.0 0 - 1.68 0 0.520
12 60 37 6.0 0 1.68 0 0.621
13 60 20 5.0 0 0 - 1.68 0.505
14 60 20 7.0 0 0 1.68 0.587
15 60 20 6.0 0 0 0 0.655
This study emphasized the statistic experimental analysis (the significance of the factors were
tested using the analysis of variation, F-test and t -
test), the use of the graphic method (Pareto´s chart) and the interpretation of the interactions among variables. Through these results, if
necessary, the model was refined, excluding
irrelevant variables. Pareto´s chart (Figure 2) showed that the model needed adjustment in terms
of supernatants, temperature and pH, a variation towards lower pH was found to be necessary (since the pH bar crossed the red pointed line
indicating that this result is not within the 95% confidence level).
These results can be attributed to the levels of variables, which were probably close to optimal levels. The distance between the quantitative lower
(-) and the highest level (+) of both variables, was probably insufficient to promote significant
differences. However, by the evaluation of the values of extinction, growth in all microbiological testing of samples in the central point (test 15) of
planning (Table 2), especially in relation to the attributes temperature and pH, are noticed.
Figure 2 -Paretto´s charts .
The model was optimized using the values of absorbance as described in (Table 4). ANOVA was used for the adjustment of the variance model
analysis and to verify the significance of the regression and if exist evidences of lack of adjustment. The results of variance analysis
(ANOVA) to describe answers are showed in (Table 5), where it can be noticed the correlation coefficient and the percentage of variance, explaining 89,07% of variance.
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 15
Table 4.Variance analysis (ANOVA) for the factorial design
Source of Variation
Sum of Square
Degrees of Significance
Mean Square
F- ratio
(model Significance) p- value
(1) Supernatant(L) 0.007508 1 0.007508 8.38159 0.033984
Supernatant(Q) 0.008916 1 0.008916 9.95268 0.025243
(2) Temperature(L) 0.007303 1 0.007303 8.15274 0.035602
Temperature(Q) 0.008191 1 0.008191 9.14362 0.029287
(3) pH (L) 0.004284 1 0.004284 4.78166 0.080431
(3) pH (Q) 0.012509 1 0.012509 13.96342 0.13476
1(L) by 2(L) 0.001301 1 0.001301 1.45173 0.282162
1(L) by 3(L) 0.00648 1 0.00648 0.72336 0.433874
2(L) by 3(L) 0.001682 1 0.001682 1.87760 0.228928
Error 0.004479 5 0.000896
Total SS 0.040983 14
R2 0.8907; Adj 0.6939
Table 5. Factorial designs main effects and interactions analysis for antagonistic between Lactobacillus sakei on Escherichia coli, Listeria monocytogenes and Staphylococcus aureus
Factor Effect Std. Error t-value p- value
Mean/Interc. 0.660089 0.029743 22.19310 0.000003**
(1) Supernatant(L) 2.891638 0.016205 2.89510 0.033984**
Supernatant(Q) -3.14787 0.024352 -3.15479 0.025243**
(2) Temperature(L) 2.851404 0.016205 2.85530 0.035602**
Temperature(Q) -3.0189 0.024352 -3.02384 0.029287**
(3) pH (L) 2.183716 0.016205 2.18670 0.080431
(3) pH (Q) -3.73057 0.024352 -3.73677 0.013476**
1(L) by 2(L) 1.203235 0.021164 1.20488 0.282162
1(L) by 3(L) 0.8493424 0.021164 0.85050 0.433874
2(L) by 3(L) 1.368385 0.021164 1.37025 0.228928
(L): linear factor; (Q): quadratic factor; *Significant factors (p < 0.05); **Significant factors (p < 0.10) .
After the calculation of coefficient errors and reliable intervals, it was verified in the below
equation that describes the power of inhibition in these conditions:
ŷ=- 0.66- 2.916x1 - 3.14787X1 2 +2.8514x2 - 3.0189X2 2 - 3.73057x3 2 (1)
Thus, one can say that this model showed no
statistical significance and that it cannot be used for predictive purposes.
The scale of Figure 3clearly shows the relation among temperature in its higher level (37 °C) and the concentration of the supernatant used in both the levels. Both variable influenced positively in the increase of the absorbance causing a reduction in kinetic of the growth of the microorganism
when using refrigeration temperature (8 °C), and
the supernatant in its lesser precipitated concentration can be noticed of (30 mL).
Also in Figure 2, the Zaxis refers to predict absorbance yield, values in contour plot represent
absorbance gradients for corresponding
supernatant and temperature level. Hollow squares in surface plot represent values of supernatant and
temperature for corresponding experiments.
As from the data of Table 2 is possible verify that the increase in some of the variable will not increase the value of the analytical reply, will not be necessary to carry out an ascension to the maximum to locate the excellent experimental
conditions, therefore these already are enclosed in
our planning. It is now enough to select the condition where the difference enters the answers of the adjusted models is maximum, this can
directly be made using equations 2 and 3 (this
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 16
comparison is facilitated, therefore the error standard of the coefficients of the model is very next), or with the aid of (Figure 4).
The more favorable experimental condition for the analytical reply is to use the lesser concentration of the variable x3 (pH) since in the tested interval, the increase in the levels of this changeable only
increases the values of absorbance. Thus, we can use level -1 (that correspond pH 5).
Figure 3 - Response surface plot describing effect of supernatant and temperature (°C).
Figure 4- Response surface plot describing effect of supernatant and pH .
Another possibility is to raise the levels of the
variable that contributes more for the increase of the analytical reply (bigger coefficient in the
equation or greater inclination the reply surface). In this in case that, x1 is about the variable (supernatant), that it can be fixed in +1 (that corresponds to 90 mL). This condition is not the ideal, therefore the absorbance of the kinetic growth also increases, still with less intensity, but
in then it does not remain another alternative. Yet,
the variable x2 (temperature) presents practically the same contribution in the two models, be ing impossible to define a condition clearly favorable
to the analytical reply (Figure 5).
Figure 5- Response surface plot describing effect of temperature (°C) and pH.
CONCLUSIONS
The factorial planning demonstrated to be suitable for evaluating the average effect of the factors on
the kinetic of growth of the pathogen microorganism. Inhibition of undesired pathogenic
microorganism by LAB may be due to the effe ct
of one, or synergism between several mechanisms. It was possible to observe strong influence of three factors that had inhibited the growth: supernatant,
temperature, and pH.Despite the fact that this study appliedfractionary factorial planning to
investigate the influence of the factors on t he kinetic of growth pathogenic microorganisms .
It was possible to observe groups of microstructures that they had formed in function of the factors of the planning, what demonstrated the utility of this statistical tool when is desired to
make exploratory research on variable whose effect is total or partially unknown. L. sakeiinhibited growth of a mixture of three strains of L. monocytogenes, E. coli and S. aureus used in our experiments.
These LAB strains should, therefore be adapted to
growth in the products and to survive in the production facilities.
RESUMO
A atividade antimicrobiana de uma cultura comercial probiótica deLactobacillus sakei (ATCC 1521) foi
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012
Portella, A. C. et al. 17
avaliada em termos de concentração celular (30, 42, 60, 78 e 90 mcg / ml, 105 - 107UFC/mL), temperatura de
incubação (4, 10, 20, 30, 37 ° C) e pH inicial do meio de cultura (3.5, 4.0, 4.5, 5.0 e 5.5) contra três patógenos alimentares, Escherichia coli, Listeria monocytogenes e Staphylococcus aureus. O efeito antagônico da cultura probiótica in vitro foi realizada através da aplicação de método quantitativo espectrofotométrico. Os resultados indicaram que uma substância inibitória presente em caldo de cultura de 24 horas poderia ser vantajoso quando mantendo a cultura dominante durante fermentações mais prolongadas. Para a maior produção de ácido láctico (1,04%v/v), o período de incubação da bactéria em caldo MRS foi de 24 horas em condições aeróbicas e a 37 ºC, com um valor de pH mínimo do
sobrenadante (3,5). Os dados sugerem que o sobrenadante pode ter atividade bacterio stática significativa contra E. coli,L. monocytogenes e S. aureus, e pode fornecer carnes curadas com um grau de
protecção contra este microorganismo, particularmente se empregada com uma combinação de pH ácido, e de refrigeração adequada.
Palavras-chave:Atividade antimicrobiana, substratos complexos, superfície de resposta
REFERENCES
Box, G. E. P.; Hunter, W. G.; Hunter, J. S. (1978), Statistics for experimenters: An introduction to design, data analysis, and model building. In
Bradley, R. A.; Hunter, J. S.; Kendall, D. G.; Watson, G. S. Eds., Wiley series in probability and mathematical statistics, Wiley, New York, 653p.
Brock, T. D.; Madigan, M. T.; Martinko, J. M.;
Parker, J. (1994),Biology of Microorganisms. New Jersey: Prentice-Hall, 909p.
Cebeci, A. and Gurakan, C. (2003), Properties of potencial probiotic Lactobacillus plantarum
strains, Food Microbiology,20, 511- 518.
Chang, S. T.; Wu, J. H.; Wang, S. Y.; Kang, P. L.; Yang, N. S.; Shyur, L. F. (2001), Antioxidant
activity of extracts from Acacia confusa bark and heartwood. Journal of Agricultural and Food Chemistry,49: 3420- 3424.
Hugas, M.; Garriga, M.; Aymerich, T.; Monfort, J. M. (1993), Biochemical characterization of
Hurst, A. (1983), Nisin and other inhibitory substances from lactic acid bacteria. In: BRANEN, A. L.; DAVIDSON, P. M. Ed. Antimicrobial in
Foods. Marcel Deckker Inc., New York, USA, p. 327- 351.
Jarvis, B. (1989), Statistical aspects of the microbiological analysis of foods. Elsevier Science
Publishers, Amsterdam, The Netherlands.
Klaenhammer, T. R. (2001), Probiotics and prebiotics. In: Doyle, M. P.; Beuchat, L. R. ; Montville, T. J. Food microbiology: fundamentals
and frontiers. 2.ed. Washington: ASM, p. 797-811 .
Koutsoumanis, K. P. and Sofos, J. N. (2005), Effect of inoculum size on the combined
temperature, pH and aw limits for growth of Listeria monocytogenes. International Journal of Food Microbiology,104:83- 91.
Man, J. C.; Rogosa, M.; Sharpe, M. E. (1961). A medium for the cultivation ofLactobacilli. Journal of Applied Bacteriology,23:130- 5.
Parker, R. B. (1974), Probiotics: the other half of the antibiotics story.Animal Nutrition and Health ,
29, 4- 8.
Schillinger, U.; Lücke, F. K. (1990), Lactic acid bacteria as protective cultures in meat
products.Fleischwirtsch, 70, 1296- 1299.
Shah, N. P. (2000), Probiotic bacteria: Selective enumeration and survival in dairy foods. JournalDairy Science, v.83, p.894- 907.
Silva, N.; Junqueira, V. C. A.; Silveira, N. F. A. (2001), Manual de métodos de análise microbiológica de alimentos. 2. ed. São Paulo:
Livraria Varela, 317p.
Vandenberg, P. A. (1993), Lactic acid bacteria, their metabolic products and interference with microbial growth. FEMS Microbiology Reviews ,
12:221- 238.
lactobacilli from dry fermented
sausages.International Journal of Food
Microbiology,18:107- 113.
J. Biotec. Biodivers. v. 3, N.1: pp. 10-17, Fev. 2012