415

Journal of Food Protection, Vol. 65, No. 2, 2002, Pages 415–418

Copyright Q, International Association for Food Protection

Research Note

The In uence of Nisin on the Thermal Resistance of

Bacillus cereus
THEREZA CHRISTINA VESSONI PENNA* AND DANTE AUGUSTO MORAES

Department of Pharmaceutical Technology, School of Pharmaceutical Sciences of the University of São Paulo, Rua Antonio de Macedo Soares, 452,
São Paulo, SP, Brazil 04607-000

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/65/2/415/1674046/0362-028x-65_2_415.pdf by guest on 12 January 2023

MS 01-215: Received 15 June 2001/Accepted 10 September 2001

ABSTRACT
Decimal reduction times (D-values) at cooking and autoclaving temperatures (80 to 1208C) of spores of Bacillus cereus
ATCC 1479-8 in rice and milk (13% wt/vol) supplemented with nisin (25 mg/ml) were evaluated. The mean D-values at
97.88C in cooked white rice, phosphate buffer (pH 7.0), and rice water (pH 6.7) were 3.62, 1.99, and 1.34 min, respectively.
From 80 to 1008C, the mean reduction in D-values due to the addition of nisin to milk was 40%. The D-value at 1108C was
;0.86 min for milk (control) and milk with nisin. The z-values ranged from 7.328C (phosphate buffer) to 10.378C (milk
control).

The child patient is highly subject to diseases, and
feeding thus becomes a therapeutic factor. Cleaning and hy-
gienic controls are indispensable and should be incorporat-
ed in all of the operations carried out in the hospital’s milk
bottle preparation room (lactarium). Such controls help re-
duce the initial bioburden in food, guaranteeing the smallest
probability of pathogen survival after terminal bottle auto-
claving (15).
The main survivors are Bacillus cereus spores (5),
which are relatively common contaminants of foods, such
as dairy and cereal products (10) and dried spices (8). Rice
water and cooked rice are used in the lactarium as an al-
ternative source of carbohydrates to feed weak babies and
also to control diarrhea in infants. According to the Bra-
zilian Sanitary Food Vigilance Agency (1), the maximum
permitted population of B. cereus is 2 3 102 CFU/g in
ready-to-eat foods.
Nisin, an antimicrobial natural additive produced by
Lactococcus lactis subsp. lactis, is the only bacteriocin with
practical (14) and legalized (2) application for preserving
foodstuffs (7). Considering an initial load of 102 -activated
(808C/10 min) spores, populations greater than 105 B. ce-
reus/ml, which are considered critical for infants (5), were
attained after 12 h at 258C, 8 h at 308C, 6 h at 35 and 408C
for milk (control), and 12 h at 30 and 358C, and no viable
cells were detected at 408C for milk with nisin added (16).
In reconstituted rice cereal with milk, inoculated with 102
B. cereus/g and stored at 15 (72 h), 21 (48 h), and 308C
(12 h), enterotoxin was detected for a population .7 log10
CFU/g (11).
For comparison, thermal resistance parameters were
evaluated for spores of B. cereus in (i) phosphate buffer
* Author for correspondence. Tel: (55-11) 3818-3694; Fax: (55-11) 3815-
6386; E-mail: tcvpenna@usp.br.
(pH 7.0); (ii) rice water and rice at cooking temperatures;
and (iii) reconstituted milk (control), along with the addi-
tion of nisin, the associated in uence of which was also
studied.
MATERIALS AND METHODS
Spores of B. cereus (ATCC 14579-8) were developed (338C/
72 h) on Trypticase soy agar (TSA; Difco Laboratories, Detroit,
Mich.), harvested, centrifuged (2,000 3 g/30 min), and suspended
in chilled (i) phosphate buffer, pH 7.0 (aqueous mixture in a pro-
portion of 2:3 of sodium phosphate monobasic at 2.78% [wt/vol]
and sodium phosphate dibasic at 5.365% [wt/vol]); (ii) rice water,
pH 6.7 (Ž ltered mixture obtained from trituration to homogeni-
zation of 50 g of sterile grains of rice in 500 ml of sterile water);
and (iii) distilled and deionized (1.0 mS) water, pH 6.7 6 0.1,
stored at 48C. The viability of heat-shocked (808C/10 min) spore
suspensions (CFU ml2 1) was estimated by the pour plate tech-
nique on TSA (338C/24 h).
Three milliliters of inoculated (106 spores per ml2 1) phos-
phate buffer and rice water was dispensed into each of 32 serum
bottles (Wheaton SB205A, SP, Br), which were sealed and sub-
mitted to total heating treatments with the following come-up
times: (i) 60 and 10 min at 82 and 858C; (ii) 42 and 6 min at 88.5
and 908C; (iii) 24 and 5 min at 928C; and (iv) 24 and 0.5 min at
95 and 98.78C. Every 10 min at 82 to 858C, every 3.0 min at
928C, and every 2.0 min at 95 to 97.88C, two bottles were chilled
and immediately assayed for survivors (CFU ml2 1) by the pour
plate technique on TSA (338C/24 h). An inoculated (106 spores
per ml2 1) mixture of 698 g of sterile rice grains in 2 liters of
sterile water, transferred into a stainless steel container, was sub-
mitted to standard cooking. The temperature was controlled by a
PT 100 thermocouple (IOPE, SP, Br, calibrated to 1.0 6 0.18C)
inserted in the center of the mixture. The come-up time of 20 min
to reach the boiling point (97.88C) was equivalent to the estimated
cooking time (F value) of 10.45 min (z 5 7.918C), from the be-
ginning of heating up to the boiling point. At 97.88C, the timer
was turned on, the cooking time was recorded for 22 min, and
416 PENNA AND MORAES J. Food Prot., Vol. 65, No. 2

TABLE 1. Thermal destruction curves [Log N 5 Log N0 2 slope·t (min)] and decimal reduction times (D-values) at heating temper-
atures for spores of B. cereus into milk (group 1, control); milk with nisin added (group 2); phosphate buffer at pH 7.0; rice water
(pH 6.7); and cooking ricea
Temperature D-Value
Heating menstrum (8C) Slope (2b) SE (b) R2 (min) V(D) SE(D)

Cooking rice 97.8 20.2905 0.0205 0.9569 3.44 5.90 3 102 2 2.43 3 102 1
97.8 20.2930 0.0299 0.9144 3.41 1.21 3 102 1 3.48 3 102 1
97.8 20.2700 0.0205 0.9715 3.70 7.91 3 102 2 2.81 3 102 1
97.8 20.2859 0.0280 0.9575 3.50 1.17 3 102 1 3.43 3 102 1
97.8 20.2909 0.0299 0.9653 3.44 1.25 3 102 1 3.53 3 102 1
97.8 20.2639 0.0301 0.9561 3.79 1.87 3 102 1 4.32 3 102 1
Phosphate buffer 82 20.0030 0.0002 0.9826 333.33 6.53 3 102 2.56 3 101
pH 5 7.0 85 20.0093 0.0062 1.0000 107.53 5.16 3 103 7.18 3 101

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/65/2/415/1674046/0362-028x-65_2_415.pdf by guest on 12 January 2023

88.5 20.0456 0.0050 0.8780 21.93 5.85 3 100 2.42 3 100
90 20.0709 0.0118 0.9371 14.10 5.51 3 100 2.35 3 100
92 20.1265 0.0145 0.9493 7.91 8.24 3 102 1 9.08 3 102 1
95 20.1842 0.0196 0.9365 5.43 3.33 3 102 1 5.77 3 102 1
97.8 20.5027 0.0419 0.9665 1.99 2.75 3 102 2 1.66 3 102 1
Rice water 82 20.0087 0.0000 0.9812 114.55 1.75 3 102 2 1.32 3 102 1
pH 5 6.7 85 20.0110 0.0008 0.9856 90.91 3.84 3 101 6.20 3 100
90 20.0963 0.0030 0.9954 10.38 1.05 3 102 1 3.23 3 102 1
92 20.1544 0.0147 0.9562 6.48 3.81 3 102 1 6.17 3 102 1
95 20.1764 0.0164 0.9508 5.69 2.77 3 102 1 5.27 3 102 1
97.8 20.7471 0.0643 0.9575 1.34 1.33 3 102 2 1.15 3 102 1
Milk (control) 80 20.0377 0.0030 0.9756 26.53 4.39 3 100 2.20 3 100
90 20.1003 0.0109 0.9546 9.97 1.18 3 100 1.09 3 100
97.8 20.5559 0.0711 0.9386 1.80 5.29 3 102 2 2.30 3 102 1
100 21.0969 0.0722 0.9585 0.91 3.60 3 102 3 6.01 3 102 2
110 21.1665 0.0226 0.9989 0.86 2.76 3 102 4 1.66 3 102 2
120 22.4026 0.2325 0.9816 0.24 1.62 3 102 3 4.03 3 102 2
Milk with nisin 80 20.0628 0.0072 0.9496 15.92 3.37 3 100 1.84 3 100
90 20.1201 0.0121 0.9611 8.33 7.01 3 102 1 8.37 3 102 1
97.8 21.0303 0.1889 0.9706 0.97 3.17 3 102 2 1.78 3 102 1
100 20.9153 0.0579 0.9804 1.09 4.78 3 102 3 6.91 3 102 2
110 21.1381 0.0597 0.9891 0.88 2.12 3 102 3 4.61 3 102 2
120 22.3855 0.3286 0.9461 0.42 3.33 3 102 3 5.77 3 102 2
a Log10 N, spore population; SE, standard error (P , 0.05); R2, multiple determination coefŽ cient (P , 0.05); (2b), slope (P , 0.05);
D-value, (21/b), decimal reduction time (min); V(D), variance, [(1/b2)2 3 [SE(b)]2]; SE(D), V(D)1/2; t, time of exposition (min).

the come-up time was 2 to 4 min. Under constant homogenization,
10-g samples (pH 5 5.8 6 0.1; aw 5 0.956 to 0.968) were taken
at regular 2.0-min intervals and immediately chilled, and the spore
survivors (CFU g2 1) were estimated by the pour plate technique
on TSA (338C/24 h).
A sterile stock solution of 250 mg nisin per ml2 1 (104 IU
ml2 1) was prepared with Nisaplin (Aplin & Barrett Ltd., Sigma
Chemical Co., St. Louis, Mo., at an activity of 106 IU g2 1) in
0.02 N HCl solution, to which 0.75% (wt/vol) NaCl solution was
added (13), the Ž nal pH being adjusted to 3.0 (6).
Three milliliters of inoculated (106 spores per ml2 1) recon-
stituted milk (13% wt/vol, pH 6.71 6 0.05) was dispensed into
each of twelve 5-ml glass serum bottles (Wheaton SB205A, Br),
forming group 1 (control). Then, 2.7 ml of inoculated milk was
dispensed into each of 12 serum bottles plus 0.3 ml of the stock
solution of nisin, forming group 2, which contained 25 mg nisin
per ml (103 IU ml2 1), with a Ž nal pH value of 6.62 6 0.02. The
24 serum bottles were sealed and heat treated with the following
come-up times: (i) 60 and 10 min at 808C; (ii) 30 and 5 min at
908C; and (iii) 6 and 1 to 1.5 min at 97.8, 110, and 1208C. Every
10 min at 808C, every 5.0 min at 908C, and every 0.5 min at 97.8,
110, and 1208C, two bottles were chilled and immediately assayed
for survivors (CFU ml2 1) by the pour plate technique on TSA
(338C/24 h). Treatments up to 97.88C were carried out in a water
bath, and those above 1008C were carried out in a silicone (200/
220 CS, r 5 0.948 g/cm3) oil (Dow Corning, SP, Br) bath. The
assays were repeated at least three times for each heating men-
struum and temperature.
The decimal reduction times (D-values) were determined
from the negative reciprocal of the slopes (b) of the regression
lines, using the linear portions of the survivor curves (log10 pop-
ulation versus time of exhibition at a constant temperature). The
slope standard error [SE(b)] and the multiple determination co-
efŽ cient (R2) were also calculated. The estimated value of the D-
value standard error [SE(D)] was obtained from the root of the
variance [V(D)] , which was calculated by V(D) 5 {(1/b2)2 3
[SE(b)]2} (Table 1). The z-value was calculated by the negative
reciprocal of the Thermal Death Time curve slope, obtained by
means of regression analysis applied to the decimal logarithm of
the D-values and respective heating temperatures. Simple regres-
sion variance analysis, estimated parameters, and respective con-
Ž dence intervals at signiŽ cant levels (P , 0.05) were calculated
J. Food Prot., Vol. 65, No. 2 INFLUENCE OF NISIN ON THE THERMAL RESISTANCE OF B. CEREUS
FIGURE 1. Survivor curves of Bacillus cereus spores in rice cook-
ing for 22 min at 97.88C, taking into consideration both shoulders
and tailing in semilogarithmic linear declines in the log CFU/g
with time for (1 ) maximum, (O) mean, and (2) minimum popu-
lations of survivors.
through the SGWIN program (Statgraphics Plus for Windows,
version 1.4, Statistical Graphics Corporation, Englewood Cliffs,
N.J.).
RESULTS AND DISCUSSION
Spores of B. cereus ATCC 14579-8 were selected as
biological reference indicators to evaluate the efŽ ciency of
(i) cooked rice, (ii) heat treatment of phosphate buffer at
pH 7.0, and (iii) rice water with the synergetic in uence of
nisin added to reconstituted milk.
Rice cooked for 22 min at 97.88C was able to reduce
the spore populations to less than 102 CFU g2 1 (Fig. 1).
The survivor curves of B. cereus spores in rice cooked for
22 min at 97.88C were represented, taking into considera-
tion both shoulders and tailing in semilogarithmic linear
declines, with time in the maximum, mean, and minimum
populations of survivors.
During the Ž rst 4 min of cooking, no reduction in the
population occurred because of the thermal activation of
the spores. The decimal reduction time (D-value) during the
last 18 min of rice cooking (decreasing portion of the sur-
vivor curve) ranged from 3.41 to 3.79 min. The smooth
tailing deviations of the survivor curves were considered in
the calculation of D-values.
For an initial bioburden of 6 log10, a total interval of
31 min (including the Ž rst 4 min of thermal activation)
would provide a reduction of at least 7 log10, for a Ž nal
population #102 1 CFU g2 1 , guaranteeing reasonable safety
for consumers. Since cooked rice forms part of the daily
diet of Brazilians, the selection of rice grains and ingredi-
ents and their hygienic manipulation are important factors
in the safety of the cooked product.
The lower pH of rice water (pH 6.7) compared with
that of phosphate buffer (pH 7.0) coincided with a D-value
signiŽ cantly (P , 0.05; three times lower) at 828C and a
difference between D-values of 16.62 min at 858C, respec-
tively. The effect of pH was not felt at higher temperatures,
when the D-values for buffer and rice water approached
417
each other. The spores showed greater thermal resistance in
the phosphate buffer, the z-values being 7.32 and 7.918C
for phosphate buffer and rice water, respectively. At 97.88C,
the spores in the rice-cooking procedure presented almost
three times greater thermal resistance.
Gaillard et al. (9) observed a signiŽ cant variation in
the D-values (2.0 to 5.4 min) of 25 different strains of B.
cereus, the spores of which were treated in citrate-phos-
phate buffer with pH 7.0 at 1008C (average z-value of
9.28C). The combined variation of pH (4.5 to 6.5), water
activity (aw 5 0.80 to 1.0), and temperature (85 to 1058C)
on the D-values for B. cereus in citrate-phosphate buffer is
remarkably relevant to our work. In the present study, the
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/65/2/415/1674046/0362-028x-65_2_415.pdf by guest on 12 January 2023
D-value at the boiling water point ranged from 3.55 6 0.17
min (cooked rice) to 1.5 6 0.58 min (milk) due to the
variation in composition of the heating menstruum.
The average D-values obtained for each menstruum at
different temperatures were signiŽ cantly different (P ,
0.05). The presence of nisin (25 mg ml2 1 , group 2) in milk
lowered the mean D-values of B. cereus, compared with
milk control (group 1), to ca. 40% at 808C, 16% at 908C,
and 46% at 98.78C, respectively. At 100 and 1108C, the D-
values were signiŽ cantly similar (P , 0.05), and at 1208C,
the D-value of 0.42 min was equal for both groups, the z-
values being 10.378C (group 1) and 9.118C (group 2). The
range of D808C -values (15.92 to 26.53 min) is an indication
of the thermal resistance of the activated spores in milk,
equivalent to the Ž rst 4 min of rice cooking at 97.88C. Ther-
mal activation of the spores of B. cereus at 808C generally
occurs after 10 to 20 min of exposure (3).
Wandling et al. (17) and Beard et al. (4) observed that
the addition of 2,000 IU nisin per ml of milk preparations
lowered the thermal resistance of Bacillus spores. For
spores of B. cereus in skim milk (control) and in skim milk
with nisin added, Wandling et al. (17) obtained mean D-
values of 7.0 and 4.8 min at 978C, 2.7 and 2.2 min at
1008C, and 1.5 and 0.85 min at 1038C, respectively. Be-
cause of the addition of nisin, the corresponding reductions
in D-values in relation to the controls were 32% (978C),
20% (1008C), and 42% (1038C). For spores of Bacillus
stearothermophilus in chocolate milk (with 10 to 12% fat
cocoa powder) supplemented with nisin, Beard et al. (4)
veriŽ ed that the increasing effect of nisin with a decreasing
pH caused reductions of D1308C of 19.30, 30, and 38.2% at
pH 6.4, 5.7, and 5.0, respectively.
In the hospital lactarium, autoclaving (1108C/10 min),
followed by cooling and storing of the bottles, determines
the microbiological quality of the milk preparation. For a
bioburden of 6 log10 with a D-value of 0.87 6 0.01 min,
the equivalent time, F, of 10 min at 1108C would provide
a reduction of at least 10 logarithmic cycles of spores (Ž nal
population, #102 4 CFU ml2 1 ), ensuring the level of safety
for every lot of 1,000 bottles being autoclaved. Nisin,
which alters the membrane permeability by dissipating pro-
ton motive force in gram-positive cells (14), was shown to
prolong the lag phase of microbial growth (12) and also to
reduce the thermal resistance of Bacillus spores and vege-
tative cells (4, 7). The addition of nisin to milk before it is
blended into rice water or cooked white rice prevents re-
418 PENNA AND MORAES
covery of survivor pathogens (17). Nisin addition also pre-
vents pathogen recovery during the cooling of autoclaved
bottles (1108C/10 min) and during the time that the product
remains on hospital pediatric  oors (16). Nisin, which is
degraded by proteases, may be incorporated in the diet of
infants without interfering with the native gastrointestinal
 ora.
ACKNOWLEDGMENTS
The authors thank the Brazilian Committees for ScientiŽ c Technol-
ogy Research (CNPq and FAPESP) for Ž nancial support; biologists Irene
A. Machoshvili and Dalete Nogueira Fajardo for technical support; and
Norah Duncan for the English revision of the manuscript.
REFERENCES
1. Anonymous. 1987. Brazilian Sanitary Food Vigilance Agency (DIN-
AL) approves microbiological standards for foodstuffs, p. 2197–
2200. In Federal laws and decrees (Diário OŽ cial da União), section
1. Government Printing OfŽ ce, Brasõ´lia, Brazil.
2. Association of Brazilian Industries of Foodstuffs (ABIA). 1996.
Compendium of legislation on foodstuffs, revision 6, vol. 1. ABIA,
São Paulo.
3. Baston, L. M. L. 1997. Spores of Bacillus cereus: biological indi-
cators in the hygienization of nursing bottles in lactarium. Master’s
thesis. University of São Paulo, São Paulo, Brazil.
4. Beard, B. M., B. W. Sheldon, and P. M. Foegeding. 1999. Thermal
resistance of bacterial spores in milk-based beverages supplemented
with nisin. J. Food Prot. 62:484–491.
5. Becker, H., G. Schaller, W. vonWiese, and G. Terplan. 1994. Bacillus
cereus in infant foods and dried milk products. Int. J. Food Micro-
biol. 23:1–45.
J. Food Prot., Vol. 65, No. 2
6. Davies, E. A., H. E. Bevis, R. Potter, J. Harris, G. C. Williams, and
J. Delves-Broughton. 1998. The effect of pH on the stability of nisin
solution during autoclaving. Lett. Appl. Microbiol. 27:186–187.
7. Desmazeaud, M. 1997. Bacteriocins of lactic acid bacteria (LAB)
and their interest to improve the hygienic quality of products. Cerela
8:38–43.
8. Drobniewski, F. A. 1993. Bacillus cereus and related species. Clin.
Microbiol. Rev. 6:324–338.
9. Gaillard, S., I. Leguerinel, and P. Mafart. 1998. Model for combined
effects of temperature, pH and water activity on thermal inactivation
of Bacillus cereus spores. J. Food Sci. 63:887–889.
10. Granum, P. E. 1994. Bacillus cereus and its toxins. J. Appl. Bacteriol.
76:61S–66S.
11. Jaquette, C. B., and L. R. Beuchat. 1998. Survival and growth of
psychrotrophic Bacillus cereus in dry reconstituted infant rice cereal.
J. Food Prot. 61:1629–1635.
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/65/2/415/1674046/0362-028x-65_2_415.pdf by guest on 12 January 2023
12. Naclerio, G., E. Ricca, M. Sacco, and M. Felipe. 1993. Antimicrobial
activity of a newly identiŽ ed bacteriocin of Bacillus cereus. Appl.
Environ. Microbiol. 59:4313–4316.
13. Rogers, A. M., and T. J. Montville. 1991. Improved agar diffusion
assay for nisin quantiŽ cation. Food Biotechnol. 5:161–168.
14. Verheul, A., N. J. Russel, R. V. Hof, F. M. Rombouts, and T. Abee.
1997. ModiŽ cations of membrane phospholipid composition in nis-
in-resistant Listeria monocytogenes Scott A. Appl. Environ. Micro-
biol. 63:3451–3457.
15. Vessoni Penna, T. C., I. A. Machosvili, and L. M. Baston. 1994. The
importance of autoclaving in hospital lactarium. Laes Haes 16:68–
74.
16. Vessoni Penna, T. C., I. A. Machosvili, and D. N. Fajardo. Effect of
nisin on growth kinetics of activated Bacillus cereus spores in
cooked rice and in milk. Submitted for publication.
17. Wandling, L. R., B. W. Sheldon, and P. M. Foegeding. 1999. Nisin
in milk sensitizes Bacillus spores to heat and prevents recovery of
survivors. J. Food Prot. 62:492–498.