Full text of DOI:10.1111/j.1365-2621.2003.tb08277.x

JFS: Food Microbiology and Safety
A Descriptive Model for the Kinetics of
Gas Release in Different Types of Pizzas
M.L. CABO, L. PASTORIZA, AND J.J. RODRÍGUEZ
ABSTRACT: A model based on typical equations of microbial kinetics is proposed to describe gas release (GR) in
ham, tuna, and meat pizzas packaged under different CO2-enriched atmospheres: 20% CO2, 70% CO2, and 70% CO2
plus 500 mg/kg Nisaplin. CO2-enriched atmospheres hardly influenced LAB growth but reduced GR, which points
to the importance of yeasts for GR. Nonetheless, LAB also contributed significantly to GR, so Nisaplin also delayed
GR. Gas release followed a diauxic pattern, the 2nd stage being presumably related to yeasts shifting from respira-
tory to fermentative metabolism once oxygen was depleted. However, storing pizzas for longer than 25 to 30 d does
not seem appropriate in terms of shelf life, so the model proposed appears adequate for practical purposes.
Keywords: pizza, Nisaplin, CO2, modeling, antibacterial activity
g) made up of 265 g of dough (pre-cooked at 300 °C for 2 min), 65 g
Introduction
of tomato paste, cheese substitute, and a number of different in-
gredients (ham pizza: ham, mushrooms, and olives; tuna pizza:
tuna, onions, and olives; meat pizza: meat and bacon; cheese pizza:
Emmental, mozzarella and cheddar cheeses; pepperoni pizza: pep-
peroni).
The independent variables of the study were the content of
Nisaplin (500 mg/kg or absence) and the initial proportion of CO2
in the gas mixture (20 or 70%). These values were selected in accor-
dance with a previous study (Cabo and others 2001).
Nisaplin (106 IU nisin per g) solutions were prepared in sterile
distilled water immediately prior to addition, and added by spray-
ing on top of the pizza and by mixing with the tomato paste. A same
volume of water, that is no Nisaplin, was added to a number of piz-
zas which were used as controls. The content of Nisaplin per pizza
always refers to the total added amount.
Once all the ingredients had been added, pizzas were placed
into a thermo-moulded laminate, and subsequently the gas mix-
ture (Carburos Metálicos S.A., Galicia, Spain) was injected and im-
mediately another laminate was sealed on top by using a vacuum-
compensated heat sealer (TF-1000, ULMA, Galicia, Spain). Gas
mixtures contained N2 as filler, and the system of packaging en-
abled some residual oxygen (~1%) to be left in the headspace with
the aim of preventing the growth of anaerobic pathogens (Church
and Parsons 1995).
WELLING OF PACKS DUE TO GAS RELEASE IS A MAJOR PROBLEM IN MANY
S
foods. Heterofermentative lactic acid bacteria, yeasts, and lac-
tose positive coliforms have been identified as responsible for CO2
production in most cases (salads: Birzele and others 1997; cooked
meat products: Sameshima and others 1997; Samelis and others
2000; cheese: Westall and Filtenborg 1998; fish products: Lyhs and
others 2001).
This was also the case in modified atmosphere packaged pre-
cooked ham pizzas, but the combination of packaging under a CO2-
enriched atmosphere and adding Nisaplin (106 IU nisin per g, Applin
& Barrett Ltd., Dorset, U.K.) prevented swelling and increased shelf
life significantly (Cabo and others 2001). This combination was found
to act synergistically presumably because of a complementary ef-
fect, since CO2 inhibits the growth of yeasts and nisin inhibits LAB.
These results were of a great interest for the food industry, so it was
considered that it would be useful to find out if they were confirmed
in other kinds of pizzas of the same line of production. It would allow
that only one mathematical model could be used to describe the
effects of nisin and CO2 on the kinetics of gas release. This could be
very helpful for the food industry as a first step to shelf-life predic-
tion and control of critical points of the production line.
Foods usually spoil as a result of microbial activity, which causes
severe changes in chemical and sensory properties. Modeling is
often used to describe the effects of state variables (Devlieghere
and others 1998; Koutsoumanis and others 2000; Erkmen 2000;
Castillejo Rodríguez and others 2000) on the kinetics of microbial
growth. Considering the complex microbial ecology of food sys-
tems, monitoring some of those changes would be clearly advanta-
geous in highly heterogeneous foods, such as pizzas, where several
different microorganisms contribute to spoilage.
Packed-pizzas were stored at 7 ꢀ 1 °C. This temperature was cho-
sen to simulate real conditions during distribution and storage at
retail outlets. Triplicate samples of each treatment were subjected
to a number of analyses at different periods of storage. Unless spec-
ified, sampling was interrupted whenever pizzas were rejected by
sensory panels.
This work has therefore aimed the development of a model to
describe the kinetics of gas release in various kinds of pizzas pack-
aged under different gas mixtures in the presence and absence of
Nisaplin.
Microbiological and chemical measurements
Pizzas were randomly cut into small pieces (3 to 4 g). A homoge-
neous mixture was obtained by transferring several of these pieces
(about 25 g) into a sterile plastic Stomacher bag and blending with
sterile peptone water (Panreac S.A., Spain) in the ratio 1:9 (w/v) by
means of a Seward Stomacher lab-blender (Model 400, Seward
Medical, London, U.K.). Subsequently, homogenates were serial di-
luted in peptone water, and 0.1 mL aliquots were spread on Petri
Materials and Methods
Experimental design
Each experimental unit consisted of one single pizza (about 400
996 JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 3, 2003
© 2003 Institute of Food Technologists
Further reproduction prohibited without permission

Kinetics of gas release in pizzas . . .
plates by triplicate. LAB counts were carried out on MRS agar (Pan-
reac S.A.) after incubation at 37 °C for 2 d. Yeasts were enumerated
on Malt Extract Agar (Panreac S.A) by incubation at 25 °C during 5
d.
r
rꢁ=
(6)
1+aꢁr’ × [N]+br’ × [C]
The proportions of CO2 and O2 in the headspace were deter- Aꢁ: maximum amount of gas released in the presence of Nisaplin or
mined by using a PAK 12P headspace analyser (Abiss S.A.R.L.,
France). The amount of gas released (GR) is defined as the increase
(at each sampling time) in the proportion of CO2 in the headspace
respect to that injected. In a preliminary experiment swelling was
CO2; rꢁ: specific gas release (dimensions T-1) in the presence of
Nisaplin or CO2; A: maximum amount of gas released in the ab-
sence of Nisaplin and CO2, expressed as percentage respect to the
total amount of CO2 present; and r: specific gas release in the ab-
determined only semi-quantitatively by using two gauges (22 and sence of Nisaplin and CO2 (dimensions T – 1).
28 mm high), so packs were classified in 3 different categories: low-
er than 22 mm, between 22 and 28 mm, and higher than 28 mm.
Aꢁ
GR0
zꢁ=
–1
Mathematical modeling
Bearing in mind that the inhibition of microbial growth by any
where GR0 is the amount of gas released at zero time; aA’ and bA’:
factor normally results in a decrease of the asymptote or the specific empirical parameters defining the degree of influence of Nisaplin
growth rate, the kinetics of growth of LAB (L) were defined by using and CO2, respectively, on the maximum amount of gas released; ar’
a classic logistic equation:
and br’: empirical parameters defining the degree of influence of
Nisaplin and CO2, respectively, on the specific gas release; and GR0
is conceptually zero, but Eq. 4 fails to make such an estimate and
provides values higher than zero for the origin ordinate. Conse-
quently, this estimate must be substracted from Eq. 4 (Cabo and
kꢁ
1+cꢁ× e–ꢃꢁ × t
L =
(1)
where the parameters µꢁ and kꢁ were defined as a function of the others 1999) as follows:
concentration of Nisaplin and the initial proportion of CO2 in the
gas mixture as follows:
1
1
GR= Aꢁ
– (
)
(7)
1 + zꢁ × e–r’ × t
zꢁ + 1
k
kꢁ=
(2)
1+ ak’ × [N] + bk’ × [C]
To substract the origin ordinate provides an estimate for A’ that
does not correspond to the actual value for the asymptote. The lat-
ter, however, can be determined by calculation of the limit when
time becomes infinite (Murado and others 2002):
ꢃ
ꢃꢁ =
(3)
1+ aꢃ’ × [N] + bꢃ’ × [C]
k: maximum biomass of LAB in the absence of Nisaplin and CO2,
expressed as log CFU per g of sample; µ: specific growth rate of LAB
in the absence of Nisaplin and CO2 (biomass formed per unit of
present biomass and per unit of time, dimensions time–1); kꢁ: max-
imum biomass of LAB in the presence of Nisaplin or CO2; ꢃꢁ: specif-
ic growth rate of LAB in the presence of Nisaplin or CO2.
ꢁ
z
Aꢁ max = lim Aꢁ = Aꢁ × (———)
(8)
t→ꢂ
zꢁ + 1
Fits of Eq. 1 and 5 to experimental data were performed according
to a least-squares method (quasi-Newton). All calculations were
carried out by using a Microsoft Excel program.
k
cꢁ =
–1
L0
Results and Discussion
N A PRELIMINARY EXPERIMENT, MEASUREMENTS OF SWELLING AFTER 21
I
d of storage in 100 samples of each ham, tuna, and meat pizzas
where L0 is the initial biomass of LAB; N: concentration of Nisaplin,
expressed as mg/kg of pizza; C: initial proportion (%) of CO2 in the
gas mixture; akꢁ and bkꢁ: empirical parameters defining the degree
of influence of Nisaplin and CO2, respectively, on the maximum
biomass of LAB; and aµꢁ and bµꢁ: empirical parameters defining the
degree of influence of Nisaplin and CO2, respectively, on the spe-
cific growth rate.
packed under 20 or 70% CO2 in the presence of Nisaplin (500 mg/
kg) showed similar response patterns (Figure 1). That is, most 20%
CO2 packed-pizzas had swollen above 22 mm and many even
above 28 mm, while swelling had hardly occurred in pizzas pack-
aged under 70% CO2 in the presence of Nisaplin. In contrast, pat-
terns were dissimilar for cheese or pepperoni pizzas since swelling
was lower for both types of treatments.
Considering the microbial origin of the gas released (Cabo and
others 2001), GR was also defined by a similar logistic-type equation:
Accordingly, this work was focused on the development of a
model describing the combined effects of Nisaplin and packaging
under CO2-enriched atmospheres on the kinetics of gas release in
ham, tuna, and meat pizzas, the shelf-lives of which were signifi-
cantly shorter according to previous studies (results not shown).
With this aim, a comparative study was made for each of these piz-
zas among 3 different treatments: packaging under 20% CO2 (that
is, commercial treatment), packaging under 70% CO2, and packag-
ing under 70% CO2 plus addition of 500 mg Nisaplin per kg of piz-
za.
Aꢁ
(4)
GR =
1+zꢁ × e–rꢁ × t
in which the parameters Aꢁ and rꢁ were also defined as a function of
the concentration of Nisaplin and the initial proportion of CO2 in
the gas mixture as follows:
A
Though a high number of LAB (> 107 CFU/ml) is rapidly reached
Aꢁ=
(5)
1+aA’ × [N]+bA’ × [C]
in modified atmosphere packaged-pizzas, it has a reduced signif-
Vol. 68, Nr. 3, 2003—JOURNAL OF FOOD SCIENCE 997

Kinetics of gas release in pizzas . . .
Table 1—Values for the parameters of the Eq. 1 of LAB growth in ham, tuna, and meat pizzas packaged under 20%
CO2 (A), 70% CO2 (B), and 70% CO2 plus 500 mg/kg of Nisaplin (C)
Ham pizza
B
Tuna pizza
B
Meat pizza
B
A
C
A
C
A
C
kꢁ
µꢁ
cꢁ
9.34
0.190
2.05
8.87
0.232
1.89
8.56
0.147
1.79
9.00
0.911
1.30
8.76
0.989
1.24
8.58
0.187
1.19
9.05
0.775
1.06
8.60
0.911
0.96
8.18
0.221
0.86
r2
n*
0.994
22
0.994
14
0.967
14
*number of experimental data
icance in terms of food safety and is not immediate responsible for
the development of off-flavors and off-odors (Cabo and others
2001). In fact, the loss of quality results from the activity of very
many different LAB and yeasts, which cause acidification and the
release of CO2 (Cabo and others 2001) so swelling becomes the
major limiting factor for the shelf life of MAP-pizzas.
tween experimental data (either of LAB counts or amount of gas re-
leased) and estimates from each equation were achieved for all piz-
zas (Table 1 and 2). Additionally, residual plots did not show any
suspicious regularity in any case either (Figure 2 and 3). Both sets of
results proved the goodness of fit of the 2 models.
Comparing the values for the parameters of these models is
useful to assess how much Nisaplin and CO2 reduced the spoilage
of those pizzas. For instance, the growth rates (ꢃꢁ) of LAB decreased
The growth of LAB and GR were successfully described by Eq. 1
and 5, respectively. High values for the correlation coefficients be-
Figure 1—Swelling of ham, tuna, meat, cheese and pepperoni pizzas packaged under 20% CO2 (filled bars) or 70%
CO2 in the presence of 500 mg/kg of Nisaplin (unfilled bars) after 21 d of storage.
998 JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 3, 2003

Kinetics of gas release in pizzas . . .
Table 2—Values for the parameters of the Eq. 5 of gas release in ham, tuna, and meat pizzas packaged under 20%
CO2 (A), 70% CO2 (B), and 70% CO2 plus 500 mg/kg of Nisaplin (C)
Ham pizza
B
Tuna pizza
B
Meat pizza
B
A
C
A
C
A
C
Aꢁ
rꢁ
zꢁ
46.42
0.450
37.13
28.55
0.194
22.45
20.46
0.120
15.81
42.53
0.822
325.0
20.37
0.690
155.15
16.99
0.291
129.24
44.24
0.683
138.06
20.56
0.550
63.62
20.31
0.220
62.84
r2
n*
0.983
29
0.996
20
0.991
21
*number of experimental data
significantly as a result of the addition of Nisaplin (Table 1), and this and others 1990; Borch and others 1996), the present case followed
is presumably related to the decreases in the rates of gas release (rꢁ) the general trend (Lannelongue and others 1982; Layrisse and
when Nisaplin was added (Table 2).
Matches 1984), and so the value of ꢃꢁ was even slightly higher for
The most relevant finding, however, results from the fact that pizzas stored under 70% CO2 than for those subjected to commercial
significant decreases in Aꢁ and rꢁ when pizzas were packed under treatment. On the contrary, yeast counts were lower in pizzas stored
70% CO2 (compared with 20% CO2) did not correspond with similar under 70% CO2 than in those under commercial treatment (Table
decreases in kꢁ and ꢃꢁ. That is, GR did not follow the kinetics of LAB 3). The inhibition of yeasts by high CO2 concentrations is well-
growth (Figures 4 and 5). In fact, although a retarding effect of CO2
on the growth of LAB has been noted in a few cases (Ahvenainen
Figure 2—Residual plots for the Eq. 1 of LAB growth in ham
(a), tuna (b), and meat (c) pizzas
Figure 3—Residual plots for the Eq. 5 of the amount of gas
released (GR) in ham (a), tuna (b), and meat (c) pizzas
Vol. 68, Nr. 3, 2003—JOURNAL OF FOOD SCIENCE 999

Kinetics of gas release in pizzas . . .
Table 3—Yeast counts (log CFU/g) in ham, tuna, and meat pizzas packaged under 20% CO2 (A), 70% CO2 (B), and 70%
CO2 plus 500 mg/kg of Nisaplin (C)
Ham pizza
B
Tuna pizza
B
Meat pizza
B
Days
A
C
A
C
A
C
3
8
13
17
2.30
2.92
3.95
3.94
<1
<1
<1
<1
<1
<1
<1
<1
5.00
4.50
4.95
4.96
3.20
3.47
3.14
3.15
3.04
3.55
3.20
3.21
4.38
5.00
5.80
5.81
3.10
3.57
3.70
3.60
3.23
4.00
3.53
3.62
known (Jones and Greenfield 1982; Kuriyama and others 1993). Nisaplin reduced ꢃꢁ (of LAB) notably, and this is presumably related
Therefore, the dissimilarity between the kinetics of GR and LAB to the lower rꢁ for pizzas containing Nisaplin.
growth seems to make clear the important role of yeasts in the re-
lease of CO2.
Nevertheless, the spoilage ability of LAB and yeasts does not
only depend on the rate of growth, but also on the specific metabolic
Nonetheless, LAB also contributed significantly to gas release. activity of the different species/groups growing on or in the prod-
This is clear when comparing the kinetics of LAB growth and GR for uct (Samelis and others 2000). Accordingly, to estimate the actual
pizzas stored under 70% CO2 with or without Nisaplin. Nisin inhib- contribution of yeasts and LAB, the activity of heterofermentative
its the growth of LAB but has no effect on yeasts. The addition of LAB, homofermentative LAB and yeasts would have to be deter-
Figure 4—Growth of LAB in ham, tuna and meat pizzas stored under 20% CO2 (broken line, ᭹), 70% CO2 (unbroken
.
black line, ᭺) and 70% CO in the presence of 500 mg/kg of Nisaplin (grey line, ). Lines represent estimates from fits
᭺
2
of Eq. 1 to experimental data (symbols).
Figure 5—Gas release in ham, tuna, and meat pizzas stored under 20% CO2, 70% CO2, and 70% CO2 in the presence
of 500 mg/kg of Nisaplin. Estimates and experimental data follow the notations of Figure 4. Lines represent estimates
from fits of Eq. 5 to experimental data (symbols).
1000 JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 3, 2003

Kinetics of gas release in pizzas . . .
mined. Considering the highly complex microbial ecology of pizzas,
1. Gas release would have resulted mostly from an increased
monitoring GR is a clear advantage to assess the effects of CO2 and fermentative activity of yeasts.
Nisaplin on the system.
2. LAB would still contribute too despite having reached the sta-
Another feature of particular relevance shown in Figure 5 was that tionary phase due to a secondarization of the production of CO2
the amount of gas released started to increase markedly in some (Luedeking and Piret 1959).
cases after 24 to 30 d of storage. The type of pizza, the presence or
However, some of the previous results would seem to support the
absence of Nisaplin and the proportion of CO2 in the gas mixture former hypothesis. For instance, the release of gas during the sec-
determined when this step started to occur. This result seems to in- ond stage was less noticeable in ham pizza, which showed the lowest
dicate that some effect has been missed by Eq. 5, and therefore the yeast counts (Table 3), than in tuna or meat pizzas. Or it was most
model is only valid prior to that increase. Consequently, subsequent marked when pizzas were subjected to the commercial treatment,
experimental data were excluded from fits of Eq. 5.
that is, with the lowest CO2 concentration, since yeasts were least in-
To verify whether or not GR actually stepped up markedly dur- hibited.
ing the last days of storage, ham pizzas were processed as usual
It is also clear from Figure 6 that the proportion of oxygen fol-
(see Materials and Methods) and packaged under 50% CO2 and lowed an inverse relationship with GR. The depletion of oxygen
10% O2 in the presence of Nisaplin (500 mg/kg). Additionally, they below a critical concentration might have triggered a shift in the
were stored for 50 d so that the whole profile of GR was defined. It metabolism of yeasts from respiratory to fermentative, and it would
is to be expected that the addition of Nisaplin would inhibit the account for the high rates of gas release in the second stage.
growth of LAB and as a result favor the activity of yeast due to a
Although it seems evident that GR followed a diauxic profile,
lesser degree of competition. The concentration of oxygen seemed storing pizzas for longer than 25 to 30 d does not seem appropriate
to be critical (data not shown), since in pizzas containing a low ini- in terms of shelf life. Consequently, the use of Eq. 5 appears to be
tial proportion of oxygen (1%) GR started to increase markedly once adequate for practical purposes.
oxygen had decreased to very low values (< 0.5%) (Cabo and others
Conclusions
2001). It could be expected that a higher O2 concentration could
show this point better and might help to clarify the role of LAB and
yeast in gas release.
As clearly shown in Figure 6, GR followed a diauxic pattern. Both
LAB and yeasts would have contributed to the release of CO2 dur-
ing the first stage (see above). To account for the release of gas
during the second stage, 2 alternatives can be considered:
OST MODELS HAVE BEEN AIMED AT DESCRIBING THE KINETICS OF
M
growth of spoilage or pathogenic microorganisms in re-
sponse to different factors (spoilage microflora: Mayer-Miebach
and others 1997, Koutsoumanis and others 2000; Listeria: Castillejo
and others 2000, Erkmen 2000). Modeling changes in chemical or
sensory properties of complex food systems due to microbial activ-
ity is much less common. However, it may help to simplify the study
of heterogenous food systems, and from an industrial viewpoint,
even to control the critical points of production lines.
References
Ahvenainen R, Skyttä E, Kivikataja R-L. 1990. The influence of modified atmo-
sphere packaging on the quality of selected ready-to-eat foods. Lebensm Wiss
Technol 23:139-48.
Birzele B, Orth R, Kraemer J, Becker B. 1997. Influence of psychrotrophic yeasts
on the spoilage of delicatessen salads. Fleischwirtschaft 77:331-3.
Borch E, Kant-Muermans ML, Blixt Y. 1996.Bacterial spoilage of meat and cured
meat products. Intl J Food Microbiol 33:103-20.
Cabo ML, Murado MA, González MP, Pastoriza L. 1999. A method for bacteriocin
quantification. J Appl Microbiol 87:907-14.
Cabo ML, Pastoriza L, Bernárdez M, Herrera JJR. 2001. Effectiveness of CO2 and
nisin to increase shelf life of fresh pizza. Food Microbiol 18:489-98.
Castillejo-Rodriguez AM, Barco-Alcala E, Garcia-Gimeno RM, Zurera-Cosano
G. 2000. Growth modeling of Listeria monocytogenes in packaged fresh green
asparagus. Food Microbiol 17:421-7.
Church IJ, Parsons AL. 1995. Modified atmosphere packaging technology: A re-
view. J Sci Food Agric 67:143-52.
Devlieghere F, Debevere J, Impe J van. 1998. Effect of dissolved carbon dioxide
and temperature on the growth of Lactobacillus sake in modified atmospheres.
Intl J Food Microbiol 41:231-8.
Erkmen O. 2000. Predictive modeling of Listeria monocytogenes inactivation
under high pressure carbon dioxide. Lebensm Wiss Technol 33:514-9.
Jones PR, Greenfield PF. 1982. Effect of carbon dioxide on yeast growth and fer-
mentation. Enzyme Microb Tech 4:210-23.
Koutsoumanis PK, Taoukis PS, Drosinos EH, Nychas G-J. 2000. Applicability of an
Arrhenius Model for the combined effect of temperature and CO2 packaging
on the spoilage microflora of fish. Appl Environ Microbiol 66:3528-34.
Kuriyama H, Mahakarnchanabul W, Matsui S, Kobayaski H. 1993. The effects of
pCO2 on yeast growth and metabolism under continuous fermentation. Bio-
technol Lett 15:189-94.
Lannguelongue M, Finne G, Hanna R, Nickelson R, Vanderzant C. 1982. Micro-
biological and chemical changes during storage of swordfish (Xiphias gladius)
steaks in retail packages containing CO2-enriqued atmospheres. J Food Pro-
tect 45:1197-203.
Layrisse ME, Matches JR. 1984. Microbiological and chemical changes of spotted
shrimp (Pandalus platyceros) stored under modified atmospheres. J Food Prot
47:453-7.
Luedeking R, Piret EL. 1959. A kinetic study of the lactic acid fermentation. Batch
process at controlled pH. J Biochem MicrobTechnol Eng 1:393-412.
Lyhs U, Korkeala H, Vandamme P, Bjorkroth J. 2001. Lactobacillus alimentarius:
a specific spoilage organism in marinated herring. Intl J Food Microbiol 64:355-
60.
Figure 6—Patterns of CO released (black line, upper
graph), O content (grey lin2e, upper graph), LAB (᭹, lower
graph) an2d yeasts (᭺, lower graph) in ham pizza packaged
under 50% CO2:10% O2 in the presence of 500 mg/kg of
Nisaplin.
Mayer-Miebach E, Zanoni B, Speess WEL. 1997. A model to predict microbial
Vol. 68, Nr. 3, 2003—JOURNAL OF FOOD SCIENCE 1001

Kinetics of gas release in pizzas . . .
contamination of blanched spinach. Lebensm Wiss Technol 30:536-42.
Murado MA, González MP, Vázquez JA. 2002. Dose-response relationships. An
Westall S, Filtenborg O. 1998. Yeast occurrence in Danih feta cheese. Food Micro-
biol 15:215-22.
overview: A generative model and its application to the verification of de- MS 20020424 Submitted 7/16/02, Revised 9/12/02, Accepted 10/1/02, Received
scriptive models. Enzyme Microb Tech 31:439-55.
10/10/02
Samelis J, Kakouri A, Rementzis J. 2000. Selective effect of the product type and
packaging conditions on the species of lactic acid bacteria dominating the
spoilage microbial association of cooked meats at 4 °C. Food Microbiol 17:329-
40.
The authors thanks Luisa Iglesias, Lorena Barros, Irene Tarrío and Carlos Suárez for their
technical assistance. This work was supported by CICYT (Inter-ministerial Committee of
Science and Technology, PETRI project 95-0324-OP).
Sameshima T, Takeshita K, Akimoto M, Yamanaka H, Miki T, Arihara K, Itoh M,
Kondoh Y. 1997. Effect of various sugars on the swelling spoilage of cooked
meat products caused by heterofermentative lactic acid bacteria. J Jpn Soc Food
Sci Technol 44:855-61.
Authors Cabo, Pastoriza, and Rodrguez are with the Inst. de Investigacións
Mariñas. C/ Eduardo Cabello, 6. 36208 Vigo (Pontevedra). Direct inquiries
to author Cabo (E-mail: marta@iim.csic.es).
1002 JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 3, 2003