Impedimetric method for selective enumeration of specifie yoghurt bacteria with milk-based culture media

Lait (1995) 75, 285-293
© Elsevier/INRA
Original article
Impedimetric method for selective enumeration
of specifie yoghurt bacteria
with milk-based culture media
F Pirovanol, 1Piazza-, F Brarnbillaê, T Sozzi '
1 Silia- YOMO SpA, Centro Ricerche,
Str per Merlino 3, 20060 Zelo Buon Persico;
2 Sitia-YOMO
SPA, Laboratorio
Microbiologico,
20080 Pasturago
di Vernate, MI, Italy
(Received
13 May 1994; accepted 21 March 1995)
Summary
- The suitability
of the impedimetric
method
for the selective
enumeration
of specifie
microorganisms,
particularly yoghurt bacteria, Streptococcus
thermophilus
and Lactobacillus
delbruec-
kii subsp bulgaricus, was investigated
in comparison
with the standard plate cou nt technique.
A series
of samples were analyzed
by using new milk-based
culture media (milk added with saccharose
and
acidified milk) for the impedance
technique,
whereas for the conventional
counting method M17 and
MRS culture media were applied. It was found that the milk-based
culture media are suitable for the
selective enumeration
of specifie yoghurt bacteria: the values detected by the impedimetric
measure-
ment (Bactometer
instrument)
and those resulting
from the standard
plate count showed
a good
correlation. The suggested milk-based culture media are shown to be suitable for the selective counting
of yoghurt
bacteria
when applied
in the impedimetric
technique.
Furthermore,
since the bacterial
counting may be determined
within 12 h, the impedimetric
method is also faster than the conventional
method.
impedimetric
method
/ selective
enumeration
/ yoghurt
/ milk-based
culture
medium
Résumé
- Méthode
impédimétrique
pour
le dénombrement
sélectif
de la flore
spécifique
du
yaourt
en utilisant
des milieux
de culture
à base de lait. La validité de la méthode de l'impédance
pour le dénombrement
microbien de la flore spécifique du yaourt, Streptococcus
thermophilus
et Lac-
tobacillus
delbrueckii
subsp bulgaricus,
a été comparée
avec la détermination
des microorganismes
réalisée par la méthode traditionnelle
de numération
en boîtes de Pétri. Plusieurs échantillons
ont été
analysés en appliquant les nouveaux milieux sélectifs de culture à base de lait (lait sucré et lait acidifié)
pour ce qui concerne la méthode impédimétrique
et en utilisant, parallèlement,
les milieux de culture
M17 et MRS pour la méthode de référence.
On a vérifié que les milieux de culture à base de lait sont
les plus appropriés
pour la détermination
de la population
microbienne
du yaourt par la mesure de
l'impédance.
Les donnés obtenues par la mesure de l'impédance
et ceux de la méthode convention-
nelle montrent une très bonne corrélation.
Les milieux de culture à base de lait, en appliquant
la mé-
thode impédimétrique.
peuvent assurer une détermination
très simplifiée
et rapide de la flore bacté-
rienne du yaourt. Cette méthode est beaucoup plus rapide que la méthode de référence parce qu'on
peut obtenir les données de la numération
bactérienne
dans les 12 h.
méthode
impédimétrique
/ dénombrement
microbien
/ yaourt
/ milieu
de culture
à base de lait

---

286
F Pirovano et al
INTRODUCTION
.
carico,
1983; Nieuwenhof
and Hoolfwerf,
1987), and other dairy products (Khayat et al,
Bacterial
enumeration
carried
out by the
1988; Story and White, 1988) as weil as for
plate count technique (CFU), which was de-
those belonging to other production
fields
veloped
in the past century,
has already
(pharmaceuticals, cosmetics, etc) (Kahn and
been accepted as a reference procedure for
Firstenberg-Eden,
1985;
Edmondson,
the hygienic-sanitary
and marketing evalu-
1988).
ation of a product and, furthermore,
is the
only officially recognized method of analysis.
Furthermore,
the method
is used for a
rapid
and accu rate detection
of bacte-
Nevertheless,
the method
has been
riophages
in starter cultures and, generally
shown not to be free from restrictions, mostly
speaking,
in every fermentation
process
because
of the long reaction times which
(Hose et al, 1983; Waes and Bossuyt, 1984).
sometimes
do not comply
with those of
people working in a production sector. If that
Besides
estimating
the total bacterial
is the case, the end-product, in order to be
count in a particular product such as yoghurt,
marketed, should be manufactured accord-
is of great importance to determine the num-
ing to the regulations and the laws in force.
ber of specific
microorganisms,
Strepto-
The impedimetric technique has also been
coccus thermophilus and Lactobacillus del-
proposed in the past century, but only since
brueckii subsp bulgaricus, which must be
the last 10 years it has been used as rapid
present
in high numbers
and in a correct
counting
method,
alternatively
with the
ratio.
standard
plate count. The principle of the
method, ie the measurement of the changes
ln a previous study, Neviani et al (1991)
occurring in a substrate as evidence of a bac-
obtained satisfactory results in the selective
terial metabolism,
has been demonstrated
count of the two microorganisms specific for
(Quesneau, 1983; Richardson, 1985): this is
yoghurt by using the Malthus instrument and
one of the points which mostly differs from
the modified culture media M17 and MRS.
the traditional
plate count based upon the
evaluation of the visible biomass (colonies).
The aim of the present study is to use milk-
based culture media which are quite different
The impedimetric technique is weil known
from M17 and MRS. As weil as being of easy
and usually applied in quality control labo ra-
preparation, they allow cost reduction in la-
tories of large industries, particularly for the
bour and materials and may be easily applied
total enumeration
of microorganisms,
esti-
also in a production field.
mation of contaminants and pathogens.
This technique
allows to determine
in a
rapid, repeatable, accu rate way and most of
MATERIALS
AND METHODS
ail, at lower costs, the shelf-life and bacterio-
logical quality of raw mate rials for different
Cultures
food-products
(Hardy et al, 1975; Williams
and Wood,
1982; Zindulis,
1984; Easter
and Gibson, 1985; Van Spreekens,
1986:
Cultures
used were: i) pure strains of Strepto-
Gibson, 1987; Banks et al, 1989) and par-
coccus thermophilus
and Lactobacillus
delbruec-
ticularly in meat (Martins and Selby, 1980;
kii subsp
bulgaricus
belonging
to the Centro
Firstenberg-Eden,
1983;
Whittingham,
Ricerche
Vomo collection;
H) starters
used for
1989), milk (Gnan and Luedecke,
1982;
industrial manufacturing
of yoghurt; and iii) yog-
O'Toole,
1983; Firstenberg-Eden
and Tri-
hurt samples from factories.

---

Impedimetric
method for yoghurt bacteria
287
Determination of lactic acid bacteria
This relationship
between DT and initial con-
centration
of microorganisms
allows the instru-
ment to be used for the bacterial counting.
Conventional method (FIL-lOF:
1986,1988)
Selective enumeration by the
impedance of Streptococcus
Bacterial
content
has been
estimated
by the
count on plate of decimal dilutions
in M17 and
thermophilus and Lactobacillus
MRS culture
media.
MRS was acidified
at pH
delbrueckii subsp bulgaricus
5.40 by addition of acetic acid.
in mixed cultures
The plates containing
M17 culture medium,
specifically
developed
for the enumeration
of
streptococci,
were incubated
at 37°C for 48 h,
Culture media
whereas
MRS plates (pH 5.40), applied for the
enumeration
of lactobacilli,
were incubated
an-
Skimmed milk powder (Oxoid, Unipath Lld, Bas-
aerobically
at 37°C for 72 h.
ingstoke,
Hampshire,
UK) reconstituted
at 10%
was modified as follows: i) by addition of 10, 15,
20% saccharose,
and sterilizing
in autoclave
at
Impedimetric method
110°C for 20 min for streptococci
test; ii) acidifi-
cation at pH 5.00 by 2 N HCI, after sterilization
in
autoclave at 110°C for 10 min for lactobacilli test.
The instrument
Bactometer
M 128 (Biomérieux,
Rome, Italy) was used for measuring
the detec-
tion time (DT) of the tested sample.
Determination of conductance curves
The DT is the time elapsing between the start
of a test and the detection
of an accelerating
The impedimetric
detection
was carried out on
impedance
signal by the instrument.
samples at 24 and 72 h from their preparation.
The DT is a function of numerous parameters,
Before being tested, the samples were diluted
including the initial concentration
of microorgan-
1: 10 in the culture medium to be applied for the
isms, their lag phase, generation
time and meta-
evaluation
and subsequently
homogenized
in
bolic activity.
Stomacher
for 1 min.
54.0
48.0
" 42.0
36.0
c
30.0
24.0
18.0
y
-:
12.0
6.0
0.0
-6.0
0.0
2.4
4.8
7.2
9.6
12.0
14.4
Hi.B
19.2
21.6
24.0
HOURS
Fig 1. Conductance
curves of a yoghurt culture in the following culture media: ...... milk;
milk
+ 15% saccharose;
- - - acidified milk, pH 5.00.
Courbes de conductance
d'une culture pour yaourt dans les milieux de culture suivants
:
lait ;
___
lait + 15% saccharose;
- - -lait acidifié pH 5,00.

---

288
F Pirovano et al
Table
1. Selectivity
factor
of milk
+ 15%
Table Il. Selectivity
factor of acidified milk (pH
saccharose
for Streptococcus
salivarius subsp
5.00)
for
Lactobacillus
delbrueckii
subsp
thermophilus
(inoculum
1: 1000).
bulgaricus
(inoculum
1:1000).
Facteur de sélectivité
du lait + 15% saccharose
Facteur de sélectivité
du lait acidifié (pH 5,00)
envers
Streptococcus
salivarius
subsp
envers
Lactobacillus
delbrueckii
subsp
thermophilus
(inoculum
1:1000).
bulgaricus
(inoculum
1:1000).
Yoghurt
Initial
Selectivity
Yoghurt
Initial
Selectivity
culture
CFUlml
tector
culture
CFUlml
tector
(x 1e1)
(x 1e1)
St
0.890
20
St
0.630
1
Lb
0.194
1
Lb
0.210
170
2
2
St
1.100
30
St
5.450
1
Lb
0.150
1
Lb
0.150
500
3
3
St
0.830
40
St
0.215
1
Lb
0.145
1
Lb
0.300
100
4
4
St
0.790
27
St
0.855
1
Lb
0.144
1
Lb
0.230
200
5
5
St
0.510
48
St
0.535
1
Lb
0.163
1
Lb
0.260
180
6
6
St
0.770
45
St
0.300
1
Lb
0.190
1
Lb
0.380
100
• Selectivity
factor: N/No relationship
between
• Selectivity
factor: N/No relationship
between
the CFU/g at DT and initial CFU/g.
the CFU/g at DT and initial CFU/g.
Decimal dilutions were obtained trorn the in-
calibration
lines for the selective enumeration
of
itial dilution and placed into the incubator
of the
both species.
system (BPU).
The cultures in sweetened
milk and those in
Check of selectivity
acidified
milk were incubated,
respectively,
at
42°C for the enumeration
of streptococci,
and at
Selectivity of culture media, both sweetened
milk
47°C for the enumeration
of lactobacilli
(fig 1).
and acidified
milk, alter
being inoculated
with
different percentages
(1, 0.1, 0.01 %) of yoghurt
cultures was evaluated
by using the plate count
technique
as described above.
Relationship
between detection time
and microbial content
Determination
of a selectivity factor'
Detection
times for ail sam pie were correlated
with the corresponding
logarithm of the bacterial
A selectivity factor NINa was established,
where
contents determined
by the standard plate count.
the two parameters
N and N° indicate,
respec-
The Iinear regression
of these data gives the
tively, the CFU/g at the DT and the initial CFU/g.

---

Impedimetric
method for yoghurt bacteria
289
Table III. Average values of bacterial enumeration
with standard plate count and with impedimetric
method.
Valeurs moyennes du dénombrement
microbien en boîtes de Pétri et avec la méthode impédimétrique.
Yoghurt
Sample
Plate count
Bactometer
count
number
(x 1(l;g)
(x 1(l;g)
St
Lb
St
Lb
Natural whole milk
24
450
200
380
250
Whole milk apricot
20
180
200
300
260
Whole milk cherry
19
200
210
320
190
Whole milk citrus fruits
20
310
260
540
240
Whole milk ananas
18
240
225
210
195
Whole milk ban ana
22
280
230
360
190
Whole milk strawberry
26
260
230
420
240
Whole milk bilberry
22
300
250
360
230
Whole milk plum
14
260
220
350
260
Whole milk apple
9
330
310
310
300
Malt
18
240
300
260
290
The selectivity
factor
indicates
the cellular
appear to develop regularly with a selectivity
growth of Streptococcus
thermophilus
or Lacto-
factor equal or higher than 20. Changes in
bacillus delbrueckii subsp bulgaricus,
inoculated
conductance to determine the DT are there-
1: 1000 in the corresponding
culture media (milk
fore only due to the occurrence
of strepto-
added with saccharose
or acidified
milk) in the
time period
occurring
between
the incubation
cocci.
time and the DT.
Acidified
milk-based
culture medium
RESUL TS AND DISCUSSION
for the selective
enumeration
of
Milk-based
culture medium added
L delbrueckii
subsp bulgaricus
with saccharose
for the selective
enumeration
ofS thermophilus
It was found that by acidifying the milk and in-
creasing the incubation temperature to 47°C,
It was observed that the addition of 15% sac-
the streptococci growth was delayed without
charose to the milk allows to delay remark-
influencing the lactobacilli development.
As
ably the development
of Lactobacillus
del-
shown in table Il, in ail six samples of yoghurt
brueckii subsp bulgaricus without negatively
culture the selectivity factor for streptococci is
affecting the growth of streptococci.
substantially
equal to 1. This suggests that
As shown in table l, in ail six samples of
their cellular growth is practically
insignifi-
yoghurt cultures the selectivity factor for lac-
cant. Lactobacilli may, on the contrary, de-
tobacilli
and streptococci
is substantially
velop regularly and show selectivity factors
equal to 1. This is to show that their growth,
equal or higher th an 100. So changes in con-
as increasing number of cells, is practically
ductance to determine the DT are solely due
insignificant.
On the contrary, streptococci
to the occurrence of lactobacilli in the culture.

---

290
F Pirovano et al
Table IV. Average values of bacterial enumeration
with standard plate count and with impedimetric
method.
Valeurs moyennes du dénombrement
microbien en boîtes de Petri et avec la méthode impédimétrique.
Yoghurt
Sample
Plate count
Bactometer
count
number
(x 1(lIg)
(x 10%)
St
Lb
St
Lb
Natural skim-milk
31
510
220
550
230
Skim-milk peach
+ passion fruit
13
550
240
590
250
Skim-milk strawberry
+ ban ana
12
500
200
660
200
Skim-milk ananas
+ grapefruit
13
600
250
540
220
Skim-milk orange
+ tangerine
13
600
220
500
200
Skim-milk plum
12
610
250
550
270
Skim-milk lemon
13
710
260
880
300
Skim-milk strawberry
11
650
280
640
260
Skim-milk wild fruits
10
690
260
610
250
Skim-milk natural corn
11
590
270
660
330
Comparative
enumeration
of lac tic
From the tables III, IV and V it follows that
acid bacteria by using the standard
a remarkable
correspondence
occurs be-
plate method (Petri dishes)
tween the values obtained by the traditional
and impedimetric
method
plate count technique and those calculated
by the impedimetric
method: this becomes
Ca 400 samples of yoghurt taken from stor-
more evident when the correlation coefficient
(never below -0.98 for lactobacilli and -0.97
age-room at +4°C after 24 h of storage were
for streptococci) among the points of different
analysed by the plate count and impedance
calibration
curves is evaluated
(figs 2, 3).
to generate different calibration curves which
Furthermore,
figures
2 and 3 show good
correspond to the same number of products.
correlation of values both for a low and high
Because of the high number of samples as
bacterial content.
weil as the heterogeneity of their composition
(whole milk yoghurts, whole milk fruit yog-
CONCLUSIONS
hurt, skim-milk yoghurt) it is possible to verity
the repeatability of the method and the relia-
From an evaluation
of the results the two
bility of the suggested culture media. In order
suggested
culture media (sweetened
milk
the determine the repeatability of the method
and acidified
milk) may be considered
as
the standard ISO 5725 was used: the repea-
suitable for the selective enumeration of spe-
tability was 8% for streptococci and 20% for
cific yoghurt bacteria utilizing the impedance
lactobacilli, largely below the 30% reported in
measurement
detected by the Bactometer
Standard FIL 117A:1988.
instrument.

---

Impedimetric
method for yoghurt bacteria
291
Table V. Average values of bacterial enumeration
with standard
plate count and with impedimetric
method.
Valeurs moyennes du dénombrement
microbien en boites de Petri et avec la méthode impédimétrique.
Yoghurt
Sample
Plate count
Bactometer
count
with cream
number
(x 1cflg)
(x 106;g)
St
Lb
St
Lb
Type 1
17
380
7
390
9
Type 2
11
330
8
360
8
Type 3
21
390
7
350
7
10.0
9.0
L
8.0
0
7.0
G
6.0
C
F
5.0
U
1
4.0
m
1
3.0
2.0
1.0
0.0
0.0
1.4
2.8
4.2
5.6
7.0
8.4
9.8
11.2
12.6
14.0
HOURS
Fig 2. Regression
line of Streptococcus
thermophilus
in whole milk yoghurt
with strawberry.
Log
CFU/ml: -0.42*DT + 8.44. Correlation:
-0.97.
Droite de régression
de Streptococcus
thermophilus
dans le yaourt entier à la fraise.
The reliability of the proposed method is
longing to the specifie yoghurt flora which is
confirmed by the satisfactory, sometimes ex-
not so important for the plate count and, on
cellent, correspondence
between the values
the contrary, is very significant for the results
detected by the Bactometer and those result-
detected by the Bactometer.
ing fram the plate cou nt in factory both with
To conclude,
the impedimetric
method
high and low values of bacterial content.
could represent an effective alternative to the
Any under-evaluations
or overestimations
traditional
plate cou nt technique
since the
of the bacterial content are, in our opinion,
data obtained by the impedimetric technique
due to the metabolic activity of the strains be-
with their milk-based
culture
media are

---

292
F Pirovano et al
10.0
9.0
L
8.0
0
7.0
o
G
6.0
C
5.0
F
U
4.0
o
1
o
3.0
m
o
1
2.0
1.0
0.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
HOURS
Fig 3. Regression
line of Lactobacillus
delbrueckii
subsp
bulgaricus
in whole
milk yoghurt
with
strawberry.
Log CFU/ml: -0.62'DT
+ 8.05. Correlation:
-0.98.
Droite de régression
de Lactobacillus
delbrueckii
subsp bulgaricus
dans le yaourt entier à la fraise.
largely comparable to those obtained by the
Edmondson
AS (1988) Use of the bactometer for
conventional method. Furthermore, the cul-
testing medical devices.
In: 1988 Bactometer
ture media used for the selective enumera-
Workshop (Paper presented
at the Bactome-
tion provided good results because of the ex-
ter Workshop, Windsor, UK, 10-11 May, 1988)
Bactomatic
Lld, Henley-on-Thames,
UK
cellent
factors
of selectivity,
the good
correlation coefficients of calibration curves,
FIL-lOF 1986: E-DOC-249
- Group E44 - Iden-
tification
and
enumeration
of characteristic
the good repeatability, the easiness of prep-
microorganisms
in fermented
milks
aration and the lower costs. Another point in
favour
of this innovative
method
is the
FIL-lOF 1988: Standard 117 A - Yoghurt. Enume-
ration of characteristic
microorganisms.
Colo-
rapidity of results, which may be determined .
ny count technique
at 37°C
in a short time, within 12 h.
Firstenberg-Eden
R (1983) Rapid estimation
of
the number of microorganisms
in raw meat by
impedance
measurement.
Food Techno/37,
REFERENCES
64-70
Firstenberg-Eden
R, Tricarico
MK (1983) Impe-
Banks JG, Rossiter LM, Clark AE (1989) Selec-
dimetric determination
of total, mesophilic and
tive detection
of Pseudomonas
in foods by a
psychrotrophic
counts in raw milk. J Food Sci
conductance
technique.
In: Rapid
Methods
48, 1750-1754
and Automation
in Microbiology
and Immuno-
Gibson
DM (1987)
Some modifications
to the
10gy(Balows A, Tilton RC, Turano A, eds) Proc
media for rapid automated detection of Salmo-
Fifth Int Symp on Rapid Methods
and Auto-
nellas by conductance
measurements.
J Appl
mation in Microbiology
and Immunology,
Flo-
Bacteriol63,
299-304
rence 4-6 November,
1987, Brixia Academie
Gnan S, Luedecke LO (1982) Impedance
meas-
Press, Brescia, Italy
urements
in raw milk as an alternative
to the
Easter MC, Gibson DM (1985) Rapid and auto-
standard plate count. J Food Prot45,
4-7
mated detection
of Salmonella
by electrical
Hardy D, Dufour SW, Kraeger SJ (1975) Rapid
measurement.
J Hyg 94, 245-262
detection of frozen food bacteria byautomated

---

Impedimetric
method for yoghurt bacteria
293
Impedance
measurements.
Institute of Food
Quesneau
R (1983) La mesure de l'impédance.
Technologists,
Annual Meeting, Atlanta, Geor-
Tech Lait 974, 62-64
gia
Richardson
GH (1985) Alternative
microbiologi-
Hose
H, Sozzi
T, Bugnard
C, Marchesini
B
cal methods: electrical impedance
method. In:
(1983)
/mpedimetric
detectian
of bacteria-
Standard
methads
for examina tian of dairy
phages.
Leatherhead
Food RA Publication,
products. American Public Health Association.
Technical
Notes Nr4, 13-17
Washington,
DC, 15th edn, 165-171
Kahn P, Firstenberg-Eden
(1985) An Impedance
Story DA, White CH (1988) Impedance microbio-
test for the microbia/agica/
screening
of phar-
logy as a total quality assurance
tool for the
maceutica/ water. American Society for Micro-
dairy industry. J Dairy Sei 77, Suppl 1, 112
biology, Annual Meeting, Washington,
DC
Khayat FA, Bruhn JC, Aichardson
GH (1988) A
Van Spreekens
KJA, Stekelenburg
FK (1986)
survey
of coliforms
and Staphy/acaccus
au-
Rapid estimation
of the bacteriological
qualily
reus in cheese using impedimetric
and plate
of fresh fish by impedance
measurements.
count methods. J Food Prot 51, 53-55
App/ Micrabia/ Biatechna/24,
95-96
Martins SB, Selby MJ (1980) Evaluation of rapid
Waes GM, Bossuyt RG (1984) Impedance meas-
method for the quantitative
estimation
of coli-
urements
to detect bacteriophages
problems
forms
in meat
by impedimetric
procedures.
in Cheddar
cheesemaking.
J Food Prot 47,
App/ Environ Microbia/39,
518-524
349-351
Neviani E, Bossi MG, Brambilla F, Gatti M (1991)
Whittingham
L (1989) A medium for the selective
Conteggio differenziato
di Streptacaccus
ther-
impedimetric
detection of Staph aureus in raw
maphilus
et Lactabacillus
de/brueckii
subsp
meat. Poster presented
at the Society for Ap-
bu/garicus
in yogurt
utilizzando
la tecnica
plied Bacteriology Annual Meeting, Edinburgh,
conduttimetrica.
/nd Latte 27, 35-54
11-15 July
Nieuwenhof
FFJ, Hoolwerf JD (1987) Impedance
measurement
as an alternative
to the plate
Williams AP, Wood JM (1982) /mpedimetric
esti-
count method for estimating
the total count of
mation
of mau/ds.
Leatherhead
Food
Re-
bacteria in raw milk. J Food Prot 50, 665-668
search Association,
Technical
Circular N 791
O'Toole
DK (1983) Methods
for the direct and
Zindulis J (1984) A medium for the impedimetric
indirect assessment
of the bacterial content of
detection
of yeast in foods. Food Microbia/1,
milk: a review. J App/ Bacteria/55,
187-201
159-167

---