CHAPTER 20
Pathogens and Food-Poisoning
Bacteria in Cheese
20.1 INTRODUCTION
Milk is a highly nutritious medium of almost
neutral pH and, therefore, many bacteria, includ-
ing spoilage and pathogenic bacteria, can grow
in it. Numerous outbreaks of food poisoning
have been traced to milk. Although cheese is
equally nutritious, it has been shown to be re-
sponsible for relatively few food-poisoning out-
breaks. These are summarized in Table 20-1.
There were 21 confirmed outbreaks of food poi-
soning in Western Europe during the years
1970-1997, 7 outbreaks in the United States
from 1948 to 1997, and 4 outbreaks in Canada
from 1970 to 1997 due to consumption of cheese
("Food Safety and Cheese," 1998; Johnson,
Nelson, & Johnson, 199Oa, 199Ob, 199Oc). Dur-
ing the period 1970-1997, an estimated
235,000,000 tonnes of cheese were produced in
Western Europe, the United States, and Canada.
Such data imply that cheese is a relatively safe
food product, but 28% of the outbreaks involved
cheeses made from raw milk.
Several organisms were involved in cheese-
related outbreaks of food poisoning, but Salmo-
nella spp., Staphylococcus aureus, and Listeria
monocytogenes were the most common (Table
20-1). The species of organism involved in the
European outbreaks were more diverse than
those in the United States, which may reflect the
greater diversity of cheese varieties made in Eu-
rope compared with the United States and
Canada. The primary reasons for these cheese-
related food-poisoning outbreaks were poor
starter activity (due to phage, antibiotic residues
in the milk, etc.), poor hygiene in the plant, gross
environmental contamination, and faulty pas-
teurization.
In recent years, the most important pathogens
found in cheese have been L. monocytogenes
and enteropathogenic Escherichia coll. The
former is the more important since the outbreaks
involved several deaths. The outbreaks due to E.
coll 0157 involved cheese made from raw milk,
and only a few cases were involved. Neverthe-
less, a major outbreak due to E, coli 0157 has
the potential to be very serious. The major
cheeses involved were soft surface-ripened
cheese and cheese with a low acidity (e.g., Mexi-
can-style cheese). There is no indication that
Mycobacterium bovis and Brucella abortus and
the so-called emerging pathogens, Campylo-
bacter jejunii, Yersinia entercolitica, and
Aeromonas hydrophilia grow during manufac-
ture of cheese. M. bovis, which causes tubercu-
losis in cows and sometimes in humans, and B.
abortus, which causes abortion in cows and un-
dulant fever in humans, are sometimes excreted
in the milk of infected cows. These microorgan-
isms were probably important causes of human
disease in the past but are not important now,
since TB and brucellosis are controlled in almost
all dairy herds in developed countries. An out-
break of food poisoning in Canada involving
Table 20-1 Food-Poisoning Outbreaks Associated with Cheese in the United States (1948-1997), Canada (1970-1997), and Europe (1970-
1997)

Country Variety No. of No. of Causative

of Origin of Cheese Year Cases Deaths Organism(s) Reference

US Colby 1958 200 Staph. aureus Allen and Stovall, 1960

US Cheddar 1958 200 Staph. aureus Hendricksetal., 1959
US Cheddar, Kuninost, and 1965 Staph. aureus Zehren and Zehren, 1968
Monterey
US Cheddar 1976 >339 Salmonella Fontaine etal., 1980
US Homemade 1983 16 Sc. zooepidemicus Sharpe, 1987
US Mexican style 1985 142 48 L. monocytogenes Linnan etal., 1988
US Mozzarella 1989 164 O Salmonella javiana; Hedberg etal., 1992
Salmonella oranienberg
Mexico Mexican style3 1975 3 Brucella Eckman, 1975

Canada Emmental 1977 12 Staph. aureus Toddetal., 1979, 1981b

Canada Cheese curd 1980 62 Staph. aureus Toddetal., 1981 a
Canada Cheddar, other types 1982 Salmonella Sharpe, 1987
Canada Cheddar 1984 >2700 1 Salmonella typhimurium D'Aoustetal., 1985; D'Aoust, 1994
Salmonella Sharpe, 1987
Finland Farmhouse 1983 35 E. co//O124:B17 Marieretal., 1973
France Camembert, Brie, 1971 387 O
Coulommiersa
France Brieb 1983 >3000 E. co//O27:H20 MacDonaldetal., 1985
France Goat milk cheese0 1993 273 1 Salmonella paratyphi B Desenclos etal., 1996
France Brie de Meauxc 1995 20 4 Listeria monocytogenes Gouletetal., 1995
France Fromage fraisc 1992 NR 1 Veratoxic E. coll Public Health Laboratory Service,
London, 1994

France, Switzerland Soft cheese 1974 77 Cl. botulinum Kauf et al., 1974; Sebald et al.,
1974
Greece Homemade, unripened 1983 23 O Brucella Sharpe, 1987

Ireland Soft cheese0 1989 42 O Salmonella dublin Maguire etal., 1992

continues
Table 20-1 continued

Country Variety No. of No. of Causative

of Origin of Cheese Year Cases Deaths Organism(s) Reference

Italy Mascarpone 1996 8 1 C. botulinum Aureli etal., 1996

Malta Soft cheese 1995 135 1 Br. melitensis Public Health Laboratory Service,
London, 1995
Scandinavia Brie 1982 >50 Bacillus; Shigella sonnei Sharpe, 1987

Switzerland Vacherin Mont d'Or0 1983/87 122 34 L monocytogenes BiIIe, 1990; Malinverni et al., 1985

Switzerland, France Vacherin Mont d'Or 1985 >40 O Salmonella typhimurium SadiketaL, 1986
Switzerland, France Doubsc 1995 25 5 Salmonella dublin Valliant et al., 1996

UK Cheddar 1983 2 Staph. aureus Sharpe, 1987

UK Cheddar 1994 >84 O Salmonella gold-coast Public Health Laboratory Service,
London, 1997a
UK Sheep milk cheese0 1984 >13 O Staph. aureus Bone etal., 1989
UK Farmhouse cheese 1992 >20 O E co// O1 57 Public Health Laboratory Service,
Edinburgh, 1994
UK Lancashire0 1997 2 O E. co// O1 57 Public Health Laboratory Service,
London, 1997b
a
Consumed in the United States.
b
Consumed in the United States, Sweden, and The Netherlands.
c
These cheeses were made from raw milk.
Cheddar cheese made from raw milk was traced
to a farm where one cow was shedding about
200 cfu of salmonella per milliliter of milk.
The infective dose of pathogens can be low.
For instance, S. heidelberg has an infective dose
of 100-500 cells, and the infective dose for E.
coli O157:H7 is thought to be 10 cells. In con-
trast, the infective dose for S. aureus is high be-
cause the actual cause of the food poisoning is
not the organism itself but a number of closely
related, heat-stable protein toxins (able to with-
stand 10O0C for > 30 min) produced by it. These
toxins are called staphylococcal enterotoxin A
(SEA), SEB, SEC, etc., and some of them (e.g.,
SEC) are further subdivided into SECl, SEC2,
SEC3, etc. The amino acid sequences of all the
enterotoxins are very similar. Some of them
(e.g., SEA and SEE) are produced during expo-
nential growth, while others (e.g., SEB and
SEC) are produced mainly during the stationary
phase of growth. Those produced in the expo-
nential phase are the more common causes of
staphylococcal food poisoning. The minimum
number of staphylococci and the level of toxin
required to cause food poisoning are thought to
be 105 cells/g and 1 ng/g of food ingested. The
strains of S. aureus present in raw milk are pri-
marily those that cause mastitis, and about 20%
of these strains produce enterotoxin.
Table 20-2 Microbiological Quality (cfu/ml) of Milk Produced in Normandy, France
n Winter
Total count 39 71,000 27,000
Enterococci 37 74 150
Coliforms 29 57 2,400
S. aureus 25 450 1,700
L monocytogenes 39 4 positive samples
Salmonella 39 1 positive sample
Y. entercolitica* 39 19 positive samples
Campylobacter 39 1 positive sample
3
OnIy 1 of 61 isolates was a potential pathogen.
20.2 PATHOGENS IN RAW MILK
Recent surveys of raw milk quality show that
the incidence of pathogens in raw milk is low.
For example, about 5% of individual Irish, En-
glish, and French farm milks are contaminated
with Listeria and less than 1 % with Salmonella
(Desmasures, Bazin, & Gueguen, 1977; Minis-
try of Agriculture, Fisheries and Food, 1997;
O'Donnell, 1995; Rea, Cogan, & Tobin, 1992).
The microbiological quality of farm milks in
Normandy (France), around the area where raw
milk Camembert is made, is generally very
good; 83% of 69 samples had total bacterial
counts below 20,000 cfu/ml, and the samples
had an average somatic cell count of 176,000/ml
(Table 20-2). The average number of coliforms,
enterococci, and S. aureus was 77, 79, and 350
per milliliter, respectively, in summer milk.
Winter-produced milk was also good, with
counts of 57, 74, and 450 per milliliter for the
above bacterial groups, respectively. The inci-
dence of Yersinia entercolitica was relatively
high and that of Campylobacter low. All these
data suggest that hygiene and cooling of milk
were effective, although 14 of the 43 milks ex-
amined did not meet the European Union crite-
rion for S. aureus in milk destined for cheese-
making (< 500/ml) (Table 20-3).
n Spring/Summer
30 86,000 21,000
25 79 400
19 77 5,000
18 350 280
30 No positive samples
30 1 positive sample
30 6 positive samples
30 No positive samples
Table 20-3 European Union Guidelines (Standards) for Milk for Cheesemaking and Cheese

Soft Cheese Soft Cheese

from Raw Milk from Pasteurized Milk Raw Milk

Listeria Absent in 25 g* Absent in 25 g

m 5 5
c O O

Salmonella Absent in 25g Absent in 25 g

m 5 5
c O O

S. aureus
mt 1,000/g 100/g 500/ml*
M 10,000/g 1000/g 2,000/ml
n 5 5 5
c 2 2 2

Conforms
m 10,000/g
M 100,000/g
n 5
c 2

E. coll
m 10,000/g 100/g
M 100,000/g 1,000/g
n 5 5
c 2 2
Total plate count <100,000/ml
Somatic cells <400,000/ml

Key: n = number of samples, m = threshold value; the result is satisfactory if the number of bacteria in all sample units does not
exceed m. M = maximum value; the result is unsatisfactory if the number of bacteria in > 1 sample units exceeds M. c = number of
samples for which the counts may lie between m and M; the sample is considered acceptable if the numbers are < m in the other
sample units.

* The 25 g sample should consist of five 5 g portions taken from different parts of the same product.

t For products labeled "made from raw milk."

* Geometric mean of 2 results, with > 2 samples per unit.

Geometric mean of 3 results, with > 1 sample per unit.

20.3 PATHOGENSINCHEESE quently handled during ripening. Consequently,

Soft cheeses contain high levels of moisture,
and the pH, particularly at the surface, increases
during ripening. In addition, some of them are
made from raw milk, and smear cheeses are fre-
soft cheeses are more prone to microbial growth
than hard or semi-hard varieties. A recent survey
of the microbiological quality of soft cheeses in
the United Kingdom market is summarized in
Table 20-4. Cheeses from several countries, in-
Table 20-4 Microbiological Quality of Soft Cheese Made from Raw and Pasteurized Milk

10 to 102 to 103 to 104 to

ND* <1& <1& <1& <704 <1& >105

Raw milk cheeses (72)

Coliforms 32 8 5 3 9 2 13
E.co// 4 8 7 1 0 4 1 0 2
S . aureus 70 O 1 1 0 0 0
L . monocytogenes 7 1 1 O O O O O
Other Listeria spp. 68 3 O 0 1 0 0

Pasteurized Milk Cheeses (405)

Coliforms 284 13 38 19 23 9 19
E . coli 3 8 3 7 9 4 1 0 2
S . aureus 401 O 2 O 1 1 0
L . monocytogenes 4 0 3 2 0 0 0 0 0
Other Listeria spp. 4 0 0 5 0 0 0 0 0

Unlabeled cheeses (960)

Coliforms 611 37 71 54 96 30 61
E . coli 8 9 1 2 2 2 5 9 6 2 5
S . aureus 9 5 8 O O 1 1 0 0
L . monocytogenes 947 13 O O O O O
Other Listeria spp. 926 30 O 1 1 0 2

* Not detected.

+Total per gram.

eluding Scandinavia, France, Germany, Greece, 20.4 LISTERIOSIS

Cyprus, Italy, and the United Kingdom, were
analyzed. Salmonella were not found in any of
the 1,437 cheeses examined. Overall, the micro-
biological quality of both raw and pasteurized
milk cheese was quite good. One (1.4%) of the
raw milk cheeses failed to meet the European
Union criteria (Table 20-3) for Listeria in raw
milk cheese, and 2 (3%) failed to meet the crite-
ria for E. coli. There are no standards for coli-
form numbers in raw milk cheese, but 28 of the
405 (7%) samples of pasteurized milk cheese
examined failed the coliform standard, and 7
(2%) failed the Listeria standard (Table 20-3).
Coliforms are killed by pasteurization, and the
high numbers in some of the pasteurized milk
cheeses probably reflect the spread of contami-
nation during the smearing of surface-ripened
cheese (see Chapter 10).
Listeriosis is caused by L. monocytogenes, es-
pecially serotypes l/2a and 4b, and it affects
mainly pregnant women, immunocompromised
people (e.g., those who are HIV positive or are
recovering from chemotherapy after treatment
for cancer), and the elderly. Prominent symp-
toms include vomiting and diarrhea, which may
lead to meningitis and bacteremia. Infection of
the blood stream, the central nervous system, the
fetus in utero, and infants by mothers who show
no obvious signs of infection during birth are the
most common forms of listeriosis.
Two relatively recent major outbreaks of lis-
terosis have been traced to cheese, one in the
United States, involving Mexican-style cheese,
and one in Switzerland, involving Vacherin
Mont d'Or cheese, which is a soft cheese made
from raw milk. Both outbreaks involved fatali-
ties, 48 in the United States (20 fetuses, 10 in-
fants, and 18 nonpregnant adults) and 34 in
Switzerland (Table 20-1). Poor hygiene was the
major factor in both outbreaks, and improper
pasteurization was also implicated in the case of
the Mexican-style cheese. However, the fact that
both cheeses also had low salt levels and that the
Mexican-styIe product is a low-acid cheese,
made without the deliberate addition of starter
cultures, while Vacherin is a surface-ripened va-
riety, in which the pH increases during ripening,
contributed to the outbreaks.
L. monocytogenes is a Gram-positive rod that
can grow at temperatures from -0.40C to 450C,
at pH values of 4.4 to 9.4, and in the presence of
10% NaCl. Generation times at O0C and I 0 C are
131 and 62 hr, respectively, and the lag times are
33 and 3 days, respectively. Therefore, holding
cheese as close to O0C as possible will help to
prevent the growth of listeria. The optimum pH
and temperature are 7.0 and 370C, respectively.
These properties make this microorganism par-
ticularly problematical. It is an ubiquitous or-
ganism and is found in soil, water, silage, and so
on. Sometimes the organism is found inside
phagocytes (neutrophils and macrophages) in
milk, and this location was thought to protect the
cells from inactivation during pasteurization.
The general consensus now, however, is that the
organism is inactivated by pasteurization whe-
ther the cells are inside phagocytes or not.
20.5 PATHOGENIC ESCHERICHIA COLI
The normal habitat of E. coll is animal (and
human) feces whence it can contaminate raw
milk, particularly if the animals have been lying
in their own dung and the udders have not been
properly washed before milking. Most strains of
E. coli are harmless, commensal organisms, but
some are pathogenic. They are differentiated
from each other on the basis of the serological
detection of somatic (O), flagellae (H), and cap-
sular (K) antigens. To date, 174 O antigens, 56 H
antigens, and 80 K antigens have been detected.
Pathogenic strains are generally subdivided into
enteropathogenic (EPEC), enterotoxigenic
(ETEC), enteroinvasive (EIEC), and enterohem-
orrhagic (EHEC), depending on how they cause
infection.
Five outbreaks of foodborne disease due to
pathogenic E. coli have been traced to the con-
sumption of soft cheeses. These involved ETEC
O27:H20, EIEC O124:B17, and EHEC O157
(Table 20-1). The outbreak due to E. coli
O124:B17 occurred in the United States but in-
volved French Camembert, Brie, and Coulom-
miers cheese made in the same plant over a 2-
day period. Despite this, the cheese was widely
distributed, because outbreaks occurred in 14
states from Connecticut in the east to California
in the west; 389 people developed food poison-
ing, but no deaths were reported. In the outbreak
involving E. coli O27:H20, the cheese involved
was from two different lots made 46 days apart,
suggesting that contamination was intermittent.
It is not clear if the cheeses in these outbreaks
were made from raw or pasteurized milk. E. coli
O157:H7 is an emerging food pathogen that has
been associated with several severe food-poi-
soning outbreaks involving meat products and
also a small outbreak involving a raw milk
cheese (Table 20-1).
The symptoms of the E. coli 0124 infection
included diarrhea, fever, and nausea; cramps,
chills, vomiting, aches, and headaches were less
common. The median time before onset was 18
hr (range, 2^48 hr) and the median duration of
the illness was 2 days (range, less than 1 day to
15 days). The symptoms of the E. coli O27:H20
infection were fairly similar; the average time
before onset was 44 hr (range, 6-144 hr), and the
symptoms lasted for 4.4 days (range, 1-14
days). The typical symptoms of O157:H7 food
poisoning include production of stools contain-
ing blood and mucus, as a result of hemorrhagic
colitis, and acute renal failure, particularly in
children. Generally, a long incubation period (up
to 9 days, average 4 days) and duration of illness
(up to 9 days, average 4 days) elapse before the
onset of symptoms and the occurrence of deaths.
In contrast, in food poisoning due to other patho-
genic strains of E. coli, the symptoms occur
within 2 days of ingestion of the contaminated
food, and no deaths have been reported. Oral
challenges using human volunteers suggest that
10M010 cells of EPEC, 108-1010 cells of ETEC,
and 108 cells of EIEC are required to cause diar-
rhea. In contrast, only 10-100 cells of EHEC are
required to cause illness.
The precise mechanism by which E. coli
O157:H7 causes disease has not been fully eluci-
dated, but isolates produce 1 or 2 toxins that are
cytotoxic to Vero cells, an African green mon-
key kidney cell line. For this reason, E. coli
O157:H7 is also called verotoxigenic E. coli
(VTEC). One of these toxins is structurally and
immunologically indistinguishable from the
shiga toxin of Shigella dysenteriae. How E. coli
O157:H7 acquired the shiga toxin is not clear.
There is relatively little information on the in-
cidence in E. coli in raw milk or cheese, but the
evidence suggests that it is low. A relatively re-
cent survey of milks throughout the production
season (Rea, Cogan, & Tobin, 1992) showed
that more than 60-100% of samples contained
fewer than 10 E. coli per ml, depending on the
date of sampling (Figure 20-1). Very few milks
contained more than 100 E. coli per mililiter.
None of 568 raw milks examined in the United
Kingdom contained E. coli O157:H7 (Neaves,
Deacon, & Bell, 1994). In a limited survey, E.
coli O157:H7 was not detected in 19 soft and
semi-soft American-made cheese, while other
strains of E. coli were found in 11 samples
(Ansay & Casper, 1997). In a Spanish study of
221 raw milk cheeses and 75 pasteurized milk
cheeses, 3 cheeses (1.4%), each of which had
been produced from raw milk, showed the pres-
ence of toxigenic E. coli (Quinto & Cepeda,
1997).
E. coli O157:H7 has a minimum growth tem-
perature of 80C, an optimum of 370C, a maxi-
mum of 440C, and it does not withstand pasteur-
ization. E. coli strains normally do not tolerate
low pH values, but E. coli O157:H7 is an excep-
tion and can grow at pH 4.5 in media adjusted
with HClbut not if the pH is adjusted with lac-
tic acid. The organism does not grow in cheese
at pH values at or below 5.4.
20.6 GROWTH OF PATHOGENS DURING
CHEESE MANUFACTURE
The major factors responsible for the control
of microbial growth in cheese have been de-
scribed in Chapter 10. These factors are also in-
volved in controlling the growth of pathogens.
The main reason for the low incidence of food-
poisoning outbreaks caused by cheese is that
most milk for cheesemaking is pasteurized,
which kills all pathogens in the raw milk. This is
probably the most important factor in control-
ling the growth of potential pathogens in cheese.
There is some evidence that phagocytosis of L.
monocytogenes by somatic cells present in raw
milk increases the heat resistance of L. mono-
cytogenes, but the consensus is that this organ-
ism is still killed by normal pasteurization. Sig-
nificant amounts of cheese are made from raw
milk in countries bordering the Mediterranean
(~ 15% of all French cheese is made from raw
milk), and in some cases no starter is used (e.g.,
artisanal production of the Spanish cheeses
Manchego and Cabrales), which implies that
cheese made from raw milk is also safe, since
few outbreaks have been attributed to such prod-
ucts. Pasteurization is normally carried out at
720C for 15s, but lower heat treatments (e.g.,
650C for 16-18 s) will destroy all the likely
pathogenic microorganisms commonly found in
milk except, perhaps, L. monocytogenes. Sub-
pasteurization heat treatments are also used in
some countries (e.g., Canada) for milk for
cheesemaking. The reason for this practice is
that stronger flavored cheeses are produced from
milk that has not been pasteurized at all or has
been subpasteurized than from fully pasteurized
milk (see Chapter 15).
M. paratuberculosis causes paratuberculosis
(Johne's disease) in cattle, and there is some evi-
dence that this organism also may be involved in
the etiology of Crohn's disease in humans. Be-
cause of this evidence, milk has been suggested
as a possible vehicle for the transmission of the
organism from cattle to man. Whether the organ-
ism withstands normal pasteurization is impor-
tant. There is conflicting evidence on this point
% of samples
Week of year
Figure 20-1 Incidence ofEscherichia coli in raw milks during the year.
in the literature, and some cheese manufacturers
have recently increased the temperature of pas-
teurization by a few degrees and/or the time of
pasteurization by a few seconds to ensure that
the milk is free of this organism. These more se-
vere conditions may damage the rennetability of
the milk.
The presence and survival of pathogens in
cheese is influenced by several factors, includ-
ing
species
initial numbers
physiological condition of the microorgan-
ism
rate of acid production by the starter and
consequent decrease in pH
tolerance of the pathogen to acid and salt
tolerance to the cooking temperature to
which the cheese curd is subjected
postmanufacturing contamination
<10 du/ml
10-<100 du/ml
>100 cfu/ml
biochemical changes that occur in the
cheese during ripening
ripening and storage temperature of the
cheese
composition of the cheese
The time-temperature profile during cheese-
making, the initial low rate of decrease in pH,
and the cooking temperature, which can vary
from 330C for Camembert (essentially no cook-
ing) to 360C for Dutch-type cheese, 380C for
Cheddar, and 540C for many Swiss and Italian
cheeses, play major roles in promoting the
growth of pathogens in cheese during manufac-
ture. A temperature of 25-4O0C is conducive to
the growth of pathogens, if they are present. The
cooking temperature (540C) and the length of
time for which Swiss-type cheese curds are held
at this temperature (~ 60 min) will kill most, if
not all, pathogens. In soft cheese, a cooking tem-
perature of 350C or lower is used, which is ideal
for the growth of pathogens. Obviously, if the
starter is active, the pH will decrease quickly and
the growth of the pathogens will be retarded.
The reverse also occurs. That is, if growth of the
starter is slow due to phage contamination and/
or antibiotic residues in the milk, considerable
growth of pathogens can occur. Therefore, a fast
acid-producing starter is one of the best means
of controlling the growth of pathogens in cheese.
In studies on the growth of pathogens in
cheese, the milk is normally inoculated with the
pathogens, pregrown under favorable condi-
tions. The numbers found in naturally contami-
nated milk would be much lower and probably
in a more stressed physiological state than those
grown in laboratory media. In such a stressed
state, bacteria are probably less resistant to the
effects of pH and temperature, but experimental
evidence for this hypothesis is not available, pri-
marily because of the difficulty of obtaining
naturally contaminated milk. Many raw milks
are, in fact, free of pathogenic microorganisms.
Data for the growth of E. coli O157:H7, L
monocytogenes, and S. aureus in Cheddar
cheese curd during manufacture are shown in
Figure 20-2. E. coli and S. aureus multiplied
during manufacture but L. monocytogenes did
not. When interpreting these data, one must re-
member that the moisture content in the curd de-
creases at each stage of manufacture, resulting in
an apparent increase in bacterial numbers due to
the concentration effect. Based on this phenom-
enon, it appears that a small decrease in the
number of listeria occurred during manufacture.
Considerable growth of E. coli occurred be-
tween the beginning of manufacture and cutting
the coagulum, when little acid production would
have occurred. No data were reported for S.
aureus between these two stages, but growth of
this organism is also likely to occur. These data
refer to Cheddar cheese curds, in which acid pro-
duction is relatively rapid (acid production will
also have a major rate-limiting effect on
growth). Growth of pathogens in the curd of
most other varieties would probably be greater
than in Cheddar because of slower acid pro-
duction.
The growth of most pathogens will slow down
and eventually cease owing to the decrease in
pH (as significant amounts of lactic acid are pro-
duced) and also to the increase in temperature
during cooking.
20.7 GROWTH OF PATHOGENS IN
CHEESE DURING RIPENING
What happens during ripening of a cheese de-
pends on the variety. Each cheese is a unique
microbial ecosystem and should probably be
considered individually. Nevertheless, broad
generalizations can be made. Hard and semi-
hard cheeses, if made properly, are safe, since
almost all pathogens die off during ripening; in
contrast, significant growth of pathogens can oc-
cur in soft cheese.
20.7.1 Hard and Semi-Hard Cheeses
Coliform bacteria die off at a rate of 0.3 log
per week in Cheddar and 0.7 log per week in
Gouda. The fate of several pathogens in Em-
mental and Cheddar cheese is shown in Figures
20-3 and 20-4. Both Emmental and Cheddar
cheeses are hard cheeses with similar pH values
(~ 5.2) immediately after manufacture. None of
the pathogens, except S. aureus at very low lev-
els, was detected in the Emmental cheese within
1 day of manufacture. This is most likely a result
of the high cook temperature (~ 520C) used in
the manufacture of this cheese. In Cheddar
cheese, S. aureus, E, faecalis, E. coli, and a Sal-
monella species all decreased during ripening,
and the Gram-negative bacteria decreased at a
faster rate than the Gram-positive organisms
(Figure 20-4). One of the problems with S.
aureus is that, even though the numbers of the
organism decrease significantly during ripening,
sufficiently high numbers may have been pres-
ent during the early stages of ripening to produce
the small amounts of enterotoxin necessary to
cause food poisoning. Therefore, it is possible
that a cheese with a low level of S. aureus may
contain a high level of enterotoxin. Enterotoxins
are proteins, and whether they are hydrolyzed by
 E. coli
L. monocytogenes
S. aureus
Cfu/ml or g

Curd after Cooking

Curd after Salting

Curd at Pressing
Curd after Heating
Curd after Cutting
Beginning of
manufacture

Figure 20-2 Growth of E. coli 0157, L. monocytogenes, and S. aureus in Cheddar cheese curd during manufac-
ture.

chymosin or bacterial proteinases during cheese Cheddar, and a pH of 5.7 indicates poor starter
ripening does not appear to have been studied. In activity, either as a result of phage contamina-
the United States, storage of cheese at 20C for 60 tion or antibiotic residues in the milk.
days may be used instead of pasteurization. In Tilsit, a semi-hard cheese, the numbers of
The number of L. monocytogenes in Cheddar all the pathogens tested decreased during ripen-
cheese also decreases during ripening (Figure ing, except L. monocytogenes, which remained
20-5), but some variations in individual trials fairly constant during ripening for 30 days, after
occur. There was also some variation in the rate which a gradual decrease of about 1 log cycle
of die-off of L. monocytogenes in cheese ripened occurred over the following 2 months (Figure
at 130C and in cheese ripened at 60C, but gener- 20-3). The stability of L. monocytogenes in this
ally the differences were small. E. coli O157:H7 cheese was attributed to the relatively low cook-
died off relatively rapidly (2 log cycles in 25 ing temperature and short cooking time (420C
days) in Cheddar cheese during ripening at for 15 min), which were bacteriostatic rather
6.50C (Figure 20-6). than bactericidal. pH may also be important. The
The effect of the pH of the cheese is also criti- pH of the Tilsit cheese increased during ripening
cal. Data for Salmonella in Cheddar cheese are from 5.2 at day 1 to 5.8 at day 90. Commercial
shown in Figure 20-7. At pH 5.03 and 5.23, they samples normally have a pH of about 6.2 at 90
died off quite quickly, but at pH 5.7 they did not days. L. monocytogenes can grow over a wide
die at all. A pH of 5.23 is typical of a well-made range of temperature, from -I0C to 450C, and
 Emmental cheese

L. monocytogenes
E. coll
S. aureus
Y. enterocolitica
Log cfti/ml or g

P. aeruginosa
S. typhimurium

Milk Curd Cheese Cheese Cheese Cheese Cheese

after lday 7 days 30 days 60 days 90 days
cooking

Tilsit cheese L. monocytogenes

E. coli
S. aureus
Y. enterolytica
P. aeruginosa
S. typhimurium
A. hydrophila
Log cfu/ml or g

C. jejuni

Milk Curd Cheese Cheese Cheese Cheese Cheese

after lday 7 days 30 days 60 days 90 days
cooking

Figure 20-3 Growth of L. monocytogenes, E. coli, S. aureus, Y. enterocolitica, P. aeruginosa, S. typhimurium,

Aeromonas hydrophila, and Campylobacter jejuni in Emmental and Tilsit cheese during ripening.
 S. aureus
E. faecalis
E. coli
Salmonella
Log cfti/g

Days

Figure 204 Decrease in numbers of S. aureus, E. faecalis, E. coli, and Salmonella in Cheddar cheese during
ripening at 120C.

this ability may be important for its survival in
Tilsit cheese.
20.7.2 Soft Cheeses
The situation in soft mold- and smear-ripened
varieties like Camembert, Brie, and Limburger
is quite different, and many pathogens can grow
readily in such cheeses. The reasons for this are
as follows:
These cheeses have a relatively high mois-
ture content.
They are ripened at a temperature (10
150C) at which bacterial growth can occur.

Trial 1
Trial 2
Trials
Count, cfa/g

Time, days
Figure 20-5 Growth of L. monocytogenes Scott A in three trials of Cheddar cheese during ripening at 60C. The
variation that occurred in different trials is clearly seen.

Log, cfu/g
Days
Figure 20-6 Survival of E. coli O157:H7 in Cheddar
cheese during ripening.
The pH increases during ripening, espe-
cially at the surface, owing to metabolism
of lactate by Geotrichum candidium and the
Penicillium spp., to a point where growth of
bacterial contaminants can occur (see
Chapter 10).
The increase in the pH of soft cheese during
ripening is significantly greater at the surface
Salmonella count, cfu/g
Time, days
Figure 20-7 Effect of pH on the survival of Salmonella in Cheddar cheese during ripening.
than in the interior of the cheese (see Chapter 10)
and is conducive to the growth of some patho-
gens. Growth of L. monocytogenes Scott A, E.
coli B2C, Enterobacter aerogenes MFl, and
Hafnia strain 14-1 occurred in Camembert
cheese during manufacture. The strain of E. coli
used was an enterotoxigenic strain. Hafnia are
closely related to coliform bacteria, and only one
species, H. alvei, appears to occur in water and
the feces of humans and animals. What happens
to these organisms during ripening is shown in
Figure 20-8. The number of L. monocytogenes
decreased initially during ripening but increased
again once the pH rose above 6. The number
also increased in the core but not to the same ex-
tent, probably because the pH increased more
slowly. In contrast, the number of E. coli and
Ent. aerogenes increased during manufacture
but began to decrease once the pH of the curd
reached 5.0, and the number continued to de-
crease during ripening. This pattern of response
probably characterizes all coliforms, with
Hafnia strain 14-1 as an exception to the rule.
The number of Hafnia strain 14-1 increased until
the pH fell to around 5; then it remained constant
and decreased to 10 cfu/g during the first week
pH 5.03
pH 5.25
pH 5.53
pH 5.70
 L. monocytogenes Scott A
log cfu/g
Hd
Time, weeks
Hafnia, strain 14-1
log cfu/g
Hd
Time, weeks
Figure 20-8 Growth of L monocytogenes, E. coli, Enterobacter aerogenes, and Hafnia of strain 14-1 and the
increase in pH in Camembert cheese during ripening.
of ripening. The number began to increase again
as soon as the pH began to increase, reaching a
final cell number of 108 cfu/g. The rate of in-
crease in the pH of the three Camembert cheeses
varied. This was probably due to differences in
the manufacturing procedures and in the rate of
growth of the different strains of yeast and P.
camemberti used. Of course, it is the combined
effect of the temperature during ripening, the
salt concentration, and the decrease in pH that
really determine the extent of growth of patho-
gens.
Blue cheese is also a soft variety, but L. mono-
cytogenes dies out during ripening even though
the pH increases from about 4.6 at day 1 to 6.2
after 10 days. The death of Listeria in Blue
cheese has been attributed to inhibition of their
>tf. aerogenes MFl
Log cell no
Hd
Time, weeks
E. coli B2C
log cfu/g
H*
Time, weeks
growth by the high level of salt ( 10% salt-in-
moisture) in these cheeses. The growth of other
pathogens on the surface of soft cheeses does not
appear to have been investigated.
Significant differences in pH between the
core and the surface of soft cheese develop dur-
ing ripening. This difference creates problems in
obtaining representative samples of these
cheeses for analysis. Wedge-shaped samples are
the most representative of soft cheese (see Chap-
ter 23).
20.8 RAW MILK CHEESES
Cheese made from raw milk has a much stron-
ger flavor than the same cheese made from pas-
teurized milk, and this is an important marketing
advantage for raw milk cheeses. Nevertheless,
from the foregoing it is clear that soft cheeses
can be problematic, and those made from raw
milk are particularly so. S. aureus is a common
cause of mastitis in dairy cows and therefore is
probably present in most raw milks. Approxi-
mately 20% of the S. aureus strains present in
raw milk produce enterotoxin. Growth of such
strains to high numbers could therefore cause
problems in cheeses made from raw milk. It is
also likely that E. coli 0157:H7 is present in raw
milk, as its major source is bovine feces. Despite
this, only two food-poisoning outbreaksand
these were small oneshave been traced to
cheese containing E. coli O157:H7 (Table 20-
1). Small numbers of L. monocytogenes may
also be present in raw milk and may subse-
quently grow in the cheese. S. aureus and E. coli
grow during cheese manufacture and are poten-
tial problems in cheeses made from raw milk. In
addition, soft mold- and smear-ripened cheeses
with a high moisture content and in which the
pH increases, especially at the surface, during
ripening are potentially hazardous, especially
when they are made from raw milk.
To produce a good-quality raw milk cheese,
free from pathogens, the raw milk should
be of good quality, with bacterial counts
below 20,000 cfu/ml and somatic cell
counts below 400,000/ml
be produced under extremely hygienic con-
ditions
be free of pathogens
be held at a low temperature (< 40C)
be made with an active (fast acid-produc-
ing) starter
have good quality control procedures in
place, including hazard analysis critical
control points (HACCP)
HACCP was developed by the U.S. space pro-
gram to ensure the microbiological safety of
foods for astronauts. It involves identifying the
microbiological hazards and preventive mea-
sures that can be taken at each step in the manu-
facture of the product. The points at which con-
trol is critical to managing the safety of the
product are then identified and limits are set.
These critical control points are then monitored
and recorded during the manufacture of each
subsequent batch of product. In cheese manufac-
ture, pasteurization and the pH of the curd at a
predetermined time after addition of the starter,
which estimates the rapidity of acid develop-
ment, are obvious critical control points.
The European Union standards for different
pathogens in raw milk and in soft cheese made
from raw or pasteurized milk are shown in Table
20-3. Listeria and Salmonella must be absent in
25 g, and a distinction is made between soft
cheeses produced from raw milk and those made
from pasteurized milk, with more stringent stan-
dards being set for the former. This reflects the
propensity of pathogens to grow in raw milk
cheeses during manufacture.
20.9 CONTROL OF THE GROWTH OF
PATHOGENS
To prevent the growth of pathogens, it is im-
perative to prevent contamination of the milk
and cheese and to be meticulously hygienic. To-
day, much cheese is made in automated systems,
but small-scale artisanal production involves
manual manipulation of the curd during manu-
facture, molding, and ripening. Good hygiene is
critical at each of these steps. Implementation of
HACCP systems is also very effective in pre-
venting the growth of pathogens in cheese. An
active, phage-free starter and pasteurization are
major critical control points. The activity of the
starter should be assessed by determining the pH
of every batch of cheese at a preset time after
starter inoculation, such as 10 hr and 24 hr in the
case of soft cheeses and Cheddar, respectively.
Comparisons of the data on a daily basis will in-
dicate if starter activity is normal. Soft cheeses
are small and will cool quickly. Therefore, keep-
ing the ambient temperature high is important
when the curds are in the molds.
Good hygiene is particularly important in the
manufacture of smear cheeses, especially where
old smear is used to inoculate the fresh cheeses.
Old smear may be contaminated with L. mono-
cytogenes and may infect all cheeses. The use of
old smear is traditional in the production of these
cheeses, and efforts are being made to develop
defined-strain smear starters to overcome the
problem. For example, much attention is being
focused on identifying smear bacteria that pro-
duce bacteriocins active against L. mono-
cytogenes. Application of such cultures to the
surface of a cheese should be very useful in help-
ing to prevent the growth of Listeria spp. The
direct application of bacteriocins produced by
lactic acid bacteria (LAB) that inhibit Listeria on
the surface of cheese is also being advocated as
an effective method for controlling listerial
growth on cheese.
As already indicated, several factors are in-
volved in controlling the growth of pathogens
(and other organisms) in cheese: pH, tempera-
ture, and the level of salt are probably the most
important. The combined effect of these factors
on growth is much greater than each factor's in-
dividual effect, and in recent years there has
been a major effort to develop models for pre-
dicting the growth of pathogens in foods based
on their growth responses to combinations of
salt, temperature, andpH. These predictive mod-
els have been developed mainly from experi-
ments carried out in complex media, and the re-
sults are felt to reflect the worst-case scenario in
foods, since growth in foods at the same tem-
perature, salt concentration, and pH value is gen-
erally less than in model systems. For example,
soft cheeses often have a salt concentration of
1.5%, have a pH of 6.5, and are stored at 50C.
Using these parameters, the model predicts that
an initial level of 10 L monocytogenes cells/g
would multiply to 10,000/g in 10 days (a genera-
tion time of about 1 day).
20.10 ENTEROCOCCI
Enterococci are found at high numbers in
many cheeses, particularly those made around
the Mediterranean. Many of these are artisanal
raw milk cheeses made at the farmhouse level.
Enterococci are considered to be important in
the development of flavor in these cheeses,
where they comprise a significant proportion of
the microflora of the raw milk and starter. Their
ability to metabolize lactose and their tolerance
to salt and heat make them ideal candidates as
starters. Enterococci at levels above 107/g have
been found in such cheeses, and these high num-
bers would almost certainly play a role in flavor
development.
There is considerable debate as to whether en-
terococci should be considered pathogens. Dur-
ing the past few decades, they have been impli-
cated in several diseases, including bacteremia,
urinary tract infections, and endocarditis. Many
strains are very promiscuous and easily pick up
plasmids that encode vancomycin resistance.
Many of these plasmids also are conjugative and
can be transferred naturally from cell to cell by
sexual combination. Vancomycin is a glycopep-
tide antibiotic that acts by inhibiting cell wall
biosynthesis. The incidence of vancomycin-re-
sistant enterococci (VRE) in hospitals has in-
creased dramatically. The use of avoparcin,
which is also a glycopeptide antibiotic, as a
growth promoter in animal feed has been impli-
cated in the increased occurrence of VREs in
farm animals, including pigs and poultry. Be-
cause of this, the use of avoparcin has been
banned recently in several European countries.
Many VREs are difficult to deal with because
they are also resistant to other therapeutic antibi-
otics, implying that alternative antibiotic therapy
may not be available. However, many bacteria,
including starter LAB like Lactobacillus, Pedio-
coccus, and Leuconostoc, are intrinsically resis-
tant to vancomycin.
There is little information on how rapidly En-
ter'ococcus spp. grow in milk, but in Cheddar
cheese they remain fairly constant during ripen-
ing. Data for other commercial cheeses are
sparse in the literature, but data for some
artisanal Spanish and Italian cheeses are shown
in Figure 20-9. Casar de Caceres and La Serena
are made from raw ewe milk and Afuega'l Pitu
from raw cow milk. No starters are used for any
of the three cheeses. Pecorino Umbro is made
from pasteurized ewe milk and a mesophilic
starter is also used. A surface microflora devel-
ops on some of these cheeses, but the counts in
Figure 20-9 are for the internal part of each
cheese. The first point on each line in the figure
is the number of enterococci present in the milk
at the beginning of manufacture. The data show
that considerable growth occurred during manu-
facture and during the first days of ripening, af-
ter which the numbers remained constant, except
for Afuega'l Pitu cheese, in which the number
decreased. The numbers of enterococci in Casar
de Caceres and La Serena cheese were well in
excess of 106/g and probably contribute to ripen-
ing. Sometimes, it is difficult to distinguish be-
tween lactococci and enterococci. However, the
Enterococci, log cfu/g
counts of enterococci shown in Figure 20-9 are
reliable, as selective media were used to enu-
merate them.
20.11 BIOGENIC AMINES
Biogenic amines can be formed through de-
carboxylation of amino acids by some strains of
nonstarter LAB, particularly Lb. buchneri, dur-
ing cheese ripening. Tyramine, produced from
tyrosine, is probably the most important. These
amines, which can cause food intoxication
within a few hours of ingestion, are discussed in
Chapter 21.

Casar de Caceres

La Serena

Afuega'l Pitu

Pecorino Umbro

Time, days

Figure 20-9 Growth of enterococci in Casar de Caceres (+), La Serena (), Afuega'l Pitu (A), and Pecorino
Umbro () cheeses during ripening. The first point on each line is the count in the milk.
REFERENCES
Allen, V.D., & Stovall, W.D. (1960). Laboratory aspects of
staphylococcal food poisoning from Colby cheese. Jour-
nal of Milk and Food Technology, 23, 271274.
Ansay, S.E., & Casper, C.W. (1997). Survey of retail
cheeses, dairy processing environments and raw milk for
Escherichia coli O157:H7. Letters in Applied Microbiol-
ogy, 25, 111-114.
Aureli, P., Franciosa, G., & Pourshaban, M. (1996). Food-
borne botulism in Italy. Lancet, 348, 1594.
Bille, J. (1990). Epidemiology of human listeriosis in Europe
with special reference to the Swiss outbreak. In AJ.
Miller, J.L. Smith, & G.A. Somkuti (Eds.), Foodborne
listerosis. Amsterdam: Elsevier.
Bone, F.J., Bogie, D., & Morgan-Jones, S.C. (1989). Staphy-
lococcal food poisoning from sheep milk cheese. Epide-
miology and Infection, 103, 449^58.
D'Aoust, J.Y. (1994). Salmonella and international trade.
International Journal of Food Microbiology, 24, 1131.
D'Aoust, J.Y., Warburton D.W., & Sewell, A.M. (1985).
Salmonella typhimurium phage type 10 from Cheddar
cheese implicated in a major Canadian foodborne out-
break. Journal of Food Protection, 48, 1062-1066.
Desenclos, J.C., Bouvet, P., Benz-Lemoine, E., Grimont, F.,
Desqueyroux, H., Rebiere, L, & Grimont, P.A. (1996).
Large outbreak of Salmonella enterica serotype para-
typhi B infection caused by a goats' milk cheese, France,
1993: A case finding and epidemiological study. British
Medical Journal, 312, 91-94.
Desmasures, N., Bazin, F., & Gueguen, M. (1997). Micro-
biological composition of raw milk from selected farmers
in the Camembert region of Normandy. Journal of Ap-
plied Microbiology, 83, 53-58.
Eckman, M.R. (1975). Brucellosis linked to Mexican cheese.
Journal of the American Medical Association, 232, 636
637.
Fontaine, R.E., Cohen, M.L., Martin, W.T., & Vernon, T.M.
(1980). Epidemic salmonellosis from Cheddar cheese:
Surveillance and prevention. American Journal of Epide-
miology, 111, 247-253.
Food safety and cheese. (1998). International Food Safety
News, 7, 6-10.
Goulet, V., Jacquet, C., Vaillent, V., Rebiere, L, Mouret, E.,
Lorente, C., Maillot, E., Stanier, F., & Rocourt, F. (1995).
Listeriosis from consumption of raw-milk cheese. Lan-
cet, 345, 1581-1582.
Hedberg, C.W., Korlath, J.A., D'Aoust, J.Y., White, K.E.,
Schell, W.L., Miller, M.R., Cameron, D.N., MacDonald,
K.L., & Osterholm, M.T. (1992). A multistate outbreak of
Salmonella javiana and Salmonella oranienberg infec-
tions due to contaminated cheese. Journal of the Ameri-
can Medical Association, 268, 3203-3207.
Hendricks, S.L., Belknap, R.A., & Hausler, WJ. (1959). Sta-
phylococcal food intoxication due to Cheddar cheese: 1.
Epidemiology. Journal of Milk and Food Technology, 22,
313-317.
Johnson, E.A., Nelson, J.H., & Johnson, M. (199Oa). Micro-
biological safety of cheese made from heat-treated milk.
1. Executive summary, introduction and history. Journal
of Milk and Food Technology, 53, 441^52.
Johnson, E.A., Nelson, J.H., & Johnson, M. (199Ob). Micro-
biological safety of cheese made from heat-treated milk:
2. Microbiology. Journal of Milk and Food Technology,
53,519-540.
Johnson, E.A., Nelson, J.H., & Johnson, M. (199Oc). Micro-
biological safety of cheese made from heat-treated milk:
3. Technology, discussions, recommendations, bibliogra-
phy. Journal of Milk and Food Technology, 53, 610623.
Kauf, C., Lorent, J.P., Mosimann, J., Schlatter, L, Somanini,
B., & Velvart. J. (1974). Botulismusepidemic vom type
B. Schweizerische Medizinische Wochenschrift, 104,
677-685.
Linnan, MJ., Mascola, M., Lou, X.O., Goulet, V., May, S.,
Salminen, C., Hird, D.W., Yonekura, L., Hayes, P.,
Weaver, R., Andurier, A., Plikaytis, B.D., Fannin, S.L.,
Kicks, A., & Broome, C.V. (1988). Epidemic listeriosis
associated with Mexican-style cheese. New England
Journal of Medicine, 319, 823-828.
MacDonald, K.L., Eidson, M., Strohmeyer, C., Levy, E.,
Wells, J.G., Puhr, N.D., Wachsmuth, K., Nargett, N.T., &
Cohen, M.L. (1985). A multistate outbreak of gastrointes-
tinal illness caused by enterotoxigenic Escherichia coli in
imported semisoft cheese. Journal of Infectious Diseases,
757,716-720.
Maguire, H.C.F., Boyle, M., Lewis, MJ., Pankhurst, J.,
Wieneke, A.A., Jacob, M., Bruce, J., & O'Mahony, M.
(1992). An outbreak of Salmonella dublin infection in
England and Wales associated with a soft, unpasteurised
cow's milk cheese. Epidemiology and Infection, 109,
389-396.
Malinverni, R., Bille, J., Perret, C., Regli, F., Tanner, F., &
Glauser, M.P. (1985). Listeriose epidemique. Schweizer-
ische Medizinische Wochenschrift, 115, 2-10.
Marier, R., Wells, J.G., Swanson, R.C., Callahan, W., &
Mehlman, IJ. (1973, December 15). An outbreak of en-
teropathogenic Escherichia coli foodborne disease traced
to imported French cheese. Lancet, 2, 1376-1378.
Ministry of Agriculture, Fisheries and Food. (1997). Micro-
biological survey: Raw cow's milk on retail sale. Food
Safety Information Bulletin, 12-13.
Neaves, P., Deacon, J., & Bell, C. (1994). A survey of the
incidence of E. coli 0157 in the UK dairy industry. Inter-
national Dairy Journal, 4, 679696.
O'Donnell, E.T. (1995). The incidence of Salmonella and
Listeria in raw milk from farm bulk tanks in England and
Wales. Journal of the Society of Dairy Technology, 48,
25-29.
Public Health Laboratory Service, Edinburgh. (1994). E. coli
0157 phage type 28 infections in Grampian. Communi-
cable Diseases and Environmental Health, 28 (No. 94/
46), 1.
Public Health Laboratory Service, London. (1994). Two
clusters of haemolytic uremic syndrome in France. Com-
municable Disease Report, 4(1}, 29.
Public Health Laboratory Service, London. (1995). Brucel-
losis associated with unpasteurised milk products abroad.
Communicable Disease Report, 5(32), 151.
Public Health Laboratory Service, London. (1997a). Salmo-
nella gold-coast and Cheddar cheese [Update]. Commu-
nicable Disease Report, 7(11), 93, 96.
Public Health Laboratory Service, London. (1997b). Vero
cytotoxin producing Escherichia coli 0157. Communi-
cable Disease Report, 7(46), 409, 412.
Quinto, E.J., & Cepada, A. (1997). Incidence of toxigenic
Escherichia coli in soft cheese made from raw milk. Let-
ters in Applied Microbiology, 24, 291-295.
Rea, M.C., Cogan, T.M., & Tobin, S. (1992). Incidence of
pathogenic bacteria in raw milk in Ireland. Journal of
Applied Bacteriology, 73, 331-336.
Sadik, C., Krending, MJ., Mean, F., Aubort, J.D.,
Schneider, P.A., & Roussianos, D. (1986). An epidemic-
logical investigation following an infection by Salmo-
nella typhimurium due to the ingestion of cheese made
from raw milk. In Proceedings of the Second World Con-
SUGGESTED READINGS
Bell, C., & Kyriakides, A. (1998a). E. coli: A practical ap-
proach to the organism and its control in foods. London:
Blackie Academic and Professional.
Bell, C., & Kyriakides, A. (1998b). Listeria: A practical ap-
proach to the organism and its control in foods. London:
Blackie Academic and Professional.
Doyle, M.P., Beuchat, L.R., & Montville, TJ. (1997). Food
gress on Foodborne Infections and Intoxications (Vol. 1).
Berlin.
Sebald, M., Jouglard, J., & Gilles, G. (1974). Type B botu-
lism in man due to cheese. Annales de I 'Institut Pasteur,
125A(3), 349-357.
Sharpe, J.C.M. (1987). Infections associated with milk and
dairy products in Europe and North America, 1980-85.
Bulletin of the World Health Organization, 65, 397^06.
Todd, E.C.D., Shelley, D., Szabo, R., Robem, H., Gleeson,
T., Durante, A., Marcoux, A., Entis, P., Morrison, D.,
Purvis, U., Foster, R., Burgener, D.M., Wright, W.W.,
Maharaja, R.S., Brodsky, M., Magus, M., Ruf, F.W., &
Shab, H.W. (1979). Staphyloccal intoxication from
Swiss-type cheese: Quebec and Ontario. Canadian Dis-
eases Weekly Report, 5(26), 110-112.
Todd, E.C.D., Szabo, R., Gardiner, M.A., Akhtar, M.,
Delorme, L., Tourillon, P., Moisan, S., Rochefort, J.,
Roy, D., Loit, A., Lamontagne, Y., Gosselin, L.,
Martineau, G., & Breton, J.P. (198Ia). Staphylococcal
intoxication from cheese curds: Quebec. Canadian Dis-
eases Weekly Report, 7(34), 171-172.
Todd, E.C.D., Szabo, R., Robern, H., Gleeson, T., Park, C.,
& Clark, D.S. (198Ib). Variation in counts, enterotoxin
levels and TNase in Swiss-type cheese contaminated with
Staphylococcus aureus. Journal of Food Protection, 44,
839-846,851-852,856.
Vaillant, V., Haeghebaert, S., & Desenclos, J.C. (1996). Out-
break of Salmonella dublin infection in France, Novem-
ber-December 1995. Eurosurveillance, 1, 2, 9-10.
Zehren, V.L., & Zehren, V.F. (1968). Examination of large
quantities of cheese for Staphylococcal enterotoxin A.
Journal of Dairy Science, 51, 635-644.
microbiology. Washington, DC: American Society for
Microbiology.
Microorganisms in foods 5. (1996). In Microbiological
specifications of food pathogens. London: Blackie Aca-
demic and Professional.
Mortimore, S., & Wallace, C. (1994). HACCP: A practical
approach. London: Chapman & Hall.