631
Journal of Food Protection, Vol. 52, No. 9, Pages 631-637 (September 1989)
Copyright© International Association of Milk, Food and Environmental Sanitarians
Thermal Inactivation of Staphylococcus aureus in Retentates
from Ultrafiltered Milk
JEFFREY L. KORNACKI and ELMER H. MARTH*
Department of Food Science and The Food Research Institute. University of Wisconsin. Madison. Wisconsin 53706
(Received for publication December 28, 1988)
ABSTRACT
Cells of Staphylococcus aureus strains 196E, 481, and 425
were thermally stressed at 56°C for 10 min in milk and enumerated
on Plate Count Agar (PCA), Mannitol Salt Agar (MSA), and PCA
with an overlay of MSA. PCA recovered more S. aureus 196E and
481 than did PCA/MSA, which recovered more than MSA. PCA/
MSA recovered slightly more S. aureus 425 than did PCA, which re-
covered more than MSA. At 58°C, in order of decreasing heat
resistance, the four strains of S. aureus originally isolated from food
were 425 > 100 and 481 > 196E. Their D-values were 26,14,13, and
3.0 min, respectively. S. aureus 425 was more heat resistant in the
stationary than in the log phase when heated at58°C in whole milk.
Heat resistance at 58°C increased overall during the stationary
growth phase, but was fairly stable when the culture was from 17 to
25 h or from 41 to 49 h old. S. aureus 425 exhibited no consistent
differences in heat resistance in concentrated (4X by volume) and
unconcentrated skim or whole milk. Adjustments of protein (3.5-
4.0% to 12.6-16%), milkfat (0.28-1.12% to 10%), and lactose (ca.
4.5-5.0% to ca. 14.5-15%) contents of milk and 4X (volume concen-
tration) UF milk retentates afforded no significant thermal protec-
tion to S. aureus 425. Diafiltration of 4X skim milk reduced thermal
protection of S. aureus 425 in the retentate over that of unconcen-
trated skim milk of the same lot when tested at 63 and 74°C. S. aureus
425 had greatest D-values (min) in skim milk (0.36 + 0.05) and
permeate (0.30 + 0.14) followed by permeate from diafiltration (0.28
± 0.06) when tested at 63°C.
On-the-farm concentration of milk by ultrafiltration (UF),
which is technically feasible (4, 18), reduces transportation
costs for the farmer (19). Slack et al. (19) suggested that UF of
raw whole milk on the farm is economically viable for farmers
with herds of 100 or more cows because of lower costs for
refrigeration and for hauling of ultrafiltered rather than uncon-
centrated raw whole milk because of reduction in fluid volume
and thus weight. However, before on-the-farm use of UF
systems can became a reality the following must occur: (a) a
suitable basis of payment must be developed for UF milk, (b)
modifications must be made in cheese manufacturing (19),
and (r) an understanding must be developed of the time/
temperature parameters required to destroy pathogenic micro-
organisms in UF milk retentates.
JOURNAL OF FOOD PROTECTION. VOL. 52, SEPTEMBER 1989
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022
Evidence in the literature suggests that thermal resistance
of some bacteria in UF milk retentates might be greater than
in unconcentrated milks. Dega et al. (2) showed that the heat
resistance of several strains of Salmonella and one of Escher-
ichia coli increased as the solids content of reconstituted skim
milk was raised from 10 to 50%. Perez et al. (15) found that
whole milk was more thermally protective of Streptococcus
faecium, Streptococcus durans, Streptococcus faecalis var.
faecalis, and 5treptococcusfaecalis var. liquefaciens than was
reconstituted skim milk or 0.1 M 3-3, dimethylglutaric acid
buffer.
Speck et al. (20) found that as the solids-not-fat portion of
chocolate milk mix was increased there was an increase in the
required pasteurization time, but no evidence was obtained
that suggested sugar, stabilizer, chocolate, or milk solids
affected the results in any way other than through the solids
they contributed. Singh and Ranganathan (17) compared the
heat-resistance of three E. coli strains in skim and whole milk
of cows, and in whole milk of buffaloes. Heat resistance was
greatest in whole milk of buffaloes followed by that in whole
and skim milk of cows. The authors attributed their results to
the higher solids and milkfat contents in buffalo milk than in
cow milk.
Grieme and Barbano (5) reported one trial in which
Staphylococcus aureus MF31 had greater D-values in milk
concentrated to 1.36X by reverse osmosis than in whole milk.
Haggerty and Potter (7) found no significant difference be-
tween D-values at 62.7°C for S. faecalis and E. coli in skim
milk, a 2X skim milk retentate, a 2X skim milk retentate
diluted to 1X with water, and a 2X skim milk retentate diluted
to IX with permeate. They also obtained similar results with
S. aureus; however, the D-value for S. aureus (1.32 min) at
62.7°C in the water-diluted retentate was slightly but signifi-
cantly lower than in the other milk samples (average, 1.71
min).
More work is needed to determine the effect concentra-
tion of milk with UF has on heat resistance of microorganisms,
especially those pathogenic to humans. This report provides
additional information about the heat resistance of S. aureus
in retentates from ultrafiltered milk.
632 KORNACKI AND MARTH
MATERIALS AND METHODS
Cultures
S. aureus strains 100, 196 E, 481, and 425 were used in this
study and were obtained from M. S. Bergdoll of the Food Research
Institute, University of Wisconsin-Madison. The cultures were
examined using the gram stain and were tested for their ability to
grow and produce acid on Mannitol Salt Agar (MSA), and for
production of catalase, coagulase, and thermonuclease.
Ultrafiltration
Ultrafiltration of skim and whole pasteurized milk was done
with a Hollow Fiber Concentrator Model DC-2, and a hollow fiber
cartridge (H1P50-20) (Amicon, Danvers, MA) at (ca. 22°C) or
below (10°C) room temperature.
Heating menstruum
Pasteurized, homogenized milk was obtained from the Univer-
sity of Wisconsin dairy factory unless otherwise indicated. Before
this milk or a UF retentate prepared from it was used as a heating
menstruum, all were heated at 78-80"C for 10 min (unless otherwise
indicated) to reduce the microbial load.
Heat treatments
The method used to heat-treat the liquids was a modification of
that of Grosche et al. (6). Five ml of heating menstruum was added
to eleven 25 x 150-mm screw-capped presterilized (autoclaved 15
min at 121°C) test tubes. Eight of these were tempered in a Thelco
Model 84 water bath (Precision Scientific, Chicago, IL) at the
appropriate temperature. A standardized thermometer (Ertco, New
York, NY) certified by the National Bureau of Standards to be
accurate to + 0.1°C was placed into one of the eight tubes. Each of
three other tubes received 0.05 ml of the S. aureus culture from a
Pipetman Automatic Pipettor P200 (Gilson) equipped with a steril-
ized plastic tip. The contents were mixed, tubes were capped and
placed in an ice bath. A sample was taken and plated to determined
the initial level of inoculum. After the heating menstruum was at the
appropriate temperature, 0.05 ml of the same culture was carefully
added with the automatic pipetting device to each of the seven
remaining tubes at convenient intervals (e.g. 1 min). The level of
water in the water bath outside the tubes was always at least 1 in.
higher than the level of milk inside the tubes. The plastic tip of the
pipettor was placed immediately above the tempered menstruum
and the culture was dispensed to prevent splashing of liquid onto the
unheated portion of the test tube. Contents of tubes were mixed
carefully to avoid splashing the heating menstruum onto portions of
tubes which were not submerged in water. Immediately upon inocu-
lation, tubes were capped and a stop watch was started. After the
appropriate heating interval, 5 ml of ice cold sterile water was
injected immediately into each tube using a Cornwall Pipettor
(Becton, Dickinson and Co., Rutherford, NJ) and tubes were placed
in an ice bath. When more than one strain of S. aureus or type of
heating menstruum were to be tested, inoculation of the tubes was
alternated between strains or heating menstrua and done at regular
intervals so that any change in temperature of the water bath during
the experiment would occur similarly in the suspensions. Tempera-
tures and times used in these experiments were 58°C-21 min, 63°C-
1 min, 73°C-10 s, and 78°C-5 s. Tubes in the trial at 58"C were
inoculated 1 min apart. At the other temperatures each tube was
inoculated, mixed, capped, and cooled before another was inocu-
lated. Samples of the unheated, inoculated heating menstruum were
also diluted and plated.
JOURNAL. OF FOOD PROTECTION. VOL. 52, SEPTEMBER 1989
Enumeration of bacteria
After heat-treatment, the unheated uninoculated menstruum,
the unheated S. aureus- inoculated menstruum, and the menstruum
inoculated with S. aureus and heated were diluted appropriately with
0.1 % sterile peptone solution and plated with Plate Count Agar
(PCA) except when indicated otherwise. Counts were estimated
from plates which contained from 30-300 colonies after incubation
at 37"C for 2 d.
Analyses
Protein was measured by the Kjeldahl method (8) at the Soils
Analysis Laboratory, Madison, Wisconsin. Fat was determined at
least in duplicate by the Babcock method (12). Total solids were
determined in duplicate using an Automatic Volatility Computer
Model AVC-MP (CEM Corporation; Indian Trail, NC). The pH was
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022
measured using a combination glass electrode and a pH meter (Orion
model 601 A Digital Ionanalyzer, Cambridge, MA).
Determination ofD-values and z-values
The average plate count of the contents of the unheated S.
awn?w.y-inoculated tubes was considered to be the initial S. aureus
inoculum before heating (0 time) if the number was appreciably (ca.
three orders of magnitude) greater than in the unheated inoculated
sample. Values from duplicate plate counts of the S. aureus-'mocu-
lated heated samples were added because the samples had been
dil uted 2-fold through addition of ice-cold sterile water. Counts from
the unheated inoculated samples (milk had been previously pre-
heated at 78-80"C for 10 min; except where indicated otherwise, to
reduce the count) were subtracted from the counts of the heated S.
aureus-'moculaXe.d sample to obtain the S. aureus count after heating.
D-values were calculated as the negative inverse slope of the line
generated from a plot of the log of bacteria count versus time. For a
typical heating trial, seven D-values would thus be generated per
treatment. z-Values were calculated as the negative inverse slope of
the line generated from a plot of the log|() of the D-values vs.
temperature.
Statistical analysis
All T-tests were done using the Minitab Statistical Computing
System (Copyright, Pennsylvania State University, 1982), MACC
version 82.1-UW2.3 or Minitab data analysis software (Standard
Emulated 8087 Version, release PC 82.1.1. Copyright, Minitab, Inc.,
1984). The SAS General Linear Models Procedure (16) was used to
compare the regression lines generated from plots of the Log1(l D-
values determined for S. aureus vs. temperature in concentrated and
unconcentrated milk from the same lots.
RESULTS AND DISCUSSION
Characteristics of cultures used
All four strains of S. aureus used were previously isolated
from food (cake, ham, milk, and cheese), produced entero-
toxin A (strain 196E produced enterotoxins A and D; M. S.
Bergdoll personal communication), appeared characteristi-
cally in grape-like clusters of gram-positive cocci upon micro-
scopic examination (1), grew characteristically (1) on Manni-
tol Salt Agar (MSA), and were catalase-, coagulase-, and
thermonuclease-positive (except strain 100 which was not
tested for thermonuclease). Taken together these results es-
tablish that the strains used were S. aureus cultures.
 THERMAL INACTIVATION OF STAPHYLOCOCCUS AUREUS 633

Recovery ofS. aureus on different media TABLE 1. Effect of growth phase and retentate concentration on D-
Stationary-phase cultures ( 1 % inocula) of 5. aureus values of Staphylococcus aureus 425 in whole and 11F-whole milk at
(strains 481, 425, and 196E) were thermally stressed at 56°C 58 °C.
for 10 min in preheated pasteurized milk. The heating vessel Retentate Age of Growth D-value
Trial concentration culture (h) phase (min)
was a sterile, flat-bottomed 37 X 300-mm jacketed test tube 1 IX 5.25 Early log 20 ± 3 (5) b
with thermostatically controlled water circulated through the 1 IX ON" Stationary 189 + 91 (5)
jacket. A sterile plastic cap with sampling ports covered the 2 IX ON" Stationary 61 ± 3 ( 4 )
top of the tube. Mixing was accomplished with a magnetic 2 IX 6.2 Log 2 0 + 3(5)
stirrer. After 10 min, 1 -ml samples were withdrawn and added 2 4X ON" Stationary 139 + 3 2 ( 4 )
to 9 ml of prechilled 0.1% aqueous peptone solution and then "ON refers to cultures incubated overnight 16 to 24 h.
b
were enumerated in quintuplicate on Plate Count Agar (PCA; Number of replicates.
Difco), PCA with a Mannitol Salt Agar (MSA; Difco) overlay
5. aureus 425 in the stationary phase was more heat-resistant
(after 1 h of incubation at room temperature), and MSA alone.
(e.g. protected) in the 4X whole milk retentate than in uncon-
Initial counts of inoculated but unheated milk were deter-
centrated (1X) whole milk. The differences were significant at

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022

mined in duplicate on PCA. All inoculated plates were incu-
p < 0.05 for all data in trial 2.
bated at 37°C for 2 d. In terms of the percentage of 5. aureus
colonies recovered, PCA recovered more 5. aureus 481 and
Effect of age in the stationary phase on heat resistance ofS.
196E(39±4and6.0±2.1%,respectively) than did PCA/MSA
aureus 425
(35 + 10 and 0.25 + 0.09, respectively) which recovered more
In this trial, D-values were determined at 58°C for station-
than did MSA alone (0.015 ± 0.004 and 0.0093 ± 0.0071%,
ary phase cultures of various ages (17, 25, 41, and 49-h-old)
respectively). However, PCA/MSA recovered slightly more
grown in autoclaved (121°C, 15 min) skim milk and then
5. aureus 425 (27 ± 4%) than did PCA (15 ± 8%) which
tested in whole milk. The D-values (min) were 8.6 + 1.0 (6
recovered more than did MSA alone (0.053 + 0.020%). These
replicates), 11 + 1 (6 replicates), 20 ± 1 (6 replicates), and 21
data were all significant at p<0.03 in each comparison of
+ 1 min (5 replicates), respectively, for cells of the different
recovery by the three media for each organism. Only four
ages. The results indicate a relatively stable range for D-values
replicates were done for strains 481 and 425 enumerated on
between 17 and 25 h and between 41 and 49 h of age. The 41
MSA. Thus it was decided to use PCA to enumerate 5. aureus
to 49-h period appeared to be the most constant, but clotting
in subsequent trials.
of milk was noted with cultures of that age making them
unsuitable for use as inocula. Consequently, only inocula that
Heat resistance of strains ofS. aureus
were in the 17 to 25-h range of age were used in these
Heat resistance of the four strains of 5. aureus was experiments. This result (increasing heat resistance with age
compared using whole milk at 58°C. In this study, raw whole of culture in the stationary phase) is apparently not true for all
milk was preheated to 65°C for 30 min in a water bath to reduce 5. aureus strains. Hurst et al. (9) found the D-values of 5.
the initial microbial load before the heating trials, and then was aureus in the early stationary phase were higher (32 min) than
cooled immediately in an ice bath. In whole milk at 58°C, heat in the late stationary phase (9 min) when tested at 55°C.
resistance of 5. aureus strain 425 > 100 and 481 > 196E. The
D-values were 26 + 5, 14 + 3, 13 + 3, and 3.0 + 0.6 min,
Influence of concentration by ultrafiltration on heat resis-
respectively (5 replicates). Differences between strains, where
tance ofS. aureus
they existed, were significant (p < 0.004). Results for strains
We then examined the effect of concentrating skim and
481 and 100 were not significantly different (p < 0.51). The
whole milk by ultrafiltration on heat resistance of S. aureus
data indicate that strain 425 probably was the most heat
425 at 57.5°C (Table 2). Differences in data between uncon-
resistant of the strains we examined. Consequently, we used
this strain for the remainder of the experiments.
TABLE 2. Effect of concentration by ultrafiltration on the heat
resistance of Staphylococcus aureus in skim and whole milk reten-
Growth phase and heat resistance
tates at 57.5 "C.
We wanted to use the organism when it was most heat Trial Type of milk Concentration factor D-value (min)
resistant during growth in skim milk. In one study (Table 1, 1 Skim IX 24+ 3(5) c
trial 1) 5. aureus 425 was grown in autoclaved (121 °C, 15 min) 1 Skim 2X 23+ 3 (4)
skim milk and was tested in whole milk (preheated before 1" Skim 4X 43+ 15(5)
b
experimental trial at 65°C) for its heat resistance in the early
2
Whole IX 12+ 2(7)
log phase and the stationary phase of growth. In this trial, the 2b Whole 2X 12+ 2(7)
b
2
Whole 4X 13+ 1(7)
culture in the stationary phase was more heat resistant than in a
The D-value at 58"'C in 4X skim milk retentate is significantly
the early log phase (p < 0.014). In another study (milk
different at p<0.05 from the D-values in IX and 2X skim milk
preheated to 65°C for 30 min), similar results were obtained
retentate of the same lot in this trial.
(Table 1, trial 2). These data are consistent with those of Hurst b
The D-values in these samples prepared from the same lot of whole
et al. (9) who also found that cells of S. aureus from the log milk are not significantly different at p<0.05; however, they repre-
phase were less heat resistant (D = 3.0 min) than were cells sent a trend in this trial.
from the stationary phase (D = 9-32 min) at 55°C. Furthermore, "Number of replicates.

JOURNAL OF FOOD PROTECTION. VOL. 52, SEPTEMBER 1989

634 KORNACKI AND MARTH
centrated and 2X concentrated skim milk retentates were not
significant (p > 0.05), but the difference between unconcen-
trated and 4X concentrated skim milk retentate was significant
(p < 0.05). These results are in agreement with data of
Haggerty and Potter (7) who found that heat resistance of 5.
aureus at 62.7°C was not statistically significantly different (D
= 1.71 min) when tested in skim milk, a 2X skim milk
retentate, and a 2X skim milk retentate diluted to IX with
permeate. However, they did not test the heat resistance of S.
aureus at other temperatures, higher milk concentrations, or in
concentrated or unconcentrated whole milk. They also found
similar results with E. coli and S.faecalis. In the present study,
differences in data for whole milk and retentates (2X and 4X)
of whole milk (Table 2) were not significant at p < 0.05, but
the data indicate a possible trend. The D-values in this trial in-
creased with an increase in concentration of the heating
menstruum. Consequently, 4X retentates were used in the
remainder of the experiments. Attempts were made to produce
at 6X retentate to test the heat resistance of S. aureus 425;
however, behavior of the 6X retentate precluded its use for
experiments. Unlike the other retentates, when cooled after
preheating, the 6X skim retentate gelled. When the product
was heated in a water bath it became fluid and clotted about 2
min later.
The pH range of all milks and retentates used in this study
was 6.52 to 6.73. The average percentages of total solids,
protein, and fat from representative lots of skim milk were
10.00±0.78, 3.7±0.32, and 0.26 + 0.10 (6 lots), respectively;
for 2X skim milk retentate (1 lot) they were 14.05, 7.1, and
0.46; for 4X skim milk retentated values were 18.96 + 2.89,
12.5 + 2.3, and 0.88 + 0.42, respectively; whereas for whole
milk they were 12.21 + 0.42, 3.2 + 0.39, and 3.5 ± 0.2,
respectively. A 2X whole milk retentate had 17.60% total
solids and 1.2% fat (1 lot), whereas 4X whole milk retentates
(2 lots) averaged 27.36 ± 1.84% total solids, 10.9 + 0.0%
protein, and 10.5 + 1.4% milkfat.
Heat resistance ofS. aureus 425 over a range of temperatures
Results described thus far suggest that ultrafiltration of
milk may yield a product which has a thermally protective
effect to S. aureus 425; ho we ver, when tests were done at 74°C
(milk and retentate also preheated to 65°C for 30 min) with
whole milk and a 4X retentate of the same lot of milk, results
were otherwise. S. aureus 425 in whole milk had a D-value of
0.47 + 0.24 min (5 replicates) and in the 4X concentrate it was
0.14 + 0.03 min (5 replicates). The difference between D-
values was significant at p < 0.04. Results of this experiment
highlighted the need for more trials at several different tem-
peratures. Temperatures used in subsequent studies ranged
between 57 and 79°C.
Figure 1 depicts results of such a study. As before,
pasteurized milk and UF milk were heated to 65"C and cooled
before the heating trials were done to reduce the background
aerobic plate count. Heat treatments were at 57 or 58, 65, 72,
and 79°C for 15 min, 1 min, 5 s, and 5 s, respectively. Average
D-values are given in Table 3 (trial 1). No significant differ-
ences at p < 0.05 were noted between D-values at 65 and 79°C
(the only two temperatures at which S. aureus 425 was tested
JOURNAL OF FOOD PROTFCTION. VOL. 52. SEPTEMBER 1989
DD 80 SB TO TD 80
Temperature (°C)
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022
Figure \.LogIB of D-values of Staphylococcus aureus 425 in whole
milk (—+) and a 4X whole milk retentate (—3) of the same lot vs.
temperature (trial 1).
in both whole and 4X whole milk retentate). The z-values for
S. aureus 425 in whole milk and in the 4X retentate of the same
lot of milk were 6.0 and 7.2°C, respectively; however, lines
generated from a plot of log10 D-values vs. temperature were
not significantly different from one another (p > 0.05). Hence,
z-values were not significantly different from one another.
This experiment was repeated with another lot of whole milk
(Fig. 2: Table 3, trial 2). In this trial, whole milk appeared to
be more thermally protective than the 4X concentrate at 5 8 and
63°C (p < 0.00001), but at 68 and 73°C there was no significant
difference (p > 0.05), and at 78°C the trend appeared to be
reversed with the 4X retentate providing more thermal protec-
tion (p < 0.00001) than the unconcentrated milk. The z-values
for S. aureus 425 in whole milk and in the 4X retentate of the
same lot of milk were 7.6 and 9.2°C, respectively. Regression
lines obtained from a plot of the log1() of D-values vs. tempera-
ture were significantly different (p < 0.05). These results
suggest that under conditions of these tests 4X ultrafiltered
whole milk retentates may have afforded slight thermal pro-
tection to S. aureus 425 at 78°C. Calculated z-values from data
summarized by Mitscherlich and Marth (13) for S. aureus
strains 161-C, S-l, B-120, and S-18 tested in reconstituted
whole milk were 2.88, 7.35, 6.62, and 5.43°C, respectively.
When compared to these strains, S. aureus 425 appears to
exhibit equivalent heat resistance.
Heat resistance of S. aureus 425 over a range of tempera-
tures also was determined using skim milk and a 4X retentate
of the same lot of skim milk (Table 4, trial 1). In trial 1, D-
values for S. aureus 425 at 63, 68, 72, and 77°C were not
significantly different (p > 0.05) when the organisms were
tested in skim milk and a 4X skim milk retentate made from the
same lot of skim milk. The z-values for S. aureus 425 in skim
milk and in the 4X skim milk retentate were 13 and 11°C,
respectively. The regression lines obtained from a plot of the
log „ of S. aureus 425 D-values determined for skim milk and
a 4X skim milk retentate vs. temperature were not signifi-
cantly different (p > 0.05). Hence, the z-values for S. aureus
425 in skim and 4X skim milk retentate in this trial were not
significantly different from each other. This experiment was
 THERMAL INACTIVATION OF STAPHYLOCOCCUS AUREUS 635

TABLE 3. D-values of Staphylococcus aureus 425 obtained using whole milk and 4X whole milk retentates.
Average
D-value + Std. dev. (min)
Temperature Whole 4X Whole
Trial (°C) milk milk retentate
57 77.00 ±21.000(5)
58 174.00 ±95.000(5)"
65 0.69 ± 0.09 (5) 0.50 ± 0 . 1 1 (3)
72 0.30 ± 0 . 1 1 (6)
79 0.033 ± 0.013(4) 0.051 ± 0.017(5)

58 3.000 ± 0.000 (7) 8.000 ± 0.3 (7)

63 0.85 + 0.16 (6) 0.53 ± 0 . 1 8 (5)
68 0.54 + 0.19 (6) 0.30 ± 0 . 1 1 (7)
73 0.065 ± 0.006 (6) 0.065 ± 0.009(6)
78 0.030 ± 0.008 (6) 0.054 ± 0.009(6)

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022

Standard deviation.
b
Number of replicates.

repeated (Table 4, trial 2) with another lot of skim milk which

Temperature (°CJ
Figure 2. Logw of D-values ^/'Staphylococcus aureus 425 in whole
milk <--- + ) and a 4X whole milk retentate (—•) of the same lot vs.
temperature (trial 2).
had been preacidified to pH 6.3 with acetic acid and concen-
trated to 3.5X with another ultrafiltration system (Romicon,
model HF4SSS, Portable, Cummings Park, Woburn, MA)
equipped with a spiral wound membrane (HF 26.5-43-PM50)
(50,000 molecular weight cut off). Heat destruction of S.
aureus 425 was assessed at 62, 68, 72, and 78°C. D-values of
S. aureus 425 were not significantly different when skim milk
and 4X skim milk retentate were the heating menstrua at any
of the temperatures tested (p > 0.05). The z-values were 9.7
and 10°C, respectively. However, these z-values are not sig-
nificantly different since the lines generated from a plot of the
log]() D-values vs. temperature were not significantly different
(p > 0.05). The z-value for a strain of S. aureus in skim milk,
calculated from data obtained by Walker and Harmon (22),
was 7.2°C. Our strain was apparently more heat resistant than
this strain. Zottola and Jezeski (24) found that S. aureus 196E
and F265 had z-values of 3.3°C and 4.4°C, respectively, in

TABLE 4. D-values of Staphylococcus aureus 425 obtained using skim milk, 3.5X skim milk retentate and 4X skim milk retentates.
Average
D-value + Std. dev.' (min)
Tcmperaiure Skim Skim
Trial I^C] milk milk retentate
lt 63 0.50 ±0.09 (4)c 0.63 ±0.14 (6)
68 0.34 ±0.17 (7) 0.27 ±0.03 (6)
72 0.057 ± 0.032 (7) 0.047 ±0.011 (7)
77 0.059 ±0.013 (7) 0.049 ±0.013 (6)

62 1.16 ±0.51 (5) 0.94 ±0.03 (3)

67 0.39 ±0.10 (4) 0.35 ±0.13 (6)
72 0.071 ±0.018(7) 0.066 ± 0.027 (7)
78 0.028 ±0.008 (7) 0.030 ± 0.005 (6)
"Standard deviation.
"D-values in skim milk and 4X skim milk retentate from the same lot over a range of temperatures.
'Number of replicates.
d
D-values in skim milk and a 3.5X skim milk retentate from the same lot over a range of temperatures.

JOURNAL OF FOOD PROTECTION. VOL. 52, SEPTEMBER 1989

636 KORNACKI AND MARTH

skim milk reconstituted to 10% total solids. They also found
that these organisms had the same z-values when tested in raw
whole milk as in 10% reconstituted skim milk. S. aureus 425
appeared to be more heat resistant over a range of tempera-
tures in skim milk (z = 13, and 9.7°C for S. aureus 425 in skim
milk; Table 4, trials 1 and 2) than in whole milk (z = 6.0, and
7.6 for S. aureus 425 in whole milk: Table 3, trials 1 and 2).
Effect of protein,fat, and lactose on the heat resistance ofS.
aureus 425
A 4 X 2 3 'experimental design was used to study the
influence of two levels each of protein, fat, and lactose on
thermal destruction of S. aureus 425 (Table 5). Protein was
varied between 3.5-4.0% and 12.6-16.0% for the lower (minus)
those of Kadan (10) who found that addition of up to 14% fat
to skim milk did not influence the heat resistance of S. aureus.
Similarly, Fay (3) found that 50% lactose in broth did not
thermally protect E. coli heated therein as compared to heating
the cells in broth without lactose.
Effect of diafiltration on heat resistance ofS. aureus 425
Diafiltration is a practice which can be combined with
ultrafiltration of milk. The effect of diafiltration on heat
resistance of S. aureus 425 was investigated. In this experi-
ment, skim milk was ultrafiltered to a 4X concentration. A
volume of distilled deionized water was added to the retentate
equivalent to the volume of permeate which had been re-
moved. The liquid was again untrafiltered to a 4X concentra-

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022

and upper (plus) levels (Table 5), respectively. Lactose was tion. During diafiltration one essentially washes away most of
varied between about 4.5-5.0% and about 14.5-15%. Fluids the low molecular weight solutes while retaining macromole-
used were skim milk, a 4X UF retentate of skim milk, and both cules of higher molecular weight. In this trial, heat resistance
fortified with 4 1 % cream, and/or lactose (Table 5). Fluids of of S. aureus 425 was tested at 62 and 77°C. S. aureus was
all eight combinations of the three constituents under study (at significantly ( p < 0.01) less heat resistant in 4X diafiltered
two levels) were prepared and tests were done in triplicate or skim milk (D = 0.92 + 0.31 min; 7 replicates) retentate than in
quadruplicate. None of the three constituents, or interactions skim milk from the same lot (D = 2.0 ± 0 . 8 min; 7 replicates
between them, significantly affected (p > 0.05) heat resistance when tested at 62°C. At 77°C a similar trend was noted. 5.
of 5. aureus 425 at 72.5°C/10 s (Table 5). Moats et al. (14) aureus 425 in 4X diafiltered skim milk was less heat resistant
found that casein added to phosphate buffer at low level (1 %) (D = 0.049 + 0.011 min; 7 replicates) than in skim milk from
was no more thermally protective to Salmonella anatum than the same lot (3 = 0.066 ± 0.023 min; 7 replicates). The data
was plain phosphate buffer. Our results are in agreement with were only significant at p < 0.12; however, the high p value
may have resulted from decreased precision of the readings at
more severe temperatures (heated for only a few seconds).
Calculated z-values for S. aureus are 10 and 12°C when skim
milk and the 4X batch diafiltered retentate, respectively, were
TABLE 5. Effect of protein11, fat1', lactose' on D-values (min) of used. Diafiltration of milk served to lower the heat resistance
Staphylococcus aureus 425 at 72.5 °C: a 4X23 • design. of 5. aureus in the retentate as compared to the unconcentrated
Fluid Protein Fat Lactose D-value (min)d non-diafiltered milk. The data also suggest that lower molecu-
(%) (%) (ca.%)
Skim 4.0 0.28 5.0 0.072 ± 0.027 (4) lar weight solutes may play an important role in the thermal
4X Skim 0.075 ±0.023 (4) protection of S. aureus in fluid dairy products. In somewhat
Skim plus 16.0 1.12 5.0 comparable work, Haggerty and Potter (7) found that the heat
40% cream 0.083 ±0.032 (4) resistance of S. aureus in a 2X skim milk retentate diluted to
4X Skim plus 3.5 10.0 4.5 IX with water was significantly lower than in 2X skim milk
40% cream 0.076 ± 0.030 (4) retentate, a 2X skim milk retentate diluted to IX with perme-
Skim plus 12.6 10.0 4.5
ate, and unconcentrated skim milk. These data are consistent
lactose 0.082 + 0.013(3) with a subsequent experiment in which skim milk, permeate
4X Skim plus 4.0 0.28 15.0
from 4X ultrafiltration of the same lot of skim milk, and
lactose 0.051 ±0.006(3)
Skim plus 40% 16.0 1.12 15.0 diafiltrate (permeate) from 4X UF concentration (diafiltra-
cream, plus tion) of a 4X skim milk retentate (same lot) diluted to a IX
lactose 3.5 10.0 14.5 0.068 ±0.010 (3) volume level with distilled deionized water were compared
4X Skim plus for their influence on the heat resistance of S. aureus 425 at
40% cream, 63°C. S. aureus 425 had greater D-values (min) when tested
plus lactose 12.6 10.0 14.5 0.088 ±0.007 (3) in skim milk (0.36 ± 0.05; 7 replicates) than in permeate from
Skim milk had 4.0% protein (measured), 0.28% fat, and ca. 5% ultrafiltration of the same lot of skim milk (0.30 + 0.14; 7 rep-
lactose (assumed). Undiluted 4X retentate of the same lot had 16.0% licates), than in the diafiltrate from UF of the same lot of skim
protein, 1.12% fat, and ca. 5% lactose (assumed). milk (0.28 + 0.06; 7 replicates). However, the data are only
"Protein levels were 3.5-4.0% and 16.0% for the low and high levels, significant (p < 0.05) for the difference in results obtained
respectively.
b using skim milk and diafiltrate (p < 0.016). The diafiltrate
Fat levels were 0.28-1.12% and 10% for the low and high levels,
should have the least solids and the least solids and the least
respectively.
c
Lactose levels were ca. 4.5-5%, and ca. 14.5-15% for the low and amount of low molecular weight solutes (LMS). Permeate
high levels, respectively. should have more solids than the diafiltrate and slightly less
d LMS than skim milk, and skim milk should have the highest
D-value in minutes + standard deviation. Numbers in parenthesis
are the number of replicates. solids content and more LMS than the other solutions. Our

JOURNAL OF FOOD PROTECTION. VOL. 52. SEPTEMBER 1989

 THERMAL INACTIVATION OF STAPHYLOCOCCUS AUREUS
data suggest that some (lactose appears not to influence the
heat resistance of 5. aureus 425) LMS in milk have a much
greater influence on the heat resistance of 5. aureus 425 than
higher molecular weight constituents such as protein, and fat.
Numerous low molecular weight solutes have been shown to
influence the heat resistance of microorganisms (3,11,21).
Mg++, Mn++, Ca++, and Co++ increased the heat resistance of
Enterobacter aerogenes (28), whereas K+, Na+, glucose, ri-
bose, lactose, and EDTA had the opposite effect on heat
resistance of this organism. Lee and Goepfert(71) showed that
increasing concentrations of phosphate buffer decreased the
heat resistance of Salmonella typhimurium when compared to
heating the bacterium in water. However, they (11) found that
Mg++, spermine, and sucrose increased its heat resistance. Fay
(3) showed that 2 molal solutions of glucose or sucrose were
not thermally protective to Staphylococcus albus, but a 2
molal sucrose solution was thermally protective for S. aureus
and P seudomonas fluorescens.
Milk contains a number of dissolved salts which could
influence the heat resistance of S. aureus or other microorgan-
isms. Included are the chlorides, phosphates, and citrates of
potassium, sodium, calcium, and magnesium (23). Potassium,
sodium, and chloride are considered to be entirely in solution
and presumably completely ionized in milk (23). Phosphates,
calcium, magnesium, and citric acid are partially in solution
and partially in suspended combinations in milk (23). Loss of
heat resistance by 5. aureus in diafiltered milk may be a
consequence of the loss of thermal protection afforded by one
or more of these dissolved salts.
ACKNOWLEDGMENTS
We thank PaulG. Morin for his technical assistance. This research was
supported by the College of Agricultural and Life Sciences, University of
Wisconsin-Madison, and by the Dairy Research Foundation of the National
Dairy Council, Rosemont. Illinois.
REFERENCES
1. Bacr, E. F., R. J. H. Gray, and D. S. Orth. 1976. Methods for the isolation
and enumeration of Staphylococcus aureus, pp. 374-385. InM.L. Speck
(cd.), Compendium of methods for the microbiological examination of
foods. American Public Health Association, Washington, DC.
2. Dega.C. A., J. M.Goepfert.andC. H. Amundson. 1972. Heat resistance
of salmonellae in concentrated milk. Appl. Microbiol. 23:415-420.
3. Fay, A. C. 1934. The effect of hypertonic sugar solutions on the thermal
resistance of bacteria. J. Agr. Res. 48:453-468.
JOURNAL OF FOOD PROTECTION. VOL. 52. SEPTEMBER 1989
637
4. Fischer, R., and A. Wellinger. 1981. Determination of heat losses in
biogas installations. Int. J. Ambient Energy 2:159-171.
5. Grieme, L. E., and D. M. Barbano. 1983. Method for use of a differential
scanning calorimeter for determination of bacterial thermal death times.
J. Food Prot. 46:797-801.
6. Groschc, C. A., H. L. Lucas, and M. L. Speck. 1952. Pasteurization
requirements for concentrated whole milk. J. Dairy Sci. 35:793-799.
7. Haggerty. P., and N. N. Potter. 1986. Growth and death of selected
microorganisms in ultrafiltered milk. J. Food Prot. 49:233-235.
8. Horwitz, W. (ed). 1980. Official methods of analysis, 13th ed. Associa-
tion of Official Analytical Chemists, Washington, DC.
9. Hurst, A., A. Hughes, andD. L. Collins-Thompson. 1974. The effect of
sublethal heating on Staphylococcus aureus at different physiological
ages. Can. J. Microbiol. 20:765-768.
10. Kadan, R. S., W. H. Marten, and R. Mickclson. 1963. Effects of ingre-
dients used in condensed and frozen dairy products on the thermal resis-
tance of potentially pathogenic staphylococci. Appl. Microbiol. 11:45-
49.
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/52/9/631/1658011/0362-028x-52_9_631.pdf by Mexico user on 03 June 2022
11. Lee, A. C , and J. M. Goepfert. 1975. Influence of selected solutes on
thermally induced death and injury of Salmonella typhimurium. J. Milk
Food Technol. 38:195-200.
12. Marth.E. H. (ed.). 1978. Standard methods for the examination of dairy
products, 14th cd. American Public Health Association, Washington,
DC.
13. Mitscherlich, E„ and E. H. Marth. 1984. Microbial survival in the
environment: bacteria and rickeitsiae important in human and animal
health. Springer- Verlag, Berlin.
14. Moats, W. A., R. Dabbah, and V. M. Edwards. 1971. Survival of Salmo-
nella anatum heated in various media. Appl. Microbiol. 21:476-481.
15. Perez, B. S., P. L. Lorenzo, M. L. Garcia, P. E. Hernandez, and J. A.
Ordonez. 1982. Heat resistance of enterococci. Milchwissenschaft 37:724-
726.
16. SAS Institute Inc. 1985. SAS user's guide: Statistics, version 5 edition.
Cary, NC.
17. Singh, R. S., andB. Ranganathan. 1980. Heat resistance of Escherichia
coli in cow and buffalo milk. J. Food Prot. 43:376-380.
18. Slack, A. W.,C. H. Amundson, C. G. Hill, Jr., and N. A. Jorgcnsen. 1982.
On-farm ultrafiltration of milk. 1. Technical feasibility studies. Proc.
Biochcm. 17:6-11.
19. Slack. A. W„ C. H. Amundson, and C. G. Hill, Jr. 1982. On-farm
ultrafiltration of milk. 2. Economic analysis. Proc. Biochcm. 18:23-25;
30.
20. Speck, M. L., and H. L. Lucas. 1951. Some observations on the high-
temperature short-time pasteurization of chocolate milk. J. Dairy Sci.
34:333-341.
21. Strange, R. E., and M. Shon. 1964. Effects of thermal stress on the
viability and ribonucleic acid of Aerobacter aerogenes in aqueous sus-
pension. J. Gen. Microbiol. 43:99-114.
22. Walker, G. C , and L. G. Harmon. 1966. Thermal resistance of Staphy-
lococcus aureus in milk, whey, and phosphate buffer. Appl. Microbiol.
14:584-590.
23. Wong, N. P., R. Jenness, M. Keency, and E. H. Marth (eds.). 1988.
Fundamentals of dairy chemistry, 3rd cd. Van Nostrand Rcinhold Co.,
New York.
24. Zottola, E. A., and J. J. Jezeski. 1969. Comparisons of short-time holding
procedures to determine thermal resistance of Staphylococcus aureus. J.
Dairy Sci. 52:1855-1857.