Transcript Spores in Cancun

Monitoring Endospores and Endospore-Forming Bacterial Populations
During Commercial Skim Milk Powder Production
C. Murillo, C. Kitts, and R. Jiménez-Flores
EBI and DPTC
California Polytechnic State University.
Project Goals
Abstract
 Count spore-formers, and viable bacteria during commercial, low-heat
skim milk powder production throughout production runs at the
following points:
 Bacillus spp. were clearly indicated.
– Milk storage silo
– Most common during skimming, condensing, and drying;
low in raw milk.
– Cream separator
• B. subtilis group were found in samples at all plants and
– Evaporator (final effect)
during all visits.
– Powder port
• B. cereus group were found in samples at all plants at
 Follow the abundance of the predominant bacteria and spore formers
least some of the time.
using Terminal Restriction Fragment (TRF) patterns
 Three types of rumen bacteria (e.g. Prevotella) were found
along with 8 different Proteobacteria types.
– Rumen bacteria and Proteobacteria were prominent in raw
Milk Powder Processing Flow
milk and decreased with processing.
Raw Milk
Skim Storage
Separation
Pre-warm
Introduction
4°C
 Milk powder is an important commodity on a national and international
level and a key ingredient in many food products.
 Confectionaries
 Infant formulas
 Baked goods
7°C
Over Time and Averaged Across all Plants and Visits
Sample
Pre-heating 2
Pre-heating 1
Condensing (MVR &/or TVR)
108
76°C
50°C
71°C
Spray Drying
54°C
Sample
Sifting
230°C
Space
Raw
Skim Condensed
Powder
(72°C)
0-2 hrs
X
X
X
X
8-10 hrs
X
X
X
X
15-19 hrs
X
X
X
X
23-27 hrs
X
X
X
X
30-35 hrs
X
X
X
X
40-42 hrs
X
X
X
X
16S rDNA Digested w/
HaeIII or DpnII
40 – 42
Con1
Raw3
-5*10^5
Sep5
Con3
Sep4
0.0
Dry1
Con4
0.2
+
0
Sep3
D232
-0.2
0.0
Raw4
0.2
0.4
e.g. Plant 1
Relative Percentage
70
60
H272
50
H273
40
D265
30
D267
20
10
0
S
C
D
1B
2A
2B
Plant and Visit
3A
3B
 Possible fecal contamination was indicated by the presence of
rumen bacteria, e.g. Prevotella spp. found in all plants.
– Levels of these organisms were sensitive to pasteurization.
S
C
D
S
C
D
S
C
 Possible soil contamination was indicated by the presence of
Proteobacteria.
–
Bacillus subtilis group (Hae272 & Dpn265) and
Bacillus cereus group(Hae273 & Dpn267)
R
104
– Equipment fouled despite cleaner incoming raw milk,
possibly by a biofilm in or near the separator.
Hae166 & Dpn265
Proteobacteria
And using Percent of Total Bacteria
1x Buffer (EDTA)
Dry
Principal Component 1
Hae272 & Dpn265
Bacillus subtilis group
15 kV
Sep2
Raw1
-0.4
105
Conclusions – Other Bacteria
Sep5
Con2
Con1
0.4
Condensed
1A
Raw3
Sep4
Sep1
Raw4
-0.2
Dry5
Dry6
D265
106
D238
Dry2
D264
-0.4
Sep2
Raw1
Sep3 H166
Sep1
Dry3
Dry4
Sep6
Skim
102
-5*10^5
0
Dry1
Dry2
Dry4
H272
Dry5
Dry6
-0.2
0.0
H225
Sep6
Raw
107
103
5*10^5
Raw6
Raw5
0.2
H271
Con4
0
DpnII
Principal Component 1
D
C
D
D
40 – 42 hrs
Processing Point and Time
S
28 -35 hrs
23-27 hrs
15 -19 hrs
8 -10 hrs
0-2 hrs
Sample 2
28 – 35
108
Raw2
5*10^5
0.2
Raw2
-0.4
Bacillus spp.
23 – 27
5*10^5
Raw5
0.0
HaeIII
0.4
Raw6
-0.4
POP6 Separation Polymer in 61cm x
50m Capillary @ 60oC
Prevotella spp.
15 – 19
from Each Plant and Visit, Averaged Over Processing Time
-5*10^5
Dry3
Principal Component 2
9.9 mW Laser
Detection
5*10^5
Principal Component 2
0.4
ABI 310 Genetic
Analyzer
0
-0.2
-5*10^5
Sample 1
8 - 10
Prevotella spp. at Each Processing Step
Hae225 & Dpn238
Prevotella ruminicola
Hae272 & Dpn265
Bacillus cereus group
Fragment size (nucleotides)
0–2
 Even as the raw milk became cleaner toward the end of a run,
contamination levels in the powder did not revert to their initial
level.
e.g. Plant 2
DNA Fragments
104
– As early as 8 hours for Plant 1.
Organism ID – Peak Matching Across
Enzyme Digests using PCA
−
105
 Peaks in separator spore levels generally occurred around 1519 hours of processing.
3 processing plants
 2 visits per plant
 20 - 24 samples per visit
5’
Fragment Separation and
Detection by CGE
Dry
 Critical increases in spore levels at the separator appeared to
result in spore increases in the downstream powder .
Sampling
Strategy
CFU/g total solids
(46.5°C)
Condensed
 Bacillus spp. appeared to thrive in the separator or very near it.
Time
5’
(94°C)
106
Conclusions – Spore formers

16S rDNA PCR- Amplified
w/ Fluorescent Primer
Skim
Time Into Process (hrs)
Bacterial DNA
Extracted
Extended
107
Sample
Condensed Powder
Annealed
Raw
103
Making Terminal Restriction Fragments
Denatured
109
Condense
Storage
Pasteurization
68°C
58°C
 Critical points for the proliferation of Bacillus spp. and their
relationship to the microbial population throughout milk powder
processing is required for improvements in milk powder quality.
Fluorescence Intensity
Bacillus spp. at Each Processing Step
7°C
Sample
 Bacillus spp. are of particular interest because they are the predominant
organisms that determine the keeping quality of pasteurized milk and
milk products.
Skim
50°C
54°C
 Endospores and thermoduric bacteria survive pasteurization and spray
drying to become a concern for milk powder producers.
 Endospores inhabit powder in “dormant” state, becoming active
upon reconstitution.
 Can cause proteolysis, lipolysis, sensory defects, and produce
toxins.
Raw
Cooling
CFU/g total solids
The microflora of milk powder consists of a wide array of microorganisms of which special attention is given to
Bacillus endospores given their role in defining commercial quality standards. Bacillus endospores survive
pasteurization and spray drying and remain dormant in the powder. Once the powder is reconstituted, endospores
germinate unpredictably, and through their enzymatic activity become detrimental to product quality. The objectives of
this study were to 1) enumerate mesophilic and thermophilic endospore populations during commercial, skim milk
powder production, and 2) characterize the microbial ecology of this process using Terminal Restriction Fragment
(TRF) patterns of 16S rRNA genes, and 3) compare the changes in bacterial populations during processing of skim
milk powder. Our approach was to observe changes bacterial populations during commercial operations, following the
same lot of raw milk throughout the processing run. Fluid, condensed and powdered skim milk samples were
collected periodically from three commercial milk powder facilities for a period of two years. Sampling points
included the raw milk silo, separator, evaporator, and spray dryer. Microbial evaluation was normalized based on total
solids. Every sample was evaluated for total aerobic plate count and mesophilic and thermophilic endospore counts.
For TRF patterns, community DNA was extracted, amplified by PCR using 16S rRNA probes, and digested with an
endonuclease, either HaeIII or DpnII. Endospore formers were predominant in condensed and powdered milk, and
tended to increase in the powder with increasing processing time. In raw milk mesophilic and thermophilic endospores
ranged from < 25CFU/g to 70 CFU/g and < 25CFU/g to 100 CFU/g, respectively. In powder they ranged from <
25CFU/g to 103 CFU/g and < 25CFU/g to 105 CFU/g, respectively. Both types of endospore counts from skim milk
showed an increasing trend with run time and eventually rendered the powder out of the 103 CFU/g limit. There was a
strong correlation of spore counts with season in all samples. In commercial samples TRF patterns successfully
described microbial populations and a drastic change was observed between raw and powder milk for most runs. A
particular case is presented for the illustration of the usefulness of this molecular technique in following a single
species of microorganism through a milk process.
Microorganism Types Identified and
Followed with TRF Data
Levels of these organisms responded in the same way to
pasteurization as the Prevotella spp. (data not shown) and
they also appeared to accumulate in or near the separator.
Sponsors: