Transcript Andrew K. Benson, PhD

Science-based policy for the food industry:
Embracing the BEST Science that’s out there
Andrew K. Benson
Professor, Department of Food Science
Director, Printed Microarray Core Facility
University of Nebraska
Powered by
Microbes exist in complex ecosystems
within different food matrices
Uncontaminated sample
Contaminated sample
Autochonous
Allocthonous
Indicator organism concept
1. Easy to culture
2. Present in all mammalian feces
3. Durable in the environment
4. Associated with presence of pathogens
*E. coli is the most common indicator used
Problems associated with use of indicators
1. Relatively low abundance in feces
a. E. coli present at 106/gram
b. Others (e.g. Bacteriodes, Clostridia) present at 1013/gram
2. Some species are present in environment (problem with coliforms)
3. Correlation with pathogens questionable:
a. Instances of “indicator free” foods causing outbreaks
b. Poor correlation of indicators with actual pathogens
Deep sequencing surveys confirm that
E. coli is a minor member
Of the human fecal microbiota
Taxonomic group to which
E. coli belongs
Eckburg et al. Science. 2005. 308:1635-1637
Do the math:
1,000 Kg production lot contaminated with 1g of feces
Assumptions:
E. coli present at 106 per gram of feces
Bacteroides present at 1013 per gram
If the contamination is homogenously distributed,
each gram of the 1,000,000 g (1000 Kg) sample would contain:
1 E. coli
1,000,000 Bacteroides
How many grams must be sampled to ensure detection of E. coli?
at least 10 grams, 100 grams would be better
How many grams must be sampled to ensure detection of Bacteriodes?
10-100 mg
Culture-based methods count only the
organisms that can be grown in vitro…
And we can culture 10% (at best) or species
From any environment
Non-culture-based methods that allow us to
survey composition of the community
Extract nucleic acids
Specific PCR tests for individual species
1. Effective for specific pathogens
2. Back to the same problem for indicators
Community profiling
1. Use of 16S rDNA as a target
2. Profiles composition of entire community
3. Able to identify allocthonous species
4. Does the ecosystem contain fecal flora?
Massively parallel Pyrosequencing:
The next generation of DNA sequencing
Extract nucleic acids
PCR amplify 16S rDNA
from all bacteria
Attach to
microspheres
Deposit beads into picotitre plates
PCR amplify on beads
Picotitre plate—single bead per well, ~1 million wells
Parallel sequencing of all 1 million wells
***At least 400,000 individual sequences…this number is rapidly growing
Will be 1 Million reads per run by 2009
Samples
1
2
3
4
.
.
.
.
Looking at the output
Species
0
10
100
1000
Relative
abundance
Uncontaminated samples contaminated samples
Allocthonous species or fecal species
Indicator organism or community profile?
It’s time we step up to where science is headed