Metagenomics in food safety: What’s the added
value? Case studies from the livestock sector in
Tanzania and Uganda
Silvia Alonso, Kristina Roesel, Stephen Opiyo, Francesca Stomeo, Delia Grace
FAO Regional Meeting on Agricultural Biotechnologies in Sustainable Food Systems and
Nutrition in Sub-Saharan Africa
Addis Ababa, Ethiopia, 22-24 November 2017

Main campuses: Nairobi, Kenya
and Addis Ababa, Ethiopia
Offices in 14 other countries
ILRI offices

ILRI resources 2017
• Staff: 670+
• Budget: about US$80 million
• Senior scientists from 39 countries
• One third of ILRI staff are women
• Main campuses in Kenya and Ethiopia,
and offices in 14 other countries around
the world

CIMMYT
Mexico City
Mexico
IFPRI
Wash. DC
USA
CIP
Lima
Peru
CIAT
Cali
Colombia
Bioversity
International
Rome Italy
AfricaRice
Cotonou
Benin
IITA
Ibadan
Nigeria
ILRI
Nairobi
Kenya
Addis
Ababa,
Ethiopia
World
Agroforestry
Nairobi
Kenya
ICARDA
Beirut
Lebanon ICRISAT
Patancheru
India
IWMI
Colombo
Sri Lanka
IRRI
Los Banos
Phillippines
WorldFish
Penang
Malaysia
CIFOR
Bogor
Indonesia
CGIAR Research Centres

Safe Food, Fair Food project
2010-2015 (2 phases)
• Assessing health risks associated
with animal source foods in low
income countries
• Identification and evaluation of risk
management interventions”

Risk assessment and biotechnologies
1
Hazard identification
Hazard characterization Exposure assessment
Risk characterization
Risk communication
What harm does it cause?
How does harm depend on
dose?
Can it be present in food? Can
it cause harm?
How and to what extent does it get
from source to victim?
What is the harm?
What is its likelihood?
Participatory methods
fit well
Risk assessment framework:
Codex Alimentarius Commission

METAGENOMICS
“The study of communities of known and unknown microbial
organisms (bacteria, viruses, parasites) directly in their natural
environments, through the study of their entire genetic
information, bypassing the need for isolation and lab cultivation
of individual species”.
Aims of metagenomics in SFFF project:
• to develop a catalogue of microbe species/groups present in
different food value chains
• In terms of sensitivity, can genomic approaches complement or
even substitute conventional microbiology in food safety hazard
identification?
• How applicable and economically feasible are these new
approaches in resource-poor settings of low-income settings.

Food safety in milk
Milk from pastoralist and smallholder farms
in Tanzania
• Household milk
• Individual cow’s milk

Food safety assessment - MILK
Conventional microbiology
Hygiene indicators
Total bacterial count
Enterobacteriaceae count
Pathogens
Listeria spp.
Salmonella spp.
CP Staphylococcus spp.
Mastitis pathogens
Staphylococcus spp.
Escherichia coli
Food safety hygiene
standards
50% samples Listeria
0% samples Salmonella
All samples carriers
35% S. aureus
13% E.coli
51% S. epidermidis

Food safety assessment - MILK
16S Metagenomics*
* Caporaso et al (2012). Performed using the Illumina MiSeq Platform
0
2
4
6
8
10
12
14
16
18
20
Relative abundance of bacteria (Genus) in household milk from
Morogoro, Tanzania
Household milk
Alonso et al, upcoming

Food safety assessment - MILK
16S Metagenomics*
* Caporaso et al (2012). Performed using the Illumina MiSeq Platform
0
2
4
6
8
10
12
14
16
18
20
Relative abundance of bacteria (Genus) in household milk from
Morogoro, Tanzania
Household milk0
5
10
15
20
25
30
35
40
Lactococcus
Weissella
Sphingomonas
Acinetobacter
Enterobacter
Unclassified
Others
Staphylococcus
Macrococcus
Streptococcus
Pseudomonas
Leuconostoc
Erwinia
Plesiomonas
Vagococcus
Serratia
Enhydrobacter
Microvirus
Tolumonas
Kurthia
Viridibacillus
Clostridium
Trabulsiella
Methylosinus
Citrobacter
Bacillus
Kocuria
Corynebacterium
Curtobacterium
Streptomyces
Pseudomonas
Enterococcus
Calothrix
Bradyrhizobium
Cupriavidus
Klebsiella
CandidatusBlochmannia
Thermogemmatispora
Escherichia
Alkaliphilus
Aerococcus
Turicibacter
Yersinia
Lactobacillus
Rhodoferax
Relative abundance of Bacteria (Genus) in milk from
Morogoro, Tanzania
Household milk Cow's milk
Alonso et al, upcoming

Food safety - PORK
– Highest per capita consumption in EAC
(3.4 kg)
– Explosion in pig numbers over the past
30 years (0.19 to 3.2 million pigs)
– Mostly in hands of smallholders
– “piggy bank”
– 70% consumed in urban areas
– “pork joint” phenomenon

Food safety assessment – Pork
• 88 pork samples (fresh and processed)
from 31 butcheries (Kamuli & Mukono
districts)
• Total aerobic counts
• Metagenomics - 16 samples excluded
(poor DNA quality)
• Illumina MiSeq libraries preparation
and sequencing
• Denovo assembly and blast queries

Food safety assessment – Pork
Total bacterial count: All fresh pork and ca. 50% of the processed pork:
above national food safety standards
Metagenomics:
Ca. 300 bacteria genus including
 pig pathogens (i.e. Brachyspira spp., Haemophilus spp., Mycoplasma spp.)
 potential foodborne zoonoses (i.e. Bacillus cereus, Campylobacter coli,
Clostridium botulinum, Escherichia coli, Yersinia enterocolitica, Staphylococcus
aureus etc.)
 occupational zoonoses (Erysipelothrix spp., Streptococcus suis)
 and anthropozoonoses (Mycobacterium tuberculosis)
Blast queries showed no hits for parasitic and viral diseases
(Descriptive and spatial analysis still ongoing)
Roesel et al, upcoming

Main messages
• Comprehensive overview of microbial flora in food
• As screening process on food safety (finds more, but
may miss some…)
• Could aid pathogen discovery
• Lots of info - enormous “untapped” potential
• Substitute for conventional microbiology?
• Does not speak about “viable” cells
• May not detect microbes in small amounts (Salmonella, Listeria)
• Rather a “complement”

Main messages
• Practical aspects:
– Not necessarily faster, but could be if equipment
and expertise are in place
– Overall costs (long run) may be similar or even
smaller
– Requires “highly-specialized” expertise for
processing and data analysis

This presentation is licensed for use under the Creative Commons Attribution 4.0 International Licence.
better lives through livestock
ilri.org