ILSI-IFBiC Task Force 10: Mammalian Toxicology

Download Report

Transcript ILSI-IFBiC Task Force 10: Mammalian Toxicology

ILSI-IFBiC Task Force 10:
Mammalian Toxicology
Consensus Views on Protein
Safety Assessment
Bruce Hammond, PhD, DABT
Monsanto Company
Society of Toxicology
Workshop Session: Risk Assessment for Proteins
Introduced into Genetically Modified Crops
March 8, 2011
ILSI® International Food Biotechnology Committee
Task Force 10 Background
In 2008, the International Life Sciences Institute Food
Biotechnology Committee (ILSI-IFBiC) formed a new
task force. Its focus was to develop international
consensus recommendations when it is appropriate to
undertake animal tox studies and how to design and use
animal tox studies in the safety evaluation of biotech
crops.
This task force also addressed issues relating to the food
safety assessment of proteins introduced into biotech
crops to impart desired traits.
ILSI® International Food Biotechnology Committee
2
Task Force 10 Membership
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Sue Barlow – Independent consultant in toxicology
Andrew Bartholomaeus – General Manager Risk Assessment, Food Standards Australia
New Zealand
Genevieve Bondy – Head, Genotoxicity and Carcinogenesis Section, Food Directorate
Health Products and Food Branch, Health Canada
Amechi Chukwudebe – BASF
Bryan Delaney – Pioneer, a Dupont Company
Bruce Hammond – Monsanto Company
Corinne Herouet-Guicheney – Bayer CropScience
Joseph Jez – Department of Biology, Washington University, St Louis, Missouri
Daland Juberg – Dow AgroSciences
Hideaki Karaki – Retired (University of Tokyo)
John Kough – Biopesticides and Pollution Prevention Division in the Office of Pesticide
Programs, USEPA
Sue MacIntosh – MacIntosh and Associates Consulting
Wayne Parrott – Center for Applied Genetic Technologies, University of Georgia
Alaina Sauve – Syngenta Biotechnology, Inc.
Kate Walker, ILSI IFBiC
Flavio Zambrone – Planitox, ILSI Brazil
3
Task Force 10 Progress
The task force has met periodically and
prepared a consensus document for
publication. The manuscript is undergoing
peer review. This presentation
summarizes the consensus conclusions
of the task group regarding protein safety
assessment.
ILSI® International Food Biotechnology Committee
4
TF6 Consensus Views on Protein Safety
In 2008, ILSI-IFBiC Task Force 6 published guidance on
the safety assessment of introduced proteins (Delaney et
al., 2008). The major conclusions were:
(1) The primary safety assessment of any protein
introduced into a food/feed crop includes bioinformatics
screening and testing potential digestibility when
exposed in-vitro to simulated digestive fluids
(2) Proteins that are not structurally or functionally related
to known mammalian toxins or allergens based on
bioinformatics screening and are confirmed to be
digestible, are less likely to pose a hazard when
consumed.
ILSI® International Food Biotechnology Committee
5
TF6 Consensus Views on Protein Safety
(3) Where possible, the mode of action of the protein
should be assessed to confirm it does not pose
identifiable or anticipated safety concerns.
(4) Where there are no safety issues identified, no further
assessment of safety would be needed (unless required
as a condition of registration - Plant Incorporated
Protectants - EPA).
(5) Where there are unresolved safety concerns, further
toxicity testing may be indicated. The studies
undertaken should be hypothesis driven.
ILSI® International Food Biotechnology Committee
6
TF10 Consensus Views on Protein Safety –
New Issues*
• What is the impact of food processing on potential dietary
exposure to functionally active introduced proteins?
• For purposes of risk assessment, can the Threshold of
Toxicological Concern model be applied for proteins?
• What are the criteria for assessing “History of Safe Use” ?
• How do you assess the potential for toxicological
interactions between multiple proteins introduced into
combined trait food crops?
* The task force did not address allergenicity since this has been
well-addressed by PATC and other groups
ILSI® International Food Biotechnology Committee
7
Impact of Food Processing on
Functional Activity of Introduced Proteins
• Protein function (e.g., enzymatic activity) is dependent on
maintenance of appropriate tertiary structure. The
microenvironment surrounding the protein helps to maintain
protein tertiary structure.
• Many crops are processed to generate food products.
Processing conditions disrupt the cell microenvironment
through the use of heat, changes in pH, extraction of lipid, use
of physical shear forces. Protein tertiary structure and function
is often lost (denaturation).
• Food crops such as maize, soybean, wheat and rice are
normally processed before they are consumed.
ILSI® International Food Biotechnology Committee
8
Impact of Food Processing on Functional
Activity of Introduced Proteins
• In vitro heat stability tests have shown that introduced
proteins exposed to temperatures similar to those used
in processing lose biochemical function (e.g., enzyme
activity, or ability to control insect pests - Cry proteins
from Bacillus thuringiensis (Bt)).
• Analysis of processed food fractions has confirmed loss
of function for introduced proteins.
• Consequently, dietary exposure to functionally active
introduced proteins in processed food is likely negligible.
ILSI® International Food Biotechnology Committee
9
Threshold of Toxicological Concern - TTC
• TTC principle has been recommended for ranking and
prioritizing risk from exposure to substances present at
low levels in food where toxicology data is limited.
• Proteins were previously excluded from establishing a
TTC since a safe threshold for dietary exposure to
allergenic proteins had not been established.
• In regard to thresholds for allergens , a recent review of
human studies with 286 subjects in France suggest a
population threshold of approximately 1 mg/person for
highly sensitive peanut allergic patients. (Taylor et al.,
2010)
ILSI® International Food Biotechnology Committee
10
Threshold of Toxicological Concern - TTC
• The introduced protein could be considered for TTC if it does
not fit the profile of known allergens:
– Digestible
– Low (ppm) levels in food
– Not structurally related to known allergens
• TTC levels for consumption of proteins were estimated based
on published literature (Hammond and Cockburn, 2008)
– The highest dietary levels tested in all studies produced no test
article related effects
– The NOAELs were corrected for enzyme purity (default to 10%)
and divided by 100 and were averaged
– ~18 mg/kg/day (acute)
– 2.5 mg/kg/day (chronic)
ILSI® International Food Biotechnology Committee
11
Threshold of Toxicological Concern - TTC
• Case study - CP4 EPSPS enzyme
– An enzyme in the aromatic amino acid synthesis pathway,
from the bacterial strain CP4, that imparts tolerance to
glyphosate herbicide
– Prior to its introduction into food crops, there was no
history of consumption in food; however, homologous
EPSPS enzymes are commonly found in food crops.
• Potential chronic intake of CP4 EPSPS from consumption of
herbicide tolerant corn was estimated to be 4 µg/kg/day.
• This is 600× lower than TTC chronic limit, assuming no
denaturation during processing of corn.
• More realistic exposure – processing reduces CP4 EPSPS ~
2 orders of magnitude – 60,000× lower than TTC
ILSI® International Food Biotechnology Committee
12
History of Safe Use Considerations
• EFSA has recommended all introduced proteins with no
history of safe use (HOSU) in foods be subjected to at
least 28-day repeat dose toxicity testing unless there is
reliable information to demonstrate their safety.
• Proteins without a HOSU have been labeled as “novel”.
According to Websters’ dictionary, novel means “new
and not resembling something formerly known or used”.
• Does changing one, several, or many amino acids in a
protein with a HOSU make it truly novel?
ILSI® International Food Biotechnology Committee
13
History of Safe Use Considerations
• ~74% of the known proteins assigned to ~ 9,000 protein families
based on structural/functional relatedness.
– Within these families, evolutionary divergence accounts for considerable
variability in amino acid content of proteins of the same family
– Yet the residues forming the active sites of the proteins are conserved
– Accordingly, functional activity is preserved
• The amino acid content of EPSPS in soy, corn, and Baker’s yeast
varies considerably from CP4 EPSPS amino acid content
– 23 to 41% identity (invariant amino acid content at a given residue)
– 49 to 59% similarity (conservative substitutions of amino acids)
• Functionally related proteins can vary considerably in amino acid
content yet maintain similar functions through homologous active
sites and tertiary structures.
• Consequently, the HOSU of related EPSPS found in foods could be
considered as evidence for the safety of CP4 EPSPS.
ILSI® International Food Biotechnology Committee
14
Structures of E. coli
and CP4 EPSPS
The X-ray structures are
represented as ribbon
diagrams with the E. coli
enzyme shown in violet and
the CP4 enzyme in red.
Although E. coli and CP4
EPSPS enzymes share only
27.6% identity, X-ray crystal
structure analysis indicate
that both proteins have
evolved to fold similarly and
are superimposable.
Monsanto Scientific Literature #17600
Cytochrome c oxidase structural homology
• Cytochrome c oxidases from bacteria, fungi and animals
share less than 30% sequence identity yet have a similar
tertiary structure – catalytic site has been conserved
Modified from Voet and Voet (1995) Biochemistry, 2nd Ed
16
History of Safe Use Considerations
• There is a HOSU for tuna cytochrome c oxidase.
• Paracoccus and Rhodospirillum cytochrome c oxidases may
have no HOSU, but if they are:
1.
not related to known allergens or toxins
2.
capable of being digested
• They should also be safe to consume based on their similar
mode of action to related cytochrome c oxidases that have a
HOSU.
• This could be considered “reliable” information indicating no
need for further toxicity testing for consumption of Paracoccus
and Rhodospirillum cytochrome c oxidases.
ILSI® International Food Biotechnology Committee
17
Assessment of Potential Protein Interactions
• Through conventional breeding, it is possible to cross
parent lines containing new traits (introduced proteins) to
generate progeny with combined traits.
• Combined trait products have been generated to reduce
potential for resistance development and/or to combine
traits in the same plant that reduce environmental stress
factors (drought, weed competition for soil nutrients,
protection against insect pests, etc.)
• Mix and match traits depending on the needs of the
grower
ILSI® International Food Biotechnology Committee
18
Combining Traits to Reduce Plant Stress
19
Impact of Potential Protein-Protein Interactions
• Interactions are unlikely when the mode of action of
introduced proteins are fundamentally different.
– Enzymes that exhibit substrate specificity and/or carry out
different catalytic reactions (CP4 EPSPS or mEPSPS and PAT
enzymes that impart tolerance to structurally different herbicides)
• Interactions that could cause adverse effects on nontarget organisms are unlikely when different insect
control proteins with the same mode of action are
combined in the same crop.
– If the mode of action does not pose a risk for non-target
organisms, combinations of such proteins are also unlikely to
pose a risk – Bt Cry proteins.
•
Cry proteins whether tested individually or in
combination are not toxic to non-target organisms.
ILSI® International Food Biotechnology Committee
20
Cry Protein Mechanism of Toxicity
Environmental Health Criteria 217, Microbial Pest Control
Agent - Bacillus Thuringiensis (WHO IPCS, 1999)
21
Consensus Conclusions
Mammalian toxicology testing may be needed if there is
uncertainty about protein safety.
Toxicological evaluation of a functionally active introduced
protein may be appropriate if it is:
– structurally or functionally related to known mammalian toxins,
– stable in simulated gastric fluids and to processing conditions,
– has a mode of action that raises a toxicological concern.
When toxicology testing is considered necessary, it should
be driven by specific endpoint-related hypotheses and
employ relevant and appropriate techniques to address
the hypotheses.
ILSI® International Food Biotechnology Committee
22
Consensus Conclusions
Mammalian toxicology testing would not be
needed if the following information is available:
– An introduced protein without a HOSU is structurally
and functionally similar to proteins that do have a
HOSU (e.g., CP4 EPSPS and mEPSPS); its mode of
action is likely to be demonstrably similar and nontoxic.
– Modifications in the primary structure of a non-toxic
protein are unlikely to make it toxic when stability is
not significantly changed. As evidenced by:
• Evolutionary divergence in protein families across species
• Engineering of proteins to improve biological function (e.g.,
enzymes).
ILSI® International Food Biotechnology Committee
23
Consensus Conclusions
Most proteins denature and lose biological function when
heated or exposed to food/feed processing conditions.
Thus, human dietary exposure to functionally active
introduced proteins in such foods is likely to be
negligible, and far below an estimated TTC for proteins.
Acute and repeated-dose toxicology testing of proteins
introduced into food crops to date has found no evidence
they are toxic to non-target organisms (includes
homologous proteins with variant structures, e.g., amino
acid sequence or content)
ILSI® International Food Biotechnology Committee
24
Dankie
ILSI® International Food Biotechnology Committee
25
Backup Slides
26
EFSA 2009 View on Safety Testing
EFSA Recommendation (2008)
“Unless reliable information is provided demonstrating
the safety of the newly expressed protein, the safety
assessment of proteins with no history of safe use (for
consumption as food) should normally include a
repeated-dose toxicity test (normally 28 days) and not
rely on acute toxicity testing. Depending on the results of
this test, further testing may be necessary.”
ILSI® International Food Biotechnology Committee
27
Learnings from Protein Engineering
• For enzymes, it has been estimated that 50-70% of
random modifications in amino acid sequence are
approximately neutral in regard to enzymatic function,
30-50% are strongly deleterious to function and only
0.01 to 0.5% are beneficial to function. It is much easier
to disrupt function that to enhance it.
ILSI® International Food Biotechnology Committee
28
Learnings from Protein Engineering
• It is highly unlikely that modifications to a non-toxic
protein will make it toxic. It has been estimated that the
likelihood that 9 substitutions in the amino acid sequence
of the enzyme phytase being an exact match in
sequence to a toxic protein are just 1 in 2 x 1011 (Pariza
and Cook, 2010).
• Putting this into perspective, the probability of winning
the Powerball lottery has been estimated at 1 in 1.95 ×
108.
ILSI® International Food Biotechnology Committee
29