Limits of Concern for the Risk Assessment of GMP EFSA, Parma 24 th - - PowerPoint PPT Presentation

1

EFSA, Parma 24th May 2017

Marion Dolezel

Limits of Concern for the Risk Assessment

• f GMP

„Limits of Concern for the risk assessment

• f GMP”

2

 Federal Agency for Nature Conservation, Unit II 3.3. (Risk

Assessment GMO)

 UFOPLAN 2013 (testing & development project)  October 2013 – June 2017  Aims of the project:

 evaluation  operationalisation  exemplification

„Limits of Concern for the risk assessment

• f GMP”

3

 LoCs and protection goals  Role of LoCs in the stepwise testing approach  Relationship between the comparative safety assessment

and LoCs

 LoCs and long-term effects  LoCs for certain areas of risk  LoC values for non-target organisms practicable and

reasonable

 LoCs for species of conservation concern

„Limits of Concern for the risk assessment

• f GMP”

4

 Literature / background information  Stakeholders interviews  Topic-related workshops with scientific experts  Stakeholder-Workshop  2 scientific publications



Dolezel et al. (2017). Are Limits of Concern a useful concept to improve the environmental risk assessment of GM plants? Environ Sci Eur (2017) 29:7; doi 10.1186/s12302-017-0104-2

 Final report  public presentation

Background

5

Background

6

„…the level of environmental protection to be preserved is expressed through the setting of limits of concern…“

Background

7

 natural resources or natural resource services  EU legislation thresholds for acceptable adverse effect(s) for ERA purposes

Background

8

ERA of applicants:

 stat. sign. difference ≠ biological relevance  qualitative risk characterisation: negligible/not likely

EFSA objectives:

 stat. sign. differences = biological relevance ?  quantitative risk characterisation: effect size = LoC ?

Definition LoC (EFSA 2010)

9

“…the minimum ecological effects that are deemed biologically relevant and that are deemed of sufficient magnitude to cause harm”

LoCs and stepwise testing strategy (EFSA 2010)

10

Operationalization of the LoC concept (EFSA 2010)

11

protection goals relevant for the ERA

protection goals in the EU assessment endpoints measurement endpoints Limit of Concern

Abundance? Mortality? Weight? …

Open questions & challenges I

12

 Lack of harm definition for PGs  In ERA: damage to PGs cannot be tested directly – use of

proxies

 Ecological entities assessed ≠ protected entities  Spatial & temporal scales

 Lack of scientific knowledge on ecological significance of

adverse effects

consequence for definition of LoC

13

Integration of LoCs into current ERA system

14

Open questions & challenges II

 What shall a LoC constitute (trigger value, stop criterion)?  What is the consequence of an exceedance/non-

exceedance?

15

LoC in the stepwise testing approach

16

Open questions & challenges III

 Stat. sign. differences or non-equivalences need to be

followed up (EFSA 2010)

 Assessment of their toxicological or biological relevance

taking safety limits into account (EFSA 2010)

 Results of comparative assessment may be relevant for LoC

concept

 But usually not done in ERA practice

 Example GM soybean & lectins – effects on NTOs

17

LoCs and comparative safety assessment

18

Open questions & challenges IV

LoCs should be derived from EU-wide protection goals and should be valid for all receiving environments

19

LoCs and receiving environments

20

Open questions & challenges V

 Define LoCs if long-term effects are likely to occur and if

risk hypothesis can be formulated

 Requires risk management measures and post-market

environmental monitoring

21

LoC Definition

Acceptability threshold (quantitatively, qualitatively) for adverse effects on entities, functions, processes …

 …that trigger regulatory concern …  … that have the possibility to cause harm to the relevant protection

goal(s) (EFSA 2010)

 … or because these adverse effects are valued as being important for a

specific protection goal (see EFSA SC 2011)

22

Operationalisation of LoCs

3 Examples (risk areas):

• Outcrossing, persistence, invasiveness (HT oilseed rape)
• Impacts on NTOs (Bt maize)
• Effects due to changes in cultivation & management

methods (HT crops)

23

Operationalisation of LoCs

3 Examples (risk areas):

• Specific protection goals
• Indicators to assess effects (based on Kowarik et al. 2008)
• Existing thresholds
• Aspects to consider for setting LoC

Dolezel et al. (20xx): in prep.

24

Definition of LoCs - Example Bt maize

effect/risk effect/risk

GM pollen

possible starting point for LoCs

reduction in population size of faunal species toxic effects on test species

exposure

25

Example: Bt Maize – NT Lepidoptera

• biodiversity and agricultural protection goals
• Proposals for LoCs
• LoCs for lab studies: validation of thresholds needed
• LoCs for field studies: effects on less mobile larval stages
• exposure-based LoCs: amount of pollen deposited on field margins
• differentiation of LoCs between in-crop and off-crop (see PPPs)
• EFSA opinions (2011, 2012, 2015, 2016)
• Ecosystem service concept for ERA (EFSA SC 2016)

negligible effects for NT Lepidoptera

Example: Bt Maize – NT Lepidoptera

• ERA of Bt maize (MON810, Bt11, 1507)

Source Protection object Protection levels EFSA (2011, 2012) NT lepi in within maize fields & margins < 1 % global larval mortality EFSA (2011, 2012) NT lepi of conservation concern in protected habitats < 0,5 % local larval mortality EFSA (2015) NT lepi of conservation concern in protected habitats < 0,5-1 % larval mortality EFSA SC (2016) Lepi as service providing unit for ES < 1 % reduction in abundance < 1 % global mortality

26

27

Example: Bt Maize – NT Lepidoptera

• dose-response

relationships for Lepi larvae and Bt maize pollen (e.g. Holst et al. 2013, EFSA 2011)

• realistic pollen

deposition data on food plants (e.g. Hofmann et al. 2016)

28

Conclusions

• LoC – useful concept for ERA of GMOs
• Further elaboration and specifications
• LoC = scientific & political decision
• Potential overlap with ES concept
• Lack of scientific knowledge should not prevent defining

LoCs

• Enables harmonization between ERAs of different stressors
• Decisions on exceedances of LoC before concept is applied
• Transparency
• Increased confidence in conclusions of risk

29

Marion Dolezel Landuse & Biosafety Environment Agency Austria Tel.: +43-(0)1-31304 3120 Email: marion.dolezel@umweltbundesamt.at www.umweltbundesamt.at

dsRNA & miRNA pathways

Baseline information to support the risk assessment of RNAi-based GM plants

Petr Svoboda

Institute of Molecular Genetics AS CR, Prague

responses to double-stranded RNA

A U C G A U C G A U C G A U C G A U C G A U

RNA replication RNA basepairing Complementarity in nucleic acids

A U C G A U C G A U C G A U C G A U C G

Sources & types of dsRNA

Exogenous

viral life cycle inverted repeats convergent transcription pairing in trans

Endogenous

• frequently associated with parasitic mobile elements

genome

dsRNA response adenosine deamination interferon response

innate immunity

SEQUENCE-INDEPENDENT

RNAi

innate immunity genome defense

SEQUENCE-SPECIFIC dsRNA dsRBD

dsRNA binding domain

• ther

e.g. Staufen-mediated decay

RNA silencing RNAi & related pathways

The principle of RNA silencing

substrate

AGO Dicer

small RNA production targeting silencing (function)

RdRP AGO RdRP

amplificiation

uninjected parent not stained antisense injected

mex-3 in situ hybridization

dsRNA injected

Canonical animal RNAi pathway

RdRP Dicer

AGO AGO

AAAAA

mRNA cleavage siRNA

dsRNA

siRNA

dsRBP

aberrant RNA

AGO

dicing slicing

RISC

loading substrate synthesis

viruses, convergent transcription, inverted repeat transcription, RdRp activity, artificial

Canonical animal RNAi pathway

• defense against elements (viruses, TEs) producing dsRNA
• with or without RdRP amplification loop

RNAi: long dsRNA-induced sequence-specific mRNA degradation

dicing repression

RISC (miRISC)

loading

miRNA

Dicer

AGO

AAAAA

inhibition of translation mRNA degradation

GW182

pre-miRNA

dsRBP AGO Drosha

pri-miRNA pre-miRNA

DGCR8

Microprocessor complex

EXP5

mRNA degradation

RELOCATION TO P-BODIES

nucleus cytoplasm

Canonical animal miRNA pathway

DGCR8

• post-transcriptional regulation of gene expression
• stoichiometry (miRNA abundance/cell) matters

miRNA: regulatory RNAs produced by Dicer from genome-encoded small hairpin precursors

DICER small RNA biogenesis

Dicer

RNaseIIIa “platform” (DUF283) RNaseIIIb PAZ DEXD HELICc

NSMB 2012 19(4):436-40.

Dicer

• Dicer makes small RNAs

Three distinct small RNA populations

Dicer

• predictability of the 5’ end

ARGONAUTE small RNA loading target recognition silencing

Argonaute-mediated silencing effects

AGO

AGO2

AAAAA

mRNA cleavage by AGO2

Eukaryots

“miRNA-like”

mRNA degradation P-BODIES CCR4-NOT

AGO

AAAAA

Inhibition of translation/deadenylation GW182

Eukaryots

“RNAi-like” RdDM

DNA methylation

Plants

• ther transcriptional silencing

heterochromatin e.g. S. pombe … poorly understood in animals

POST-TRANSCRIPTIONAL TRANSCRIPTIONAL

Argonaute structure and function

AGO PIWI PAZ N MID mRNA siRNA 3’ 5’ 5’ 3’

• Argonaute structure explains principles of target recognition

and repression

Argonaute structure and function

AGO PIWI PAZ N MID mRNA siRNA 3’ 5’ 5’ 3’

• Dissociation constants for seed matching targets are in a pM range
• low abundant miRNAs unlikely to have significant regulatory effects
• seed match + abundance = siRNA off-targeting

Argonaute-mediated silencing effects

AGO

Parameters influencing silencing by small RNAs

• small RNA abundance (stoichiometry)
• target site accessibility
• complementarity with the target
• type of silencing (transcriptional/post-transcriptional)

A small RNA seed sequence defines the minimal sequence complementarity required for silencing

OFF-TARGETING Perfect vs. imperfect basepairing

nucleus cytoplasm

Ago

Class 2 hairpin (miRNA-like) Class1 hairpin (shRNA) siRNA

EXPRESSION VECTOR . TRANSFECTION

Ago2

AAAAA

Cleavage of mRNA by Ago2

Ago

AAAAA

mRNA degradation Inhibition of translation RELOCATION TO P-BODIES RISC loading short RNAs (miRNAs and siRNAs)

Argonaute – targeting & off-targeting

AGO

• off-targeting is siRNA-specific
• any siRNA has off-targeting potential

Jackson et al. (2003) Nature Biotech

Argonaute – targeting & off-targeting

AGO

Argonaute – targeting & off-targeting

AGO

Jackson et al. (2003) Nature Biotech

• off-targeting is largely concentration-dependent
• it is strongly reduced in sub-nanomolar range

transfected at 100nM

siRNA2 siRNA4 siRNA3 siRNA1 pool

Thermofisher/Dharmacom website

Argonaute – targeting & off-targeting

AGO

• siRNA pooling is a way to reduce concentrations of individual

siRNAs while keeping the constant siRNA amount in a transfection

• natural siRNA pools produced from siRNAs are highly specific

because of a highly diluted off-targeting effect

Argonaute – targeting & off-targeting

AGO

• off-targeting potential stems from seed sequence frequency
• siRNA knock-downs - usually employ nM concentrations
• hydrodynamic transfection (40 mg/mouse – Nature, 418, 38-39)

Ago

AAAAA

miRNA-like Inhibition

• f translation

seed = nucleotides 2-7

Ago2

AAAAA

RNAi-like Cleavage of mRNA by Ago2

seed = nucleotides 2-7

RNA silencing in different

• rganisms

mRNA degradation

RNAi

Dicer

Co-existence of miRNA & RNAi pathways

defense gene control inhibition

• f translation

miRNA

Dicer AGO AGO

Arthropod set up

mRNA degradation

RNAi

Co-existence of miRNA & RNAi pathways

defense gene control inhibition

• f translation

miRNA

AGO Dicer

Annelida set up (some Molluscs?)

mRNA degradation

RNAi

defense gene control inhibition

• f translation

miRNA

AGO Dicer

interferon response

PKR

Co-existence of miRNA & RNAi pathways

defense Vertebrate set up

mRNA degradation

RNAi

Co-existence of miRNA & RNAi pathways

defense gene control inhibition

• f translation

miRNA

AGO Dicer

RdRP

Nematode set up (some Molluscs?)

Nematodes

endoRNAi antiviral defense

replication dsRNA dsRNA

DRH-1

Dicer

gene control

ERGO-1

AAAAA

mRNA cleavage

RdRP

WAGOs RDE-4

1o siRNA 1o siRNA 26G RNA 22G RNA

DRH-3

2o siRNA

Dicer

immunity

RDE-1

AAAAA

mRNA cleavage

RdRP

SAGO-2 RDE-4

1o siRNA 22G RNA

DRH-3

2o siRNA

ERI ERI RDE-8

1o siRNA 22-23 nt

RNA clearance

exoRNAi

RDE-1

AAAAA

mRNA cleavage

RDE-8

RdRP

WAGOs

dsRNA injection feeding soaking 1o siRNA 1o siRNA 22-23 nt 22G RNA

DRH-3

2o siRNA

DRH-1

Dicer

RDE-4

nucleus cytoplasm

Dicer

ALG-1/2

gene control

AGO

AAAAA

inhibition of translation

AIN-1

pre-miRNA

miRNA

miRNA 22-23 nt pri-miRNA Microprocessor complex

Nematodes

RNA clearance

exoRNAi

RDE-1

AAAAA

mRNA cleavage

RDE-8

RdRP

WAGOs

dsRNA injection feeding soaking 1o siRNA 1o siRNA 22-23 nt 22G RNA

DRH-3

2o siRNA

DRH-1

Dicer

RDE-4

0.5 - 1.0x106 dsRNA molecules per each gonad arm Tabara et al., 1998

Plants

RDR6

RNA clearance (post-transcriptional)

transgene & viral silencing

AGO

dsRNA viral long hairpin 21/22 nt siRNA

SDE3

sense RNA

RDR6

SGS3

DCL4/2 DCL3

AGO4/6

24 nt siRNA

RdDM

AGO SDE3

DNA methylation (transcriptional)

HEN1 HEN1 DRB3 DRB4

sense RNA TAS loci

AGO

miRNA

RDR6 DCL4

AGO1/7

tasiRNA

21nt tasiRNA

Gene regulation during development

HEN1 DRB

miRNA pathway in plants & animals

gene control

Plants

mRNA cleavage inhibition of translation pre-miRNA

HYL1 AGO1

miRNA 21 nt pri-miRNA

SE HYL1 SE

DCL1 DCL1

AGO1

AAAAA

SUO

HEN1 HEN1

nucleus cytoplasm nucleus cytoplasm

Arthropods

AGO1

gene control

AGO1

AAAAA

inhibition of translation

GW182

LOQS

miRNA 21-23 nt

Dicer-1

nucleus cytoplasm

Mammals

AGO1-4

gene control

AGO1-4

AAAAA

inhibition of translation

GW182

TARBP2

miRNA 21-23 nt

Dicer-1

pre-miRNA pri-miRNA pre-miRNA pri-miRNA

DGCR8 DGCR8

Drosha

DGCR8 DGCR8

Drosha

Plants

AGO1

DCL2 DCL3

AGO10 AGO7

DCL1

miR-390 miR-156/166 U U A A

AGO2 MAIN miRNA PATHWAY AGO4/6/9

21 nt 24 nt

long inverted repeats (evolving miRNAs) ALTERNATIVE miRNA PATHWAY

DCL4

• highly complex RNA silencing system 4x Dicer, 10-20 Argonautes
• a number of small RNAs, TGS & PTGS effects

Plants – transcriptional silencing

Canonical RdDM Non-canonical RdDM

• highly complex RNA silencing – crosstalks & redundancy

small RNA mobility

RNAi mobility - systemic RNAi

dsRNA dsRNA dsRNA

dsRNA delivery RNAi effect Cell autonomous RNAi Systemic RNAi Environmental RNAi

dsRNA

Example

0.5 - 1.0x106 dsRNA molecules per each gonad arm

mammals

• C. elegans

some Arthropods (Tribolium) plants

• C. elegans

insects

Plants – spreading of RNA silencing

short distance long distance

Plants –> Animals

 

? ? ?

environmental & systemic RNAi circulating miRNAs

? ?

Huang et al., 2006 Baum et al., 2007 Mao et al. 2007

environmental & systemic RNAi

Plants –> Animals

Unclear/controversial issues: Mechanism of transport

• Mechanism of transport across membranes not explained
• Unclear if free or bound to a protein
• Survival in digestive tract?

Effector complex structure

• Would require binding of methylated single stranded RNAs by AGO

Targeting stoichiometry

• Concentrations estimated 68-250 fM – too low
• Authors calculate ~850 molecules per cell, cannot be verified – data not released

Plants –> Animals

• meta-study of xenomiRs of 824 datasets from human tissues and body fluids
• xenomiRs commonly present in tissues (17%) and body fluids (69%),
• low abundance, 0.001% of host human miRNA counts
• no significant enrichment in sequencing data from tissues and body fluids exposed

to dietary intake (e.g. liver).

• no significant depletion in tissues and body fluids that are relatively separated

from the main bloodstream (e.g, brain and cerebro-spinal fluids)

• the majority (81%) of body fluid xenomiRs stem from rodents, which are rare

human dietary contributions, but common laboratory animals.

• body fluid samples from the same studies are clustered by xenomiR compositions
• suggesting technical batch effects.
• feeding studies - no transfer of plant miRNAs into rat blood, or bovine milk

sequences into piglet blood.

doi: 10.1261/rna.059725.116

RNA, advanced online, Jan 6., 2017

Key points

• a targeting repertoire of a small RNA is largely determined by its

seed – nucleotides 2-8.

• not absolute rule (non-canonical binding)
• allows some predictability, especially for conserved targets
• RNAi-like cleavage or miRNA-like target repression silencing

effects are primarily defined by AGO isoform and basepairing

• targeting efficiency is determined by:
• small RNA abundance (stoichiometry)
• target site accessibility
• complementarity with the target
• vertebrates have lack systemic RNAi, an RdRP amplification

system, and highly processive Dicer -> inefficient RNAi

• plant small RNA pathways use 3’ end 2-O-methyl modification of

all small RNAs. In mammals, such modification is found only in piRNAs bound to PIWI AGO cladein the germline

CREDITS: Miloslav Nic Tomas Novotny Jan Paces END

Rana 2007

nucleus cytoplasm GW182 AGO2 AAAAA Cleavage of mRNA by Ago2 Exportin 5-mediated transport AGO AAAAA mRNA degradation Inhibition of translation relocation to P-bodies GW182 miRNA duplex

Dicer

GW182 AGO RISC loading

Mammalian microRNA pathway

pri-miRNA pre-miRNA DGCR8

Microprocessor complex

DGCR8 Dicer cleavage targeting pre-miRNA Drosha

miRISC

RNA silencing in selected taxonomic groups

Task I Mode-of-action of dsRNA and miRNA pathways

Animal Dicer evolution

• RNAi-dedicated Dicer-2 in Arthropods is a

derived character

• the mammalian “miRNA” Dicer is related

to miRNA-producing Dicer-1 in Arthropods

• Dicer in C. elegans produces efficiently

miRNAs and siRNAs

Dicer

“miRNA” Dicer

mRNA degradation

RNAi

Dicer

Co-existence of miRNA & RNAi pathways

defense gene control inhibition

• f translation

miRNA

Dicer AGO AGO

mRNA degradation

RNAi

Co-existence of miRNA & RNAi pathways

defense gene control inhibition

• f translation

miRNA

AGO Dicer

mRNA degradation

RNAi

defense gene control inhibition

• f translation

miRNA

AGO Dicer

interferon response

PKR

Co-existence of miRNA & RNAi pathways

defense

Interferon response induced by long dsRNA (>30bp)

sensing specific responses

MDA5 TLR3 PKR OAS RIG-I

common response INTERFERON RESPONSE

ISG

interferon-stimulated genes

eIF2a P RNaseL

global inhibition

• f translation

global mRNA degradation

2’,5’-OA

• The interferon response can be detected/monitored

Small RNA pathways in animals

mammals birds fish

Arthropoda Nematoda Annelida Mollusca Cnidaria Porifera Chordata ECDYSOZOA LOPHOTROCHOZOA

Chelicerata Myriapoda Crustacea Hexapoda Trilobita †

Mammals (and vertebrates in general)

OAS MDA5 TLR3 PACT

Dicer

miRNA RNAi

AGO1-4

gene control

AGO1-4

AAAAA

inhibition of translation

GW182

AGO2

AAAAA

mRNA cleavage pre-miRNA miRNA dsRNA siRNA

TARBP2 PKR OAS RIG-I

translational repression RNAse L IFN signaling

interferons & interferon stimulated genes

Interferon response

common sensors

RNA silencing

antiviral defense

dsRNA

• miRNA pathway is the main RNA silencing pathway
• main dsRNA response = sequence-independent interferon response

Annelids

Dicer

miRNA RNAi

AGO AGO

AAAAA

inhibition of translation

AGO

AAAAA

mRNA cleavage pre-miRNA miRNA dsRNA siRNA

TARBP2 ? OAS RIG-I ? MDA5 ?

RNAse L signaling

innate immunity?

dsRNA response

common sensors

RNA silencing

dsRNA

? ?

• almost no functional data, set-up seems similar to mammals

Molluscs

RdRP Dicer

miRNA RNAi

AGO

gene control & antiviral defense?

AGO

AAAAA

inhibition of translation

GW182

AGO

AAAAA

mRNA cleavage pre-miRNA miRNA dsRNA siRNA

TARBP2 PKR OAS RIG-I MDA5

translational repression RNAse L IFN signaling

interferons & interferon stimulated genes

Interferon response

common sensors

RNA silencing

antiviral defense

dsRNA

?

MX

• almost no functional data, set-up seems similar to mammals
• possible RdRP loop – would make it similar to nematodes

Arthropods

nucleus cytoplasm

AGO2

defense gene control

AGO2

AAAAA

mRNA cleavage siRNA dsRNA

PKR RIG-I MDA5

Interferon response

common sensors

RNAi

dsRNA

AGO1

gene control

AGO1

AAAAA

inhibition of translation

GW182

pre-miRNA

LOQS

miRNA

miRNA 21-23 nt pri-miRNA Microprocessor

R2D2

Dicer-1 Dicer-2

TLR3?

signaling

innate immunity?

• separated miRNA & RNAi
• sensors of the interferon response present

Nematodes

• C. elegans is an outstanding model for analyzing RNA silencing
• highly complex RNA silencing system
• one Dicer but 26 Argonautes and 3 RdRPs
• four pathways can be recognized
• miRNA
• exoRNAi
• endoRNAi
• antiviral defense
• primary and secondary RNAs (amplification of the response)
• cytoplasmic and nuclear Argonautes
• systemic RNAi, sensitive, cheap, temperate areas worldwide

0.5 - 1.0x106 dsRNA molecules per each gonad arm Tabara et al., 1998

Nematodes

nucleus cytoplasm

Dicer

ALG-1/2

gene control

AGO

AAAAA

inhibition of translation

AIN-1

pre-miRNA

miRNA

RNA clearance

exoRNAi

miRNA 22-23 nt

RDE-1

AAAAA

mRNA cleavage

RDE-8

RdRP

WAGOs

dsRNA

endoRNAi antiviral defense

replication injection feeding soaking dsRNA dsRNA

DRH-1

1o siRNA 1o siRNA 22-23 nt 22G RNA

DRH-3

2o siRNA

Dicer

gene control

ERGO-1

AAAAA

mRNA cleavage

RdRP

WAGOs RDE-4

1o siRNA 1o siRNA 26G RNA 22G RNA

DRH-3

2o siRNA

Dicer

immunity

RDE-1

AAAAA

mRNA cleavage

RdRP

SAGO-2 RDE-4

1o siRNA 22G RNA

DRH-3

2o siRNA

ERI ERI

pri-miRNA Microprocessor complex

RDE-8

1o siRNA 22-23 nt

DRH-1

Dicer

RDE-4

Plants

AGO1

DCL2 DCL3

AGO10 AGO7

DCL1

miR-390 miR-156/166 U U A A

AGO2 MAIN miRNA PATHWAY AGO4/6/9

21 nt 24 nt

long inverted repeats (evolving miRNAs) ALTERNATIVE miRNA PATHWAY

DCL4

• highly complex RNA silencing system 4x Dicer, 10-20 Argonautes
• a number of small RNAs, TGS & PTGS effects

miRNA pathway in plants & animals

gene control

Plants

mRNA cleavage inhibition of translation pre-miRNA

HYL1 AGO1

miRNA 21 nt pri-miRNA

SE HYL1 SE

DCL1 DCL1

AGO1

AAAAA

SUO

HEN1 HEN1

nucleus cytoplasm nucleus cytoplasm

Arthropods

AGO1

gene control

AGO1

AAAAA

inhibition of translation

GW182

LOQS

miRNA 21-23 nt

Dicer-1

nucleus cytoplasm

Mammals

AGO1-4

gene control

AGO1-4

AAAAA

inhibition of translation

GW182

TARBP2

miRNA 21-23 nt

Dicer-1

pre-miRNA pri-miRNA pre-miRNA pri-miRNA

DGCR8 DGCR8

Drosha

DGCR8 DGCR8

Drosha

Plants

RDR6

RNA clearance (post-transcriptional)

transgene & viral silencing

AGO

dsRNA viral long hairpin 21/22 nt siRNA

SDE3

sense RNA

RDR6

SGS3

DCL4/2 DCL3

AGO4/6

24 nt siRNA

RdDM

AGO SDE3

DNA methylation (transcriptional)

HEN1 HEN1 DRB3 DRB4

sense RNA TAS loci

AGO

miRNA

RDR6 DCL4

AGO1/7

tasiRNA

21nt tasiRNA

Gene regulation during development

HEN1 DRB

Plants – transcriptional silencing

Canonical RdDM Non-canonical RdDM

• highly complex RNA silencing – crosstalks & redundancy

Plants – spreading of RNA silencing

Timeline

• extremely large volume of literature
• majority not related to the review purpose (RNAi technology,

miRNA biology, innate immunity …)

Pubmed: RNAi OR RNA interference OR miRNA OR microRNA OR dsRNA

Timeline

1990 2000 2010 2015 discovery

• f RNAi

RNAi mechanism solved (AGO2 crystalized) co-suppression 1st miRNA Let-7 & siRNA Dicer discovered Dicer crystalized GW182:AGO2 Single- molecule analysis

• f AGO

binding pre-RNAi era

• mainly plant PTGS research
• initial miRNA research

RNA silencing core molecular mechanism deciphering

• mutation screens
• biochemical approach

miRNA research Plant co-suppression, PTGS, VIGS, TIGS etc. mechanisms RNAi research

Literature review process

• 1. Searches in bibliographic databases

n = 641 975

• 2. Citation pearl growing using publications known

to be landmark publications in the field. (Annex C)

• 3. Removal of duplicates (compilation of a

comprehensive set of scientific and grey literature).

• 4. Exclusion of references published since 2000

without DOI

• 5. Filtering for relevance to individual ELS questions
• 6. Screening of titles and abstracts
• 7. Study selection based on full-text reports

n = 682 911 n = 239 987 n = 190 734

• 1. Searches in bibliographic databases

n = 641 975

• 2. Citation pearl growing using publications known

to be landmark publications in the field. (Annex C)

• 3. Removal of duplicates (compilation of a

comprehensive set of scientific and grey literature).

• 4. Exclusion of references published since 2000

without DOI

• 5. Filtering for relevance to individual ELS questions
• 6. Screening of titles and abstracts
• 7. Study selection based on full-text reports

n = 682 911 n = 239 987 n = 190 734

Literature review process

reference database Scopus keyword search Pubmed WoS

citations of 47 landmark papers

covers highly-cited pioneering papers from the pioneering times when nomenclature was not established and uniformly adopted across the field

ProQuest

Literature review process

double strand* rna, dsrna rna interference, rnai, gene silenc*, ptgs Dicer, rnase III, argonau*, ago1, ago2, Piwi, wago, rde1 or rde-1, r2d2 tarbp2 or trbp2 mirna or microrna, sirna, 21u rna

• ligoadenylate, Pkr

Literature review process

• 1. Searches in bibliographic databases

n = 641 975

• 2. Citation pearl growing using publications known

to be landmark publications in the field. (Annex C)

• 3. Removal of duplicates (compilation of a

comprehensive set of scientific and grey literature).

• 4. Exclusion of references published since 2000

without DOI

• 5. Filtering for relevance to individual ELS questions
• 6. Screening of titles and abstracts
• 7. Study selection based on full-text reports

n = 682 911 n = 239 987 n = 190 734

• described in detail in
• 2. Data & Methodologies

SPECIFIC SET-UP FOR EACH TASK/ELS QUESTION OR TAXONOMIC GROUP:

Literature review process

MAMMALS BIRDS FISH MOLLUSCS ANNELIDS ARTHROPODS NEMATODES PLANTS

choice of keywords for reference inspection

Literature review process

references with abstracts with highlighted relevant keywords relevant/irrelevant choice chosen filtering keywords exclude include

Reis et. al (2015)

filtering keywords

Literature review process

publication type annotation

• ptional annotation buttons

Literature review process

relevant include export to Endnote

Literature review process

• 1. Searches in bibliographic databases

n = 641 975

• 2. Citation pearl growing using publications known

to be landmark publications in the field. (Annex C)

• 3. Removal of duplicates (compilation of a

comprehensive set of scientific and grey literature).

• 4. Exclusion of references published since 2000

without DOI

• 5. Filtering for relevance to individual ELS questions
• 6. Screening of titles and abstracts
• 7. Study selection based on full-text reports

n = 682 911 n = 239 987 n = 190 734

• described in detail in
• 2. Data & Methodologies

lack of 3’ overhangs induces IFN via Rig-I dsRNA > 30 bp activates PKR and 2’,5’-OAS some sequence motifs within ssRNA can activate IFN cationic lipid-RNA complexes activate IFN via TLR3 and TLR7 5’ triphosphate introduced by phage RNA polymerases activates IFN siRNA < 30 bp can activate PKR lack of 3’ overhangs induces IFN via Rig-I dsRNA > 30 bp activates PKR and 2’,5’-OAS some sequence motifs within ssRNA can activate IFN cationic lipid-RNA complexes activate IFN via TLR3 and TLR7 5’ triphosphate introduced by phage RNA polymerases activates IFN siRNA < 30 bp can activate PKR

Interferon response induced by siRNAs

24 hours 72 hours mock siRNA A siRNA B1 siRNA B2 siRNA C mock siRNA A siRNA B1 siRNA B2 siRNA C

dsRBD Zβ deaminase Zα

NLS NES

ADAR1p150 ADAR1p110 ADAR2 ADAR3

Adenosine deamination

Kono & Akiyama, 2013 DOI: 10.5772/55203

nucleus cytoplasm

ADAR1

degradation

?

AGO

Dicer Dicer

AGO

dicing

RISC-loading complex

asymmetry sensing HSP90

AGO

Dicer HSP90 Argonaute loading passenger strand removal

Argonaute loading

AGO

sense siRNA strand (passenger) antisense siRNA = targeting (active) strand!

5’-CGUACGCGGAAUACUUCGAdTdT-3’ ||||||||||||||||||| 3’-dTdTGCAUGCGCCUUAUGAAGCU-5’

• siRNA duplex undergoes loading of one of the strands on RISC
• 5' portion of the selected strand is paired less stably than its 3' portion
• ssRNA could reconstitute RISC; 10- to 100-fold higher concentrations required

relative to siRNA duplexes (Martinez et al., 2002, Cell. 110(5):563-74)

Omics and bioinformatics applied to the characterization of plant materials

24 May 2017 Esther Kok

Acknowledgements

RIKILT Wageningen UR Jeroen van Dijk Martijn Staats Marleen Voorhuijzen Martijn Slot Roberta Mariot Joseph Evaristo Rico Hagelaar

2

WUR - Biometris Hilko van der Voet WUR – Plant breeding Ronald Hutten Richard Visser University of Nijmegen – Chemometrics Jeroen Jansen

SAFETY ASSESSMENT OF A NEW GM VARIETY

3

Parent crop

Identity, phenotypic & agronomic performance History of safe use Compositional analysis

Donor, transgene(s) and delivery process

Description of donor Description of vector DNA Transgene delivery process Characterisation

• f introduced

DNA Characterisation

• f insertion site

Characterisation

• f gene

product(s)

Structure, identity and characterisation Mode of action/Specificity Toxicity Allergenicity

New GM crop

Identity, phenotypic & agronomic performance Nutritional analysis Compositional analysis Safety analysis (animal studies)

Focus: potential presence

• f unintended effects of the genetic modification

Unintended effects

4

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses) Why may potential unintended effects not be relevant?

• We have a long history of innovative plant breeding with

very few examples of adverse effects

• Plant breeders take their responsibility to develop new crop

varieties that are safe and nutritious

• It is unlikely that a safe variety is transformed into an unsafe

variety as the result of unintended effects

Unintended effects

5

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses) Why may potential unintended effects be relevant?

• A range of new and powerful techniques (Crispr-Cas,

synthetic biology) allow the rapid introduction of new RNAs, proteins and secondary metabolites, unknown to our food supply chain, possibly even unknown to nature.

• Because of the targeted and precise techniques plant breeding

programmes are becoming shorter with less time (years/harvests) to assess new varieties for altered characteristics

Unintended effects

6

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses)

• Hazard identification on the basis of:
• Molecular characterisation
• Phenotypic analysis
• Agronomic performance
• Compositional analysis (targeted analyses)

Unintended effects

7

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses)

• Hazard identification on the basis of:
• Molecular characterisation
• Phenotypic analysis
• Agronomic performance
• Compositional analysis (targeted analyses)
• Animal feeding trials with whole foods

Unintended effects

8

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses)

• Hazard identification on the basis of:
• Molecular characterisation
• Phenotypic analysis
• Agronomic performance
• Compositional analysis (targeted analyses)
• Animal feeding trials with whole foods

In the GRACE project:

• animal feeding trials with whole foods
• detailed compositional analyses - same maize materials

Compositional analysis (targeted)

9

Non-GM counterpart GM variety

Compositional analysis (targeted)

10

Non-GM counterpart GM variety Conventional variety 1 Conventional variety 2 Conventional variety 3 Conventional variety 6 Conventional variety 5 Conventional variety 4

Compositional analysis, targeted vs omics analysis

11

Targeted analyses:

• key nutrients (macronutrients/micronutrients),
• key anti-nutrients, including natural toxins

Omics analyses:

• Transcriptome: all transcribed DNA products (RNA)
• Proteome: all proteins
• Metabolome: all secondary metabolites

Unintended effects

12

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses) Targeted analyses

• Few hundreds of end-

points

• Limited coverage of

individual metabolic routes

• Advanced data analysis

is required (comparison with conventional varieties)

• Natural variation needs

to be included!

Omics analyses

• Many thousands of end-

points

• Broad coverage of

individual metabolic routes

• Advanced data analysis

is required (comparison with conventional varieties)

• Natural variation needs

to be included!

Omics analyses

13

If the DNA code is not clear If we can not interpret observed changes in the DNA We use the compositional data (targeted analyses)

Omics analyses lead to very large datasets. The question is: how to analyse for meaningful differences in the omics profiles, given the fact that there is much natural variation between plants due to e.g.

• Genotype
• Environmental conditions of growth

(soil and climatological conditions) Model developed with Wageningen UR Biometris (statisticians) and University of Nijmegen, dept of Chemometrics Basic criterium: profiles of varieties that can not be considered as safe should fall outside of the one class

14

Compare transcriptomics profiles

Omics profiles of commercial crop plants Build a one- class classification tool Classify the new profile

Omics profiles of commercial crop plants Build a one- class classification tool Classify the new profile

Within the safe one- class? Further analysis No further analysis Yes No

Omics analysis: one class model (SIMCA)

1 6

Safe

Omics analysis: one class model (SIMCA)

1 7

IN

• r

OUT?

Safe Parent GM

Omics analysis: one class model (SIMCA)

GRACE conclusions

• Unintended effects can likely be more effectively traced by informative
• mics analyses compared to animal feeding studies with whole foods:
• GRACE data have shown that the comparative safety assessment

(Implementation Regulation 503/2013) can also be adopted for omics data, e.g. using the one-class model approach: the GM variety can be compared to its closest conventional comparator, as well as to a range of conventional varieties.

• GRACE data have shown that the one-class model classifies mycotoxin-

contaminated maize samples as outside of the one ‘safe’ class, the results would provide a scientific basis for further analysis.

GRACE conclusions

• All maize varieties fed to the test animals (90-d) in the course of

GRACE were classified by the one-class model as inside of the one ‘safe’ class

• The one-class model classifies experimental potato varieties that are fit

for human consumption but genetically more distant from the lines that are currently consumed, in almost all cases as outside of the one ‘safe’ class (indicating that the one-class model represents a conservative approach)

GRACE conclusions

• Based on these observations: omics data provide qualitatively

structured details of the plant material which facilitates a non-targeted “safety“ evaluation.

• Thereby it provides a better basis for the decision on the scientific

rationale to frame the subsequent risk assessment steps, which may include the performance of an animal feeding trial with the plant- derived whole food/feed

Thank you very much for your attention!

esther.kok@wur.nl

23

1

Cartagena Protocol

• n Biosafety

and Synthetic Biology

Boet Glandorf GMO Office, RIVM The Netherlands

2

3

4

Objective Cartagena protocol

5

To contribute to ensuring the safe transfer, handling and use of LMOs resulting from modern biotechnology that may have adverse effects on the biological diversity, taking also into account risks to human health

Cartagena Protocol

6

• Negotiated under the Convention
• n Biological Diversity (CBD)
• Adopted 29 January 2000 after

4 years of intense negotiations

• Entry into force: 9 September

2003

• 170 ratifications/ accessions
• 8 meetings of the governing

body (COP-MOP)

Scope

7

Applies to: Transboundary movement, transit, handling and use of all LMOs that may have adverse effects on biodiversity, taking also into account risks to human health

How does the Protocol work?

8

The Protocol establishes rules and procedures to regulate the movements of LMOs from one country to another

Categories of LMOs

9

• LMOs for intentional

introduction into the environment (such as seeds and live fish)

• LMOs intended for direct use

as food, feed or processing, LMOs-FFP (such as agricultural commodities – corn, canola and cotton)

• LMOs for contained use (such

as bacteria for laboratory scientific experiment)

Procedures for Transboundary Movements of LMOs

Two key procedures: – The Advance Informed Agreement (AIA) procedure – Procedures for LMOs intended for direct use as food, feed or for processing (LMOs-FFP)

Precautionary Approach

Objective: Safe Transfer, Handling and Use of LMOs Biosafety Clearing-House (BCH) , Capacity-Building, Compliance and COP-MOP Supporting Mechanisms:

• Risk

Assessment

• Risk

Management

• Information

Sharing

• Public

Awareness & Public Participation

• Rules/

Procedures:

• AIA Procedure
• Procedure for

FFP

• Decision -

making

• Handling,

Transport, Packaging and Identification:

• Documentation

for Shipment

• Standards

Key Provisions of the Protocol

12

Regulation (EC) 1946/2003 regulates transboundary movements of GMOs and transposes the Cartagena Protocol on Biosafety into EU law

The Protocol sets common rules for the trans-boundary movement of Living Modified Organisms to ensure the protection of biodiversity and human health globally. The Regulation, which addresses in particular exports of GMOs, obliges EU countries to take legal, administrative and other measures to implement their commitments under the Protocol. It establishes the procedures for the trans-boundary movement of GMOs including:

• notification to importing parties
• information to the Biosafety Clearing House
• requirements on identification and accompanying documentation

Cartagena protocol

Main discussion items at last COP MOP meetings were:

• Adoption/endorsement of Guidance on Risk Assessment
• f LMOs (Road map)
• Development of further RA guidance for specific groups
• f LMOs, such as LM fish and organisms obtained by synbio

13

Synthetic biology: new and emerging issue under the CBD?

2012 Decision XI/II New and Emerging issues Noting, based on the precautionary approach, the need to consider the potential positive and negative impacts of components,

• rganisms and products resulting from synthetic biology techniques
• n the conservation and sustainable use of biodiversity, requests the

Executive Secretary, subject to availability of resources…

• Compilation of developments in synbio
• Synthesis of information on synbio
• Review by technical body (SBSTTA)
• Is synbio a new and emerging issue?

14

Synthetic biology: new and emerging issue under the CBD

2014 Decision XII/24. New and emerging issues: synthetic biology?

• Coordinated approach synthetic biology and Cartagena Protocol
• SBSTTA not clear if synbio is new and emerging issue
• Precautionary approach: only introduction after risk assessment,

risk management is in place

• Establishment of Ad Hoc Technical Working Group (AHTEG) and
• nline forum
• Mandate: operational definition synbio, difference synbio and LMO,

benefits and risks synbio, best practices for risk assessment and monitoring, framework to address impacts After review by SBSTTA, draft decision to be discussed in COP 2016

15

General Surveillance van GGO's in Nederland | 27 september 2010 16

3 opinions in total

17

Synthetic biology: new and emerging issue under the CBD

2016 Decision XIII/17 New and emerging issues: synthetic biology

• Precautionary approach: only introduction after risk assessment,

risk management is in place, also applies to organisms with a gene drive

• Current applications synbio are LMO
• Risk assessment methodology LMOs is applicable to synbio
• Unclear of some organisms obtained by synbio in the future will fall

under definition of LMO

• Collection of further info on experience with synbio, such as risk

assessments, effects (positive, negative)

• Extension AHTEG, online forum
• Mandate: to review recent technological developments, synbio
• rganisms that are no LMO, best practices, detection and

monitoring

18

Cartagena protocol

Cancun 2016 Draft decision

• Adoption of Guidance on Risk Assessment of LMOs
• Development of further RA guidance for specific groups
• f LMOs, such as for organisms obtained by synbio

2016 Decision VIII/12

• Take note of Guidance as one of the Guidances
• No extension of the AHTEG on Risk Assessment
• No development of further guidance on LM fish and synbio
• Compilation of views on:
• topics for further guidance and
• criteria to decide when such guidance is considered

necessary

19

Next step

Outcome of:

• Online forum Cartagena Protocol
• Online forum synbio
• AHTEG synbio

will be discussed in technical body (SBSTTA) in 2018 Decisions will be taken in COP MOP (Cartagena Protocol) en COP (synbio) at the end of 2018 in Egypt, based on report SBSTTA

20

21

Thanks!!