Toxicogenomics
oftewel
de zoektocht naar alternatieven voor de huidige
proefdiermodellen voor het bepalen van veiligheid
van chemische stoffen
Dept. of Health Risk Analysis & Toxicology
Maastricht University, the Netherlands
Jos Kleinjans
Vakgroep Toxicogenomics
Current protocol for toxicity testing
• Short-Term Tests for Genetic Toxicity
– Bacterial Reverse Mutation Test
– In vitro Mammalian Chromosomal Aberration Test
– In vitro Mouse Lymphoma TK +/- Gene Mutation Assay
– In vivo Mammalian Erythrocyte Micronucleus Test
• Acute Oral Toxicity Tests
• Short Term Toxicity Studies
– Short-Term Toxicity Studies with Rodents (3 dose levels, 2-4 weeks)
• Subchronic Toxicity Studies – Subchronic Toxicity Studies with Rodents (3 dose levels, 90 days)
• Carcinogenicity Studies with Rodents
• Combined Chronic Toxicity/Carcinogenicity Studies with Rodents
• In Utero Exposure Phase for Addition to Carcinogenicity Studies with Rodents
• Reproduction and Developmental Toxicity Studies – Multi-generation Reproduction Studies (3 dose levels, life time)
– Stand alone/multi-generation Developmental Toxicity Studies
• Neurotoxicity Studies
• Metabolism and Pharmacokinetic Studies
• Immunotoxicity Studies
Current demands on animal testing
in toxic risk assessment
Type of toxicological
research
Dutch %
n=63,305
EU %
n=781,866
% of
rodents
Acute lethal toxicity 11,1 6,7 37,9
Other acute toxicity 11,2 12,3 71,3
Subacute toxicity 10,8 89,4
Subchronic/chronic toxicity 5,3 13,8 90,0
Carcinogenicity 2,0 3,0 100,0
Mutagenicity 3,5 3,1 99,5
Reproductive toxicity 11,7 10,6 84,5
Teratogenicity 31,2 4, 8 50,2
Others 13,3 19,5 38,9
EU’s regulatory authorities
• Pharmaceuticals
– European Medicines Agency
• Food ingredients
– European Food Safety Authority
• Cosmetics
– Scientific Committee on Cosmetic Products
and Non-Food Products
• Industrial Chemicals
– European Chemical Agency
Increasing demands on chemical risk assessment:
• High failure rate of new drug candidates due to unmanageable toxicity, accounting for approximately 30% of this attrition
• The EU REACH program on industrial chemicals – Registration, Evaluation and Authorization of Chemicals:
– Existing and new substances should in the future be subject to the same procedure under a single system.
– Large amounts of additional tests required before 2018 • 30,000 existing chemicals already placed on the market since before 1981 and sold
at > 1 tonne per year
• EU-wide ban on animal use in cosmetics development – Council Directive 92/32/EEC (7th Amendment): testing ban prohibiting animal
testing of cosmetic products in the EU.
• Future EU regulations on food chemicals
The problem of high attrition rates of new
drug candidates
I. Kola & J. Landis. Nature Reviews/Drug Discovery 2004: 3: 711-715
~ 30 % due to human toxicity
The TeGenero case (NJEM, 2006: 354: 1869-1871)
• A t 8 a.m. on Monday, March 13, 2006, eight healthy young men entered a trial of a drug under development by the small German immunotherapeutics company TeGenero. Six of the volunteers were assigned to receive active drug, and two were to receive placebo.
• The six volunteers were to be the first humans to receive TGN1412, a humanized monoclonal antibody designed as an agonist of the CD28 receptor on T lymphocytes, which stimulates the production and activation of T lymphocytes.
• It was hoped that this product would benefit patients with B-cell chronic lymphocytic leukemia or autoimmune diseases such as multiple sclerosis or rheumatoid arthritis.
• However, after receiving injections of TGN1412, the six volunteers became desperately ill, had multiple-organ failure, and were transferred to an intensive care unit2 with what has been described as a cytokine release syndrome.
• Preclinical testing, including tests in rabbits and monkeys that used doses up to 500 times as high as the doses received by the first group of volunteers, reportedly showed no signs of toxicity.
Understanding REACH • REACH is an EU regulation to improve the protection of human health
and the environment from the risks that can be posed by chemicals, while enhancing the competitiveness of the EU chemicals industry. It also promotes alternative methods for the hazard assessment of substances in order to reduce the number of tests on animals.
• In principle, REACH applies to all chemical substances; not only those used in industrial processes but also in our day-to-day lives. Therefore, the regulation has an impact on most companies across the EU.
• REACH places the burden of proof on companies. To comply with the regulation, companies must identify and manage the risks linked to the substances they manufacture and market in the EU. They have to demonstrate to ECHA how the substance can be safely used, and they must communicate the risk management measures to the users.
• If the risks cannot be managed, authorities can restrict the use of substances in different ways. In the long run, the most hazardous substances should be substituted with less dangerous ones.
• REACH stands for Registration, Evaluation, Authorisation and Restriction of Chemicals. It entered into force on 1 June 2007.
The problem posed by REACH
• Existing and new industrial substances should in the future, following the phasing in of existing substances until 2018, be subject to the same procedure under a single system.
• According to the European Chemicals Bureau, only 14% of the highest production volume chemicals have a publicly available basic set of hazard data (as defined, but not necessarily required, in existing EU law); 65% has less than base-set and 21% has nothing
• Large amounts of additional tests required
• 30,000 existing chemicals already placed on the market since before 1981 and sold at > 1 tonne per year
• 50,000 to 120,000 intermediates
• Questions on false positive findings
• 50% of all chronically used human pharmaceuticals induce tumors in rodents, but only 20 human pharmaceutical carcinogens have been confirmed by epidemiologic studies
REACH Timeline
12
Nature 460, 1080-1081 (2009): “Some 65,000 companies made
more than 2.7 million pre-registrations for in excess of
140,000 substances”
• Claimed costs of REACH/questions on feasibiity
• time consuming: using all of Europe’s CRO capacity, REACH will take 11-40 years
• estimated to cost 2,5 – 6,5 billion Euros – Chemical industry: “Unemployment”.
“Deindustrialization of Europe”
– WWF: “Only 0,1% of the European chemical industry’s annual sales revenues”
• large ethical implications for animal welfare: REACH will use 8 – 20 million animals
• Claimed benefits of REACH
• benefits to human health amount up to €260 billion by 2020 in Europe
Ethical considerations
• Aristoteles
Hierarchical attitude towards nature which places humans above, and even separate from, nature.
Humans are different from the other animals because they not only sense but can make choices as well.
God expects, even demands, that we be stewards of His creation.
• Immanuel Kant
“The heart of a man can be judged by his treatment of animals."
• Mahamta Ghandi
“The greatness of a nation and its moral progress can be judged by the way its animals are treated"
• In their 1959 book "The Principles of Humane Experimental Technique", William Russel and Rex Burch presented the 3R principle referring to Replacement, Refinement and Reduction of animal testing.
Accepted by the European Partnership for Alternative Approaches to Animal Testing
Amendment to the latest consolidated version of
the REACH legislation REACH Regulation
1907/2006 :
• The Commission aims to reduce the number of animals used for testing, by half
• (39) The Commission, Member States, industry and other stakeholders should continue to contribute to the promotion of alternative test methods on an international and national level including computer supported methodologies, in vitro methodologies, such as appropriate, those based on toxicogenomics, and other relevant methodologies. The Community's strategy to promote alternative test methods is a priority and the Commission should ensure that within its future Research Framework Programmes and initiatives such as the Community Action Plan on the Protection and Welfare of Animals 2006-2010 this remains a priority topic. Participation of stakeholders and initiatives involving all interested parties should be sought.
“Beter dan een kristallen bol - Microarray
voorspelt het gedrag van een borsttumor”
• Onderzoekers van het Nederlands Kanker Instituut en het Erasmus Medisch Centrum hebben een grote hoeveelheid borsttumoren onderzocht
• Het vergelijken van de borsttumoren van patiënten jonger dan 55 jaar die bij diagnose een kleine tumor hadden (tot 5 cm diameter) leverde in Amsterdam een set van 70 genen op die heel goed gebruikt kunnen worden om te voorspellen of de tumor zich wel of niet gaat uitzaaien.
• In Rotterdam werd in materiaal van patiënten van alle leeftijden en met zowel grote als kleine tumoren een genexpressieprofiel van 76 genen gevonden, dat eveneens een hoge voorspellende waarde heeft.
• De genexpressieprofielen kunnen gebruikt worden om bij nieuwe patiënten het tumorweefsel te analyseren.
– Nadat de chirurg de tumor heeft verwijderd wordt een klein stukje hiervan bevroren, en naar een speciaal laboratorium gestuurd.
– Hier wordt de activiteit van de genen in de tumor vergeleken met het genexpressieprofiel. Het resultaat geeft aan of de patiënt een hoog of een laag risico heeft op uitzaaiingen.
– Afhankelijk van de uitslag kunnen de arts en de patiënt bepalen of een chemokuur of een hormoonbehandeling nodig is.
Animal
in vivo
Animal cells
in vitro
Human
in vivo
Human cells
in vitro
Development of alternatives to animal toxicity models:
in vitro-in vivo and inter-species extrapolation in
validating toxicogenomics-based predictive screens
?
Currently available ‘omics technologies for toxicology
Messenger RNA synthesis
Cellular functions/activities
DNA template
Protein synthesis
transcription
translation
Transcriptomics
Proteomics
Metabolomics
• a prospective study of more
than 600 patients from 22 US
tertiary care centers found
that acetaminophen-related
liver damage is the leading
cause of acute liver failure in
the country
• about half of such cases
involved unintentional
overdose
• toxic mechanism related to
generation of oidative stress
upon APAP metabolism
Biological pathways affected by APAP in mouse and rat liver. Heat map of
toxicological phenotypes (liver necrosis, serum ALT activity) and gene expression from
integrated mouse and rat data sets
APAP- induced gene expressions in vivo in rodents
Beyer RP et al. Toxicol. Sci. 2007
Prediction accuracies for rat liver toxicity vs. sub-toxicity by APAP
with different predictive methods as applied to blood
Bushel PR et al. PNAS, 2007
Differentially expressed genes in
blood that discriminate between
exposure to subtoxic/nontoxic or
toxic dose of APAP. Pictured is a
subgroup of genes involved in
immune response and inflammation.
Applying the human orthologue (66
genes) of the 270 rat blood gene
predictor for hepatotoxicity to blood
from APAP overdose victims versus
controls allows clear separation
Bushel PR et al. PNAS, 2007
ASAT Innovation Programme
Applying ‘omics technologies (transcriptomics and metabolomics) to a case study on human
volunteers
X X
ASAT Innovation Programme
Classic clinical liver toxicity tests
• Alkaline phosphatase (ALP)
• Gamma glutamyl transpeptidase (GGT)
• Alanine transaminase (ALT)
• Lactate dehydrogenase (LDH)
• Total bilirubin (Tbil)
• Albumin (Alb)
No significant changes after any APAP dose
ASAT Innovation Programme
Stringently selected, differentially expressed genes
vs T=0 at FDR < 10% and fold change of 1,2
(SAM analyses)
t = 1 t = 7 t = 25 Time course
0,5 g APAP/24 h
0 0 15 5
2 g APAP/24h
177 83 1811 1630
4 g APAP/24h
n.a. n.a. 1212 n.a.
ASAT Innovation Programme
Metabolomics in volunteers, upon administration of 0,5-4 g APAP/24 hr –
metabolomics data analysis
• Serum
– Polar metabolites using NMR
– (Semi-polar) metabolites using LC-TOF-MS
• Urine:
– Polar metabolites using NMR
– (Semi-polar) metabolites using LC-TOF-MS
– Identification of unknown metabolites
ASAT Innovation Programme
Dose dependency of metabolites in urine
-2
0
2
4
6
8
10
12
S-CH
3_p_g
luc
HS-p_g
luc
p_gl
uc_Nac
-glu
co/g
alac
tosa
min
e1
cyst
eine_
p_gl
uc
met
hoxy
_p_g
luc2
p_gl
uc_Nac
-glu
co/g
alac
tosa
min
e2
cyst
eine_
p
Nac-cy
stei
ne_p
_glu
c
cyst
eine_
p_SO
3
C9H10O
4N2-p
_glu
c
Nac-cy
stei
ne_h
ydro
xy-p
hydr
oxy-p
_SO
3
met
hoxy
_p_S
O3
met
hoxy
_p_g
luc1
C7H10O
3N4-p
_SO
3?
p_gl
uc
Nac-cy
stei
ne_p
etio
chol
anone
_glu
c
andr
osta
netri
ol_g
luc
p_SO
3
hydr
oxy-p
_glu
c
307_
gluc p
test
oste
rone
_glu
c
trp-p
?
0
1
2
3
4
5
6
7
8
HS-p_glu
c
cyste
ine_p_SO
3
S-CH
3_p_gluc
meth
oxy_p_glu
c2
cyste
ine_p_glu
c
p_gluc_
Nac-glu
co/gala
ctosa
min
e1
C9H10O4N2-p
_gluc
meth
oxy_p_glu
c1
hydroxy
-p_SO
3
p_gluc_
Nac-glu
co/gala
ctosa
min
e2
Nac-cys
tein
e_p_gluc
Nac-cys
tein
e_p
cyste
ine_p p
meth
oxy_p_SO
3
Nac-cys
tein
e_hydro
xy-p
hydroxy
-p_glu
c
p_gluc
trp-p
?
C7H10O3N4-p
_SO3?
p_SO3
307_gluc
androsta
netriol_
gluc
test
ostero
ne_gluc
etiochola
none_gluc
Average ratio (6 persons): MS signal 4g:0.5g
Average ratio (6 persons): MS signal 2g:0.5g
Stand.
Dev.
ASAT Innovation Programme
Results on MX and TX integration
• 0.5g APAP - t1
– Mitochondria, transcription (increase time course)
• 0.5g APAP - t25
– Transcription
• 2g APAP - t1
– Inflammatory response, ribosome (GSTO1)
• 2g APAP - t25
– Protein ubiquitination (decrease time course)
• 4g APAP - t25
– Transcription (GSTT1)
ASAT Innovation Programme
Rat-man comparison of APAP discriminating genes
Bushel PR et al. PNAS, 2007
rat human
ASAT Innovation Programme
• Few genes from Bushel’s profile are differentially expressed upon APAP administration to human volunteers at any time point at any dose level • The behaviour of this subset of genes is more comparable to Bushel’s control group than to his APAP overdose group
ASAT Innovation Programme
A. Kienhuis, M. van de Poll, H. Wortelboer, M. van Herwijnen, R. Gottschalk, A. Boorsma,
J. Kleinjans, R. Stierum, C. DeJong, J. van Delft.
Inter-species and in vitro – in vivo comparison of acetaminophen-induced gene expression profiles.
Toxicol Sci 2009 Feb;107(2):544-52.
Comparing APAP-induced changes in MetaCore pathways (all doses) at t = 24h
in human in vitro vs human in vivo
Cell process # pathways
Immune response, cytokine,chemokine-mediated signaling
7
Apoptosis/cell cycle 5
G-protein coupled receptor protein signaling
11
Transcription/translation 2
Response to extracellular stimulus
6
Small GTP-ase mediated signal transduction
1
hepatocytes in vitro
PBMCs in vivo
Using DNA microarrays, gene expression data are derived from
exposure of model systems to known toxicants (Group A, B, and C
genes). These data are compared to a set of gene expression changes
elicited by a suspected toxicant. If the characteristics match, a putative
mechanism of action can be assigned to the unknown agent.
Toxicogenomics-based screens for toxic class prediction
IP PL
037712
40
The carcinoGENOMICS
project - main outcomes -
Jos Kleinjans
Dept. of Toxicogenomics Maastricht University, The Netherlands
IP PL 037712
41
1. Genotoxicity testing
1. In vitro (2 or 3 tests on mutagenicity and
clastogenicity)
2. In vivo for in vitro positives
2. Carcinogenicity testing
1. For in vivo GTX compounds
2. For compounds to which humans will be exposed (drugs, cosmetics, some occupation settings)
Current carcinogenicity testing strategy
IP PL 037712
42
Demands for better tests: examples for genotoxicity and carcinogenicity
For pharmaceuticals, the current test battery on genotoxicity (bacterial mutagenesis, in vitro mammalian mutagenesis, in vitro chromosome aberration analysis and an in vivo chromosome stability assay) has been assessed to predict rodent carcinogenicity correctly by not more than 38 % while simultaneously producing high percentages of false positives
(Snyder RD, Green JW. A review of the genotoxicity of marketed pharmaceuticals. Mutat Res. 2001, 488:151-69 )
A recent survey of over 700 chemicals demonstrated that even 75–95% of non-carcinogens gave positive (i.e. false positive) results in at least one test in the in vitro test battery
(Kirkland D et al. Evaluation of the ability of a battery of three in vitro genotoxicity tests to discriminate rodent carcinogens and non-carcinogens. Mutat Res. 584 (2005) 1–256)
The current rodent cancer bioassays provide inadequate data to estimate human cancer risk at low dose; accuracies of approximately 60 % are achieved
(Ames BN et al. Cancer prevention, rodent high-dose cancer tests, and risk assessment. Risk Analysis, 16: 613-617 (1996) )
50% of all chronically used human pharmaceuticals induce tumors in rodents, but only 20 human pharmaceutical carcinogens have been confirmed by epidemiologic studies
For the important class of non-genotoxic carcinogens, no suitable test model is available
These assays have not been modified substantially since the initiation of their use.
IP PL 037712
43
Major aim of carcinoGENOMICS is to develop in vitro methods for
assessing the carcinogenic potential of compounds, as an
alternative to current rodent bioassays for genotoxicity and
carcinogenicity.
KEY TERMS:
Metabolome and transcriptome profiling.
Major target organs: the liver, the lung, and the kidney.
Robust in vitro systems (rat/human).
Interindividual variability.
Exploring stem cell technology.
Well-defined set of model compounds.
Phenotypic markers for genotoxic and carcinogenic events.
Extensive biostatistics to identify predictive pathways.
In silico model of chemical carcinogenesis.
Dedicated high throughput technology
IP PL 037712
44
Human proximal tubular epithelial cells:
Human Primary Cells
HK-2 human cell line
- Human pamplona virus transformed
RPTEC/TERT1 human cell line
- transfected with human telomerase (hTERT) (~ telomerase
positive)
Rat proximal tubular epithelial cells:
NRK-52E cell line
Carcinogenomics WP 3 Kidney Models :
Initial Cell Models:
IP PL 037712
45
robust human proximal tubular epithelial cell model selected and optimized based on: Morphology and characteristics Barrier function Genetic stability Metabolic characterization Transcriptomic profiling
Primary cilia
The Human - RPTEC/TERT1 Kidney model M. Wieser et al. Am. J. Physiol. Am J Physiol Renal Physiol. 2008, 295:F1365-75.
RPTEC/hTERT1 cells contain high levels of renal osmolytes e.g. betaine
Metabolites: 1:Isoleucine, 2:Leucine, 3:Valine, 4:EtOH, 5:Threonine, 6:Lactate, 7:Alanine, 8:Lysine, 9:5-Oxoproline, 10:Glutamine, 11:Pyruvate, 12:DMSO, 13:Tyrosine, 14:Glucose, 15:Myo-inositol, 16:Histidine, 17:1-methylnicotinamide, 18:NAD+/NADP+, 19:Tryptophan, 20:Contaminant, 21:GSSG, 22:Beta-alanine, 23:GSH, 24:Aspartate, 25:Phosphocholine, 26:Glutamate, 27:Phenylalanine, 28:Glycerophosphocholine, 29:Betaine, 30:Glycine, 31:MeOH, 32:ATP/ADP/AMP, 33:Formate and 34:Phenol. (nb. vertical expansion in aromatic region of both spectra)
Ellis JK et al, Mol. BioSyst., 2011, 7, 247–257.
Misclassification rates of the RPTEC/TERT1 Kidney model
• Human RPTEC/TERT1 cells were treated for 6h, 24h, and 72h with 1 concentration
• Each treatment was performed in at least 3 replicates
• Each tox class (GTX, Non-GTX, Non Carcinogen) is represented by 10 compounds
• Lowest misclassification rates obtained using:
– All experiments
– 72 h experiments
Classifier construction in the RPTEC/TERT1 human in vitro model using the Consensus dB pathway finding tool
-> Classifier was
based on 149 pre-
defined human
pathways (ANOVA p-
value<0.05) and 30
chemicals
-> Additional blinded
compounds were
correctly classified
with respect to all
three toxicity classes
-> Pre-validation study
assessed high inter-
lab reproducibility
Conclusions and future directions
• Our society needs assays for chemical safety
– which better predict human toxicity than current animal
models
– which are less costly and time-consuming
– which are preferably non-animal based
• The toxicogenomics approach applied to dedicated
human cellular models in vitro appears promising
• Future work will focus on
– integrating the latest ‘omics platforms (e.g next generation
sequencing)
– upgrading current in vitro models (e.g. 3D models, stem
cell-derived models)
– translating in vitro results to human disease (e.g. by
connecting to – drug-treated - patient-derived ‘ omics data)
– interacting with the EU’s regulatory authorities
Top Related