7. NAMs & Omics Revolution

Question 1

What are the classic measures of toxicity?

Answer 1

• Histopathology
• Clinical Chemistry
• Metabolism
• Physiology
• Enzymology
• Electron Microscopy

Question 2

What are the new approach methodologies of toxicity?

Answer 2

• Information from the exposure of chemicals in the context of hazard assessment (seen with Dr. McKeague)
• in silico approaches
• in chemico approaches
• in vitro assays (seen with Dr. McKeague)
• high-throughput screening (seen with Dr. McKeague)
• high-content imaging (seen with Dr. McKeague)

All of the -omics:
• genomics
• epigenomics
• transcriptomics
• proteomics
• lipidomics
• metabolomics

Question 3

What is the tiered protocol for toxicity testing recommended by the NIHS and Health Canada?

Answer 3

5 interacting levels in the tier:

1. Start with the chemical structure
2. What is happening in terms of structure activity relationship - in silico
3. In vitro -cell lines
4. Move to lower organisms like flies and worms. Fish are still debatable because they have backbones but a lot of work is still done on fish
5. Assess toxicity in mammals

If there is toxicity along the way you stop. Furthermore, the steps are interconnected with loops where if there are problems at one level can go back and test a different approach/other chemicals.

Question 4

What problems did we have in terms of proper toxicity testing ? What had to be done?

Answer 4

There were too many chemicals and not enough data. Therefore, compounds needed to be prioritized for more extensive toxicological evaluation, like high volume chemicals and the chemicals you think are most toxic. Also, since there are so many chemicals, we cannot test them all, so we want to be able to develop methods to predict their toxicity -> predictive toxicology.

Question 5

Define “in silico” approaches & name them. What is the purpose?

Answer 5

In silico experiments are software-based. In silico means: by means of computer modelling or computer simulation.

In silico approaches include:

1. Data management
2. Bioinformatics
3. *Quantitative structure activity relationship
4. *Read across
5. Modeling
6. Reverse dosimetry

The purpose is: From what we know about a family of chemicals, can we predict if another chemical will be toxic or not.

Question 6

What are “in chemico” approaches?

Answer 6

• Identify reactive compounds

- Use of analytical techniques

Question 7

What are the different approaches to the read-across in silico method? Label the axes on the graph. (See graph on slide 9 of L7)

Answer 7

X-axis: carbon chain length
Y-axis: Toxicity value
There are points on the graph that are placed based on chemicals that we know (analogs?) (their toxicity and carbon chain length is known). Then, the chemical that we are interested in is added to the graph in order to find its toxicity.
-> Interpolation is when the chemical falls within the carbon chain length of the known chemicals and an estimated toxicity can be determined.
-> Extrapolation is when the chemical of interest is outside of the known chemicals and we extend the trend line to estimate the toxicity.
-> Chemicals can also be compared with a “category approach” where they are compared to other chemicals that are in the same family. They can be put into families based on structure (ex: carbon chain and R-groups) or function (ex: chemicals that have estrogenic activity). It’s also possible to have “outliers” from the categories/families (ex: has the same carbon chain structure but the R-group is different).

Question 8

What is the purpose of the read-across method?

Answer 8

To be able to potentially find bad/toxic chemicals based on the computer analysis to identify whether or not we should stop the use of the chemicals.

Question 9

What is the main goal of the “omics”? (Genomics - Transcriptomics - Proteomics – Lipidomics -Metabolomics)

Answer 9

To determine whether gene, RNA, protein or metabolite expression profiles or ”signatures” can serve as markers to predict toxicity. (see if the toxicant exposure is causing harm to any of the “omics” ex: RNA, proteins, etc.)

Question 10

Define Genome.

Answer 10

All genes of an individual organism.

Question 11

Define Genomics.

Answer 11

The study of all of the genes of a cell or tissue

at the DNA level.

Question 12

Define Epigenomics.

Answer 12

The study of all epigenetic modifications, e.g., reversible modifications on a cell’s DNA or histones that affect gene expression without altering the DNA sequence

Question 13

Define Transcriptomics.

Answer 13

The study of all of the gene transcripts of a cell or tissue (RNA level) -> ex: how does the mRNA respond to a given toxicant?

Question 14

Define Proteomics.

Answer 14

The study of all of the proteins of a cell or tissue.

Question 15

Define Lipidomics.

Answer 15

The study of all lipids in an organelle or a cell. Ex: how does the toxicant modify the lipid profile of the cell?

Question 16

Define Metabolomics.

Answer 16

The study of all small chemicals in a cell.

Question 17

Describe the Omics cascade from genome to phenotype. (slide 14 L7) Why is the cascade important?

Answer 17

This cascade describes how the “omics” funnel down and relate to each other.

1. Genomics: At the DNA level we see what is possible in the cell (the potential).
2. Transcriptomics: At the RNA level we see what appears to be happening in the cell (current direction).
3. Proteomics: at the protein level we see the functional capabilities of the cell (proteins are what makes things happen).
4. Metabolomics: at the metabolite level of the cell we see what is ACTUALLY happening in the cell because metabolites are the limiting currency of the cell. Therefore, what is happening in the cell is dependent on the metabolomics.

Question 18

Define Toxicogenomics

Answer 18

Toxicogenomics: the effects of a toxicant on all of the genes of a cell or tissue at the DNA, RNA, protein and metabolite levels

Question 19

Define Toxicoepigenomics

Answer 19

Toxicoepigenomics: The effects of a toxicant on the expression of genes in cells or tissue that is not mediated by altering the nucleotide sequence

Question 20

What is the underlying hypothesis of toxicogenomics?

Answer 20

Hypothesis: The most sensitive marker of a toxicant on a cell is at the mRNA level. So… thats where you should study.
Gene expression PRECEDES protein changes and toxicity. If you see damage in a tissue that has occurred because proteins are effected, it’s most likely because transcripts of a protein are affected.
Furthermore, changes in gene expression can be measured at low doses of the toxicant.

Question 21

What are the different ways to measure gene expression via mRNA transcription?

Answer 21

1. Gene by gene analysis: Northern Blotting (look at one transcript at a time), RT- PCR (is it expressed or not expressed), qRT-PCR (quantify the amount of the transcript)
2. DNA arrays and microarrays
3. RNA sequencing

Question 22

Give a classical example of a toxicogenomics flow scheme.

Answer 22

1. Expose animal to a toxicant (multiple doses)
2. Isolate a tissue or tissues you want to look at
3. Isolate RNA and run your chips to find out which transcripts are being affected or not.

Question 23

How is toxicogenomics being used? (what are the applications) What are the goals?

Answer 23

1. Deciphering mechanism of action (pathway analysis) of toxicant
2. Response at low doses (Could see an affect at transcript level before you see an affect in the actual tissue)
3. Revealing potentially novel health effects (hypothesis generating science: Based on whats being affected you form your hypothesis, instead of the other way around.)
4. Identification of perturbed pathways – targeted follow-up

Goals:

1. Biomarker discovery
2. Predictive toxicogenomics (so you don’t have to test on animals or people)

Question 24

Describe array technology. Why was it exciting?

Answer 24

1. You have nylon membranes and you spot DNA sequences on it (ex: 300-400 base pairs) which act as a probe to detect gene expression.
2. Take RNA from tissue & reverse transcribe it to generate cDNA.
3. Put cDNA with radioactive label on the membranes and look for radioactivity to see which genes are expressed.

It was exciting because you could look at 212 transcripts at once.

Question 25

Describe microarray technology. Why was it exciting compared to array technology?

Answer 25

1. Instead of spotting on nylon membranes like in array technology, a new tech allowed you to spot on glass microscope slides (known as gene chips). 2. 60, 000 spots instead of 212 3. Use fluorescent probes (multiple probes per gene/transcript) instead of radioactivity 4. Use image analysis to tell you which genes are expressed.

Question 26

What is next generation sequencing / Shotgun library preparation? What is good about this new technique?

Answer 26

With this technique you can look at the ENTIRE genome (not just a small percentage), and this allows you to also look at regulatory elements. (I think it also looks at the non-coding part of the genome...like the part that makes micro RNA's and stuff). 1. chop up entire genome (there are cutting markers on each piece?) 2. You sequence each piece (approx. 400 bases) 3. Algorithms will rebuild all the transcripts based on the lengths (rebuilds the sequence of the entire genome)

Question 27

What is important to note about next generation sequencing technologies?

Answer 27

The more detailed sequencing you do, the more information you get. The more complex your analysis, the more expensive.

Question 28

What types of errors are encountered in RNA-Seq quantification?

Answer 28

1. Lack of depth in sequence reads: The more you run a sequence, the less errors there are (but it's more expensive and longer to analyze because there is more data) 2. Ambiguous non-matching reads: due to having a lot of repeat sequences 3. Errors in amplification 4. Errors in assignment

Question 29

What is a "clustering figure"? What does it represent?

Answer 29

In a clustering figure, you look at multiple samples relative to each other and analyze an Increase or decrease in gene expression depending on the colours in the figure (ex: red is increase and green is decrease. and a brighter colour means bigger change, black means no change.). Each lane (x-axis) will be a different replicate/sample. From this, you can see how transcripts are regulated based on the chemical it was exposed to. The y-axis represents the different genes/transcripts per sample.

Question 30

What types of statistics can tell you which genes are differentially expressed between samples?

Answer 30

1. Fold change (not a statistical test): look at how many folds of change there are. People used to say you need a 2 fold change between transcripts for it to be a significant difference. However, sometimes you may have a ten-fold change in a transcript which shows clear significance... but sometimes you can have a 1.8 fold change but it is in a key enzyme that regulates a pathway. 2. A result is considered ‘significant’ or differentially expressed if it is unlikely to have occurred by chance (this is what P-value measures). 3. p-value: P-value takes into account how many comparisons you are making. The smaller the p-value (less than or equal to 0.05), the stronger the evidence that you should reject the null hypothesis (no effect). A high P-value is evidence that the null hypothesis is true. 4. T-tests/ANOVA 5. Permutation test: 5a) MANOVA 5b) Significance analysis of microarrays (SAM) Today, studies need at least 5 replicates.

Question 31

What is the next step once you have figured out which transcripts/genes have changed due to the toxicant?

Answer 31

Gene ontology: Bioinformatics that lets you describe the characteristics of genes such as their cellular compartment, biological function, and molecular processes. You can use this to see how they are affected in different studies. You have to put all the transcripts together and see how they are related to each other (what is the major effect?) For example: are they part of a pathway (cell division, cell cycle, mitochondrial metabolism), cellular compartments, biological function, or molecular processes. This can be done using bioinformatics.

Question 32

What are KEGG pathways and what are they used for?

Answer 32

They're a collection of manually drawn pathway maps representing our knowledge on many molecular interaction and reaction networks (ex: metabolism, human diseases, etc.). using KEGG pathways you can see which transcripts in different pathways are up/down-regulated and see what process is being affected. For example are regulatory transcripts of the cell cycle being affected?

Question 33

What are different components that are necessary to design an experiment in order to study a mechanism of action?

Answer 33

1. Appropriate tissue/cells sampled (Transcriptomics works best in a purified cell type because different cells can have different up/down-regulation of transcripts and you won't be able to properly analyse. It is necessary to determine a specific cell type you want to examine). 2. Adequate sample size: Need at least 5 because of the errors in the methods in the microarrays. 3. Appropriate selection of timepoints (more than 1–early vs downstream vs disease – integration of multiple omics?) 4. Appropriate selection of treatment conditions (non-toxic): people used to use doses where the cells were dying and the same pathways would show up (apoptosis/necrosis) which doesnt tell you how the toxicant is working. You want to move backwards from a dose that's causing overt damage to the cells. 5. Sample collection–randomization or block design 6. HIGH QUALITY RNA!!! (garbage in, garbage out): if you don't take proper precaution you will breakdown the RNA 7. Implementation of QA/QC 8. Appropriate normalization and filtering 9. VALIDATION WITH ALTERNATIVE TECHNOLOGIES??: verify that what you're showing with the micro-arrays is real. Usually you would do qRT-PCR to verify and quantify the key transcripts you care about. In some cases, RNA-Seq, when it is done in enough depth does not need validation (doesn't use probes...it's actual sequencing).

Question 34

What is a central data repository? Give an example of one.

Answer 34

It is a public database where you must submit your full set of data in order to be able to publish your work so that you can't keep your data a secret. Ex: Micro-array or RNA-seq data. One recognized central data repository is "GEO" which is run by the NIH.

Question 35

What are the limitations of toxicogenomics? | The ones i put stars next to are the ones i'm memorizing because too bad i'm tired of this shit.

Answer 35

– Difficulty in analysis of high density data. – Difficulty in integration of data obtained by different technologies. – **Difficulty in linking “omics” data to specific adverse effects. – **Difficulty in translation statistical assessments into biological understanding. – Limitations of incomplete functional annotation of genome data bases. – **Incomplete knowledge of functional pathways and networks, particularly trans-genome relationship. – **Difficulty in proving that a certain chemical should be regulated based on transcriptomic data.

Question 36

What are the types of categories looked at in toxicoepigenomics?

Answer 36

1. DNA methylation 2. Histone modifications 3. Non-coding RNA 4. Chromatin organization

Question 37

Describe DNA methylation on cytosine. Why is it important?

Answer 37

- DNA methylation occurs through several DNA methyltransferases (DNMT) for example S-Adenosylmethionine (SAM). - Methyl marks added to 5' position of cytosine that modulate (usually suppresses) gene expression. - Methylating the cytosine causes it to function as a distinct base from regular cytosine. - The methylation process is very tightly regulated. Importance: - The upstream region of a transcript often has CPG islands. If they're highly methylated, the ability of the machinery to come and read the transcript will be suppressed. If they aren't methylated the machinery can read. (there are exceptions but it is a general rule) - Transgenerational effect on toxicity action depends almost exclusively on DNA methylation.

Question 38

Describe how histone modifications can regulate gene transcription.

Answer 38

Histone modifications affect how tightly DNA is wrapped around them. The histone code is basically all the different types of modifications that can be done to histones and what affect it has on the histone. The types of modifications are: 1. Acetylation: If you hyper-acetylate a histone it will have a tendency to leave DNA (open up and allow DNA to be reached by transcriptional machinery). 2. Methylation 3. Ubiquitination 4. Sumoylation 5. Phosphorylation

Question 39

What are the different types of non-coding RNA?

Answer 39

- miRNAs micro - piRNAs -> piwi interfering (discovered in testis and binds Piwi protein, smaller than miRNA, they have also been discovered in the ovary and some other tissues) - siRNAs small interference - shRNAs short hairpin - lncRNAs long non coding - snRNAs small nuclear - snoRNAs small nucleolar - tsRNAs -> tRNA derived small RNA (you can truncate tRNA to make tsRNA which is important in the load sperm carry to the zygote and in regulating gene expression in the zygote) - aRNAs antisense - exRNA extracellular

Question 40

What are Micro-RNA's?

Answer 40

- They are small cut strands of RNA that go on an mRNA that is being read to either suppress the reading of the mRNA or move the mRNA to be metabolized and recycled. - It comes from the nucleus and goes into the cytoplasm to regulate the way transcripts are read at the mRNA level instead of the DNA level. - Any miRNA can sit on many different mRNA transcripts. So, if you want to control a whole pathway, you can find one miRNA that affects all the mRNA's of the key transcripts.

Question 41

What is chromatin organization? Why is it important?

Answer 41

Chromatin can be organized in loops. As the loops get organized they form domains that get organized in certain spots in the nucleus. There is a structural organization of the nucleus essential for reading or non-reading of transcripts. It can be affected by age and by toxicants.

Question 42

Watch the lecture on the last 2 examples to see the kinds of graphs and tables he used for transcriptomics and proteomics.

Answer 42

Start at 1hr into the lecture

Question 43

What does the Transgenerational effect of toxicity depend on?

Answer 43

- Transgenerational effect on toxicity action depends almost exclusively on DNA methylation.