Week 8: Post Genomic Techniques to Understand the Relationship Between Genotype and Phenotype

How can you know if a gene is being expressed?

Northern Blotting

Challenges with N. Blotting

• mRNA hard to obtain in large quantities

• mRNA easy to obtain but degrades easily

Issue with microarrays

cross- hybridization

Challenge of known transcript probes

• genome content of organism has to be known

• synthesise probes for every gene

• put on glass slide in a known order

Problems solved by NGS

How can cells be isolated?

• FACS sorter (fluorescence activated cell sorting)

• microfluidics

Understanding immune diversity and responses to pathogens

• know that the immune system is very complicated- many different cell types

• understand process to boost immune system

• take the spleen of the mouse (which contains all Immune System cells)

• sequence cells from spleen = information about the transcriptome

• each cell’s transcriptome used as a signature for each type of cell there is in the immune system

• give immune stress to the mouse e.g bacterial infection - sequence cells and see what has changed→ size/ number of different cells

• some genes turned on/ some off

Why are some genes expressed over other genes?

• Coding sequences of genes are interspersed with non-coding sequences

• DNA motifs in non coding sequences

• DNA motifs are recognised by specific proteins called TF. or transcriptional activators that bind to sequences

• When they bind, they engage RNA polymerase and tell RNA polymerase to transcribe

• Understand regulatory information + what the motifs are → helps decode the genome/ know whether a gene would be expressed under a given condition

What can be used to identify binding sites to DNA?

Chromatin Immuno-Precipitation (ChIP)

How does ChIP work?

• Take TF of interest which you have made in the tube

• See all the binding sites for that TF

• raise antibodies to that TF using the techniques talked about → mice/ phage

• fish out from the genome every piece of DNA that that TF would bind to

• fix cells quickly so TF gets stuck chemically

• with antibody, fish out every fragment of DNA the TF is bound to

• Remove TF from DNA and sequence DNA left in tube

• Map back to genome → see more hotspots where we map more sequence reads than other parts of the genomes

• Any similar features? e.g different TFs have different motifs/ preferred sequences they would bind to

What affects the likelihood of transcription?

• DNA packaging

• Not just about TF

• Only some TF are bound → another layer of info other than if that sequence is present

• DNA is packaged into chromatin

• Nucleosomes = series of histones that wrap DNA up

• Histones can be modified (methylation, phosphorylation)- changing how the chromatin is packed → allows accessibility to TF or not

• Many different histone ‘marks’→ ChIP can suggest a ‘histone code’

What might H3K4me mean?

Used ChIP seq

Histone 3 modified with lysine (K4) + me → methyl group

histone with lysine marks transcription start site

What does DNA methylation do?

causes DNA to get packaged tightly → silences gene expression

What adds methyl groups to DNA?

Enzyme DNA methyltransferase adds methyl group to cytosine residues → CG pairings that are methylated

What can be used to determine the extent of methylation?

• Bisulphite sequencing

How does bisulphite sequencing work?

1. Denaturation: Incubate fragments of genomic DNA at 95 C

2. Conversion: Incubate with sodium bisulfite at 65 C and low pH (5-6) deaminates cytosine residues in fragmented DNA

3. Desulphonation: Incubation at high pH at room temperature for 15 min removes the sulfite moeity, generating uracil

• Use NGS to see if genomes are methylated

• Cytosine residues can be sulphinated under particular conditions

• convert from cytosine → uracil

• if cytosine residue is still there→ methylation

Where are there extensive changes in patterns of methylation?

cancer cell lines

What is the encode project

• goal is to take a genome sequence and determine whether any gene will be expressed at any given time/ under any condition/ in what cell etc.

Mapping Targets:

• DNA methylation

• Open chromatin

• RNA binding

• RNA sequences

• Modified histones

• ChiP seq.

• Transcription factors

Historically (like ~ 70s) if you wanted to know the expression of a protein, what would you do?

• immunofluorescence

• western blot → run proteins and separate by size → incubate membrane with antibody

^ looking at one protein at a time and have to raise an antibody

Understanding the Proteome

• Separate in 2 dimensions: size + charge

• stain proteins using dye

• see spots → many different proteins in tissue

• Compare to cell populations→ look at relative abundance

Challenges

• don’t know what the proteins are

• tells us there are differences but what are they?

Extract spots/ proteins from gel and run through mass spectrometer

→ gives sequence information of peptides

Challenge

• Take each individual spot and do one at a time

• need a lot of protein to see it on the gel

Solution = advances in mass spec.

How does mass spectroscopy identify expressed proteins?

• magnet bends based on charge

• flight time depends on mass

• where it hits the detector → know the precise mass of every peptide

• then can back calculate what AA combination that can give that precise mass

• computationally find out sequence

How can we compare proteomes of two different samples?

• grow cells on medium containing light or heavy AA → supplemented with 13 C or 15 N

• Slightly changes mass of the protein → so can tell if it came from sample A or sample b

What does metabolomics do?

• provides a direct functional readout of the physiological state of an organism

How can metabolites be analysed?

mass spec

If there are thousands of things different between sample A and B → how do you know what is an important difference?

more commonality → more important

How do we move from correlation to causation

• remove factor - does the process change as a result = cause

• knockout/ remove function→ homologous recombination, inducible expression; enhance gene

What are two methods of understanding gene function?

1. Forward genetics

2. Reverse genetics

How does forward genetics work?

• using naturally occurring mutations or introducing mutations with radiation, chemicals, insertional mutagenesis (e.g transposon mutagenesis)

Piggybac transposons

• a movable genetic element that efficiently transposes between vectors and chromosomes through a “cut-and-paste” mechanism

• will jump in diverse species

What can transposon insertions be detected by?

• reporter gene expression (fluorescent)

• transposon disrupts gene expression → its fluorescence is expressed instead

• transposons can integrate without a promoter –results in ‘enhancer trap’

Issues with transposon insertions

• done in single cell of embryo

• see result in the phenotype so if gene is involved in making the embryo - you won’t see this

• no gradual/ subtle phenotype → gene expressed or not - see inactivity but not more activity

• possible solution: chemical mutagenesis

Chemical Mutagenesis

• expose cells to a chemical e.g Ethyl methanesulfonate (EMS) → modified bases

• in this case, guanine converted to thymine

introduction of small, subtle changes in the genome

• could be severe- affects AA in binding site, point mutation e.g sickle cell, remove start/ insert stop codon

• or could change active site to make it more active, make a change that stops it from being destroyed → methylation/ phosphorylation → more expression

Challenges to chemical mutagenesis

• don’t know which mutation is causal

• how to find the mutation if it is a small change → have to sequence the whole genome first

Solutions:

• knockout mutants

• have the same phenotype multiple times → find commonality → eliminate based on that

How can GWAS (genome wide association studies) be useful?

• allow ‘natural’ mutations affecting a trait of interest to be identified

• can do experiments by introducing mutations and accelerating evolution→ but mutations occur naturally all the time

• enough individuals → look for commonality

  = Genome Wide Association

How would I identify genes required for head development

• Take head tissue

• look for genes preferentially expressed- RNA sequencing

• Know if gene is mutant → remove function, take a library of mutants and screen for mutants that don’t have heads or overexpress genes

• turns out gene is TF → express TF in vitro, purify, tag , raise antibodies, CHIP seq