How the UMCG genetics dpt. uses CRISPR: opportunities and limitations

Genome editing

a type of genetic engineering in which DNA is inserted, deleted, modified or replaced in the genome of a living organism. Genome editing targets the insertions to site-specific locations

• focusses on the gene in its native state (cannot be done with random insertion as this is often employed to create an overexpression of a gene)

examples of genome editing + applications

• knockout alleles

• conditional alleles

• CRISPR genome engineering

Applications:

• functional studies of genes (eQTL)

• tagging of genes to study localisation/ expression patterns in vivo

• functional genetic screens

examples random integration + applications

examples:

• transgenesis

• transposon mutagenesis

• (retro) viral mutagenesis

applications:

• functional studies of overexpressed genes

• integration of exogenous genes (e.g., RNAi recombinases)

• functional genetic screens

CRISPR works on

single nucleotides only. Cannot be used to e.g., replace entire genes

mutation

change in a gene (e.g., single nucleotide)

• geneticists use “variant”/ SNP

genetic variation

the difference in DNA sequences between individuals (that what makes us unique)

(genetic) variant

any alteration in the most common DNA sequence (reference genome)

• genetic variation is often caused by a mutation, but may also arise in other ways (e.g., recombination)

single-nucleotide polymorphism (SNP)

a genetic variant affecting one nucleotide that occurs in >1% of the population. Less than that is considered a rare variant

usually, we were only interested in the…..of the genome, as these…… Nowadays, we are also interested in….., as there…..

coding, encode for proteins (these are the effector molecules, central dogma), non-coding parts of the DNA as most of the variation sits between genes (tricky to investigate)

mendelian disorders (monogenic disorders)

• ~100% genetic cause

• 1 or 2 alleles are affected by variants (either dominant or recessive)

• high effect size

• usually very rare (<1%)

• variant usually found in coding region of the gene

• example: Sickle cell’s disease

genetic treatment is possible!!

• aim of genetic department is to prove whether a variant (or two) is disease-causing or not because:

  • give solace

  • provide information about future risks

complex genetic disorders

• ~10-50% of the diseases have a genetic cause

• combination of numerable SNPs

• low effect size of the gene variants

• very common!

• environmental factors should be taken into account

• example: Alzheimer’s

• black box concerning pathways/ genes and context

when is a variant disease causing - problems

problem 1: pathogenicity cannot always be predicted, e.g., for 20-40% of all DNA variant found in hereditary cancer
problem 2: function of many genes are unknown and variants that are found are thus not actionable

=> pathogenicity testing using CRISPR may provide insights on the variant!!

pre-CRISPR techniques

• transient transfection

• transduction & random integration

  • introduce new DNA into cells using different types of vectors (plasmids/ BACs for transfection)

  • transfection methods (electroporation/ lipofection/ cationic polymers)

    • viral vectors (transduction)

What do the pre-CRISPR approaches yield?

• transient expression of transgenes

  • effective for a couple days/ weeks (episomal plasmids)
    
    

• random integration into the genome

  • rare event, requires the use of selection genes

=> not genome editing

current genome editing tools

• meganucleases (least feasible)

• Zinc finger nucleases

• TALEN

• CRISPR/Cas9 (most feasible)

meganucleases

• microbial endonucleases (restriction enzymes) with a long recognition sequence (>14bp)

• naturally specific, unlikey that gene of interest contains the required recognition sequence

• attempts to alter recognition sequences

  • mutagenesis/ high throughput screening of meganucleases

  • generation of fusion proteins (hybrid meganucleases)

Zinc finger nucleases

• artificial nucleases comprised of (engineered) Zinc finger domains and (engineered) catalytic subunit of FokI endonuclease

  • ZNF domains are derived from TFs

  • each domain recognises 3bp sequences

• fusion protein: 3-6 Znc finger domains + FokI catalytic subunit

• new nuclease has to be created for each application => laborious!

FokI

• endonuclease with separate DNA recognition and cleavage domains

• cleavage domains activated upon dimerisation

Transcription activator-like effector nucleases (TALENs)

• proteins secreted by plant pathogenic Xanthomonas bacteria

• binds promotor sequences in host cells to activate genes that aid infection

• DNA binding through repeated domain containing 33-35 aa repeat motifs with variable amino acids at position 12/ 13 (repeat variable diresidue RVD)

• RVD is specific for single nucleotides

• fusion with FokI: protein contains series of repeat motifs + catalytic subunit

• new nuclease has to be designed for each application!!

CRISPR allows for endogenous editing of a gene

• can be used in fibroblasts/ PBMCs/ induced pluripotent stem cells

=> advantage: the controls are isogenic, the genetic background is the same so we only look at the effect of the variant

patient-derived cells: change the variant to control and study the patient variant

what makes iPSCs difficult to CRISPR?

• low transfection energy

• low viability after transfection

• spontaneous differentiation

• FACS not always possible (check whether transfection was successful)

• (in)vulnerability for antibiotics

• some loci are inaccessible due to stem cell-ness

=> endless possibilities when the iPSCs are edited sucessfully!

example applications of CRISPR-modified PSC cultures

• cancer modelling

• tracing cellular populations

• CRISPR screening using CRISPRi

• gene repair

• epigenetic editing

GWAS

Genome-wide association studies

• used to ID risk variants (SNPs) for complex disorders

  • most SNPs are in non-coding regions => biological effect can be verified using eQTL

expression quantitative trait loci

• associates an SNP to changes in gene expression

cis-eQTL

SNP X has an effect on local gene A

• altered protein A levels have an effect on the binding to the TFs binding sites of downstream genes

=> 1 QTL may have multiple eQTL effects, therefore association does not equal causation

How to use CRISPR to test whether an SNP has an effect on gene-expression?

• HDR: change SNP in the genome of a cell line/ iPSC

  • conventional CRISPR + repair template

  • base editor/ prime editor

• change the whole region

  • conventional CRISPR for deletion

  • CRISPRi/ a

  • epigenomic editing

the big advantage of CRISPR is….

that you are in control of the situation!

questions to as when designing a CRISPR experiment

1. what is my RQ?

2. which tissue should I be looking at?

3. which model should I use? (e.g., in vivo vs in vitro/ patient-derived vs cell line)

4. which genetic pertubation is needed?(KO/ missense mutation/ overexpression)

5. how do I deliver my constructs?

6. how do I select my clones?

7. what am I going to test with my model?