Techniques 3: Functional Genomics – Sequencing, Morpholinos & Genome-Wide RNAi

Functional Genomics – Position within the Module

• Techniques Session $3$ of LIFE$2072$ (The Genome and Human Disease)
  • Lecturer: Dr Bill Wickstead
  • Builds on:
  • Techniques $1$ (Next-Gen sequencing fundamentals)
  • Techniques $2$ (Northern blots & conventional expression analysis)
  • Links forward to Techniques $4$ (lentiviral biology & advanced vectors)

Learning Objectives (Re-stated & Expanded)

• After revising these notes you should be able to:
  • ❖ Explain how NGS read-outs can be repurposed to interrogate chromatin status & protein–DNA interactions.
  • ❖ Compare morpholinos and RNA interference as tools for experimentally reducing mRNA levels.
  • ❖ Describe how large-scale sequencing makes genome-wide functional screens feasible.
  • ❖ Evaluate ethical / practical considerations (off-target effects, essential genes, organismal viability).

Central Dogma Refresher – Why ‘Expression’ ≠ mRNA Only

• Canonical flow: DNA → RNA → Protein (Replication | Transcription | Translation).
• Information can be modulated at multiple layers:
  • DNA accessibility (chromatin compaction, DNA/histone modifications).
  • RNA quantity (transcriptional rate, mRNA turnover, RNA processing).
  • Translation efficiency & protein turnover.
• Therefore measuring only steady-state mRNA provides an incomplete view of gene function.

Measuring mRNA Levels

• Historical method: Northern blot

  • Complex RNA mixture probed with a single labelled DNA/RNA fragment.
  • Advantages: quantitative & shows transcript size.
  • Limitation: one transcript per assay ➔ low throughput.

• Modern method: RNA-seq

  1. mRNA purified & reverse-transcribed → cDNA.
  2. Deep sequencing on an NGS platform.
  3. Map reads to reference genome/transcriptome.
  4. Read counts ∝ mRNA abundance (must normalise for gene length & library size; e.g. $\text{TPM}$/$\text{FPKM}$).
  • Captures all transcripts simultaneously & discovers splice isoforms, allele-specific expression, etc.

Reminder: Illumina NGS Workflow (Techniques 1)

• $1$ Prepare genomic (or cDNA) fragments + ligate adapters.
• $2$ Bind single-stranded fragments to flow-cell lawn of primers.
• $3$ Bridge amplification → millions of clusters.
• $4$ Sequencing-by-synthesis generates short reads with quality scores.
• Key realisation: if we can translate any biological question into a change in read counts, we can exploit NGS power.

Beyond Sequence – Chromatin & Protein Binding

• Gene activity depends on:
  • DNA methylation, histone PTMs, nucleosome positioning.
  • Recruitment of transcription factors (TFs), polymerase, chromatin remodelers.
  • Nuclear localisation (euchromatin vs heterochromatin; lamina association).

ChIP-Seq (Chromatin Immunoprecipitation + Sequencing)

• Workflow:
  • Cross-link proteins to DNA in vivo (e.g. formaldehyde).
  • Fragment chromatin (sonication or MNase digestion).
  • Immunoprecipitate with antibody against:
  • Specific DNA-binding protein (e.g. TF).
  • Histone modification (e.g. H3K27ac) – reveals epigenetic marks.
  • Reverse cross-links, purify DNA & NGS.
• Read-mapping yields peaks = loci enriched for the protein/modification at cross-link time.
• Applications:
  • Identify TF binding motifs, promoters/enhancers, open chromatin.
  • Infer active vs repressive chromatin landscapes.

Variants Linking Other Properties to Sequencing

• DNase-Seq & ATAC-Seq – measure chromatin accessibility (nuclease or Tn5 insertion bias).
• FAIRE-Seq – formaldehyde-assisted isolation of regulatory elements (open chromatin DNA partitioning).
• MNase-Seq – map positioned nucleosomes.
• NOMe-Seq – nucleosome occupancy & CpG methylation.
• DRIP-Seq – DNA:RNA hybrid (R-loop) mapping.
• Principle: pairing a biochemical enrichment with NGS converts diverse chromatin questions into read-count data.

Manipulating RNA Levels – Functional Perturbations

Endogenous Small-RNA Regulation

• microRNAs (miRNAs) (~$22$ nt) naturally modulate translation & stability.
  • Endogenous but transient; effect sizes modest.
  • Researchers sought synthetic analogues with stronger, programmable knock-down.

Morpholinos

• Phosphorodiamidate morpholino oligomers (PMOs)
  • Backbone: morpholine ring + neutral phosphorodiamidate linkage.
  • Binds complementary RNA/DNA by standard Watson–Crick base pairing.
  • Nuclease-resistant (not degraded), therefore stable in cells/embryos.
• Functional uses:
  • Block translation initiation (bind over/just upstream of AUG).
  • Alter pre-mRNA splicing (mask donor/acceptor sites).
• Delivery:
  • Must be introduced exogenously (cannot be transcribed by cell).
  • Typical systems: microinjection into zebrafish, Xenopus, chick, sea urchin embryos.
• Example (Zhou & Chapman $2009$):
  • Targeted MagT1 + TUSC3 genes in zebrafish with two distinct morpholinos (A & B).
  • Readouts: Mg$^{2+}$ homeostasis, mRNA/protein levels, embryonic survival curves.
  • Scrambled morpholino control essential for specificity.

RNA Interference (RNAi)

• Nobel Prize $2006$ to Fire & Mello for discovery in C. elegans.
• Natural mechanism: long dsRNA → diced into $∼20$ bp siRNAs.
  • Dicer ➔ siRNA duplex ➔ loaded into RISC.
  • Guide strand directs Argonaute to complementary mRNA ➔ cleavage & degradation.
  • Biological role: antiviral defence & transposon silencing.

Lab Exploitation

• Two entry routes for dsRNA/siRNA:

  1. Exogenous delivery
  • Microinjection, electroporation, lipid-based transfection.
  • Feeding dsRNA-producing bacteria (unique to C. elegans).
  • Effect wanes when source removed.
  1. Endogenous production
  • Introduce DNA construct expressing dsRNA.
    • Long hairpin or short-hairpin RNA (shRNA) designs.
    • Driven by RNA Pol III promoters (e.g. U6, H1) for precise short transcripts.
  • Constructs can be stable (integrated or viral) & inducible.

• shRNA lentiviral vectors (Rubinson et al $2003$):

  • Efficient transduction of mammalian cells, including non-dividing types.
  • Enable pooled screens & long-term knock-down.
  • Must include control shRNA to assess off-target/IFN responses (Yao et al $2016$).

Scaling Up to Genome-Wide Screens

C. elegans Pioneer Study (Kamath et al, Nature $421, 2003$)

• Realisation: worms trigger systemic RNAi simply by feeding on bacteria expressing dsRNA.
• Pipeline:
  1. PCR amplify gene fragments for $\sim$$19{,}427$ predicted genes.
  2. Clone into T7-promoter vector in E. coli HT115 (dsRNA-competent).
  3. Generate $14{,}440$ bacterial strains (coverage $\approx86\%$).
  4. Feed worms; score phenotypes: lethality, locomotion, development, etc.
• Delivered first near-global functional annotation in a metazoan.

"Genomicising" Functional Screens via NGS Barcodes

• Concept: each perturbation (shRNA, sgRNA, transposon insertion, etc.) carries a unique DNA barcode.
  • PCR amplify the region encoding the barcode from mixed cellular DNA.
  • Sequence counts = surrogate for cellular abundance/fitness under condition X vs control.
  • Converts complex phenotypic selection into differential read-count analysis.

Pooled shRNA Screens in Mammalian Cells

• Create lentiviral shRNA library ($\sim10^{3}\text{–}10^{5}$ unique hairpins).
• Infect target cells at low multiplicity (≤$0.3$) so each cell receives ≈ one shRNA.
• Apply selective pressure (drug, oncogene activation, senescence signal…).
• Compare barcode representation before vs after selection by NGS.

Example $1$ – Drug-Target Essentiality (Huang et al $2014$)

• Library: $442$ shRNAs covering known drug targets.
• Post-growth ( $12$ days) disappearance of certain shRNAs identified essential genes ➔ prioritise druggable vulnerabilities.

Example $2$ – Senescence Bypass (Aarts et al $2017$)

• Genome-wide library: $\approx58{,}000$ shRNAs (multiple per gene).
• Selection: cells forced into senescence; survivors sequenced.
• Enriched shRNAs pinpointed factors whose suppression enables senescence escape (implications for cancer biology).

Ethical, Philosophical & Practical Considerations

• Off-target effects: si/shRNAs & morpholinos may partially match unintended transcripts ➔ false phenotypes. Mitigation: multiple independent reagents + rescue experiments.
• Essential genes: Complete knock-down can be lethal; conditional or partial suppression strategies needed.
• Delivery toxicity: Electroporation, viral vectors, microinjection may harm cells/embryos.
• Data interpretation: Sequencing counts require rigorous normalisation (library size, PCR bias, copy number variations).
• Genome editing vs knock-down: CRISPR offers knock-out/activation alternatives; choice depends on question (loss-of-function vs hypomorph).

Connections to Broader Principles / Future Directions

• Systems biology: Combining ChIP-seq, RNA-seq, ATAC-seq, proteomics yields integrative regulatory maps.
• Precision medicine: Pooled loss-of-function screens identify synthetic lethal partners for targeted cancer therapies.
• Epigenetic therapeutics: Mapping histone modifications guides development of HDAC, DNMT, BET inhibitors.
• High-content phenotyping: CRISPR or RNAi screens coupled with single-cell RNA-seq or imaging expand phenotypic resolution.

Quick-Fire Revision Points

• ChIP-seq = antibody enrichment of protein-bound DNA ➔ sequencing peaks reveal binding sites.
• Morpholinos = stable antisense oligos that block translation/splicing, widely used in embryo studies.
• RNAi relies on Dicer/RISC; can be triggered by exogenous dsRNA or endogenous shRNA expression.
• Pooled shRNA/RNAi libraries + NGS barcoding enable genome-wide functional screens in a single experiment.
• Always validate hits with secondary assays & consider off-target/compensatory effects.

Self-Check Questions

1. Why might an mRNA show high abundance by RNA-seq but the protein be undetectable?
2. Contrast ChIP-seq and ATAC-seq: what biochemical property is each capturing?
3. Describe two ways to deliver dsRNA into mammalian cells and mention pros/cons of each.
4. How does a scrambled morpholino control help interpret experimental results?
5. In a pooled screen, an shRNA barcode drops $100$-fold after drug treatment. What biological inferences can you draw?

Revisit these questions and the bullet points above until you can answer confidently without referring back to the transcript.