Biochemistry Focus ECR webinar – The complex world of RNA modification and replication

Main ideas & lessons conveyed

Webinar purpose and structure

• This is an early-career researcher (ECR) webinar in the Biochemical Society / Biochemistry Focus Webinar Series, sponsored by the Biochemical Society and Portland Press.
• The session is titled “The complex world of RNA modification and replication.”
• Viewers are encouraged to submit questions during the webinar via a screen-based question box; questions are addressed toward the end.

---

Speaker 1: Cara F. (RWTH Aachen University) — ADP-ribosylation of RNA (“ADP-ribose RNA capping”)

Core concept introduced

• ADP-ribosylation is a known protein modification discovered in the 1960s.
• Mechanism (protein-based background):
  • ADP-ribosyltransferases (human PARP family, e.g., “PARP” enzymes) use NAD to transfer ADP-ribose to substrate molecules.
  • Substrates can receive:
    • a single ADP-ribose, or
    • poly-ADP-ribose (PAR) chains.
  • The process is reversible via dedicated hydrolases:
    • enzymes that reverse PAR chains
    • enzymes that remove single ADP-ribose units to return substrates to baseline.
• Extension to RNA:
  • RNAs can also be substrates for ADP-ribosylation.
  • The talk focuses on a model where RNA receives an ADP-ribose “cap-like” structure (conceptually analogous to other RNA caps).

Why the speakers investigated it

• Two key studies (2018 and 2019) reported that:
  • yeast enzymes (e.g., Tpt1) could modify RNA/DNA,
  • human PARP family enzymes plus other enzymes could modify RNA.
• Initial “lab debate”:
  • the lab primarily focused on protein modification, so RNA modification was unexpected.
• Motivation:
  • reproduce and clarify what others observed
  • determine whether RNA modification is detectable in cells, not only in vitro.

Methodology / experimental workflow

In vitro ADP-ribosylation assays

• Starting point from literature:
  • incubate RNA (or oligonucleotides) with ADP-ribose transferases and radio-labeled NAD/ADP-ribose
  • detect incorporation by autoradiography.
• Adapted approach (no radioactivity):
  • use an RNA oligo with defined chemistry at the 3’ end (described as a 3’-phosphate / “OLe/All” approach depending on substrate design)
  • infer modification efficiency from visualizable gel shifts.

Assay substrate requirements

• Activity depends on oligo design:
  • efficient modification when oligos start with adenine/guanine
  • less/none when starting nucleotide differs
• This helped reconcile why different papers reported different apparent activity levels:
  • different oligo starting nucleotides affect observed efficiency.

Reversibility controls

• Demonstrated reversibility:
  • ADP-ribosylated RNA can be reversed by adding specific hydrolases
  • resulting RNA returns to an unmodified gel band state.

Detection in cells: antibody-based method (initial attempt)

• Slot blotting was used to detect ADP-ribose–modified RNA extracted from cells.
• A commercially available antibody was used.
• Early result:
  • little/no signal compared with wild-type controls.
• Hypotheses tested:
  • transferases may be insufficiently active in the system, and/or
  • hydrolases remove the modification quickly.
• Strategy:
  • overexpress components that might generate modification
  • block reversal by using a cocktail of siRNAs targeting different ADP-ribose hydrolases.
• Result:
  • stronger ADP-ribosylated RNA signal when reversal is prevented.

Cellular induction by stress

• Stress can induce detectable RNA ADP-ribosylation, including:
  • interferon alpha
  • EBSS (energy stress / starvation)
• Certain RNAs (e.g., specific mRNAs) may become ADP-ribosylated.

Complementary validation: LC-MS approach

• Because antibodies alone are not fully definitive, an LC-MS-based pipeline was developed.
• High-level pipeline:
  • generate ADP-ribose–modified RNA standards by enzymatically generating individual nucleotides with ADP-ribose
  • run extracted RNA from cells (including knockdown conditions removing key hydrolases)
  • detect modified nucleotide signatures by LC-MS.
• Result:
  • early signals support that modification exists in cells (quantitation still developing).

Function tests: translation and stability

Does the ADP-ribose “cap” enable translation?

• Compared translation for RNAs with:
  • no cap (control)
  • canonical m7G cap
  • ADP-ribose cap
• Findings:
  • non-capped / ADP-ribose-capped RNAs did not show canonical translation
  • m7G-capped RNAs translated as expected
• Conclusion confirmed using an in vitro translation system.

Is it a degradation/dependency signal?

• Treated ADP-ribose-capped RNAs with exonuclease/degradation machinery and hydrolases.
• Findings:
  • ADP-ribosylated RNA remained stable
  • unlike unmodified RNA
• Implication:
  • ADP-ribose capping does not primarily act as a degradation trigger (unlike behaviors attributed to NAD-capped RNA).

Identifying reversal (“de-capping”) enzymes

• Tested candidate enzymes that reverse ADP-ribosylation.
• Findings:
  • some enzymes involved in related decapping/reversal pathways can decap ADP-ribosylated RNA in vitro.
• Key insight:
  • multiple hydrolase domains can reverse the RNA modification
  • PARG-like and macro-domain involvement were highlighted.

Mechanistic comparison with NAD capping

• Conceptual parallel proposed:
  • NAD cap
    • introduced early/transcription-associated
    • blocks translation
    • marks RNA for degradation
  • ADP-ribose cap
    • introduced post-transcriptionally
    • blocks canonical translation
    • appears to protect RNA from degradation
    • reversed by RNA ADP-ribose hydrolases (notably PARP/PARG-related enzymes)

Central conclusions from Cara’s talk

• The modification likely exists in living cells, supported by:
  • antibody detection with reversal-blocking strategies
  • emerging LC-MS evidence
• It appears stress-induced and dynamic.
• The “ADP-ribose cap”:
  • likely does not enable translation
  • likely does not mainly promote degradation
  • instead seems to stabilize RNA under certain contexts
• Next functional priority:
  • determine which RNAs get modified (transcript specificity)
• Next technological priority:
  • develop a pipeline to identify modified RNAs genome-wide.

---

RNA identification / sequencing pipeline (detailed bullets)

• Goal: identify which RNAs carry the ADP-ribose cap under conditions like EBSS.
• Workflow described:
  1. Isolate total RNA (mixture of capped and uncapped forms).
  2. Remove free phosphates using a phosphatase step.
  3. Reveal the cap-associated phosphate using a macro/PARG-related hydrolase step.
  4. Use the newly exposed phosphate to attach adapters / enable library preparation.
  5. Sequence to identify modified RNA species.
• Observations so far:
  • initial sequencing produced ~150 hits
  • optimization increased to ~500 hits in a second run
  • preliminary transcript categories align with stress biology (energy starvation / mitochondrial electron transport association)

Take-home summary (Cara)

• RNA ADP-ribosylation (“ADP-ribose capping”) is real, stress-linked, reversible, and appears to protect RNA while blocking canonical translation.
• Identifying RNA targets is essential to determine biological function and disease relevance (e.g., cancer, neurodegeneration, autoimmunity; PARP dysregulation context).

---

Speaker 2: Ari Williams (Bernhard Nocht Institute / Center for Structural Systems Biology) — Cap snatching and transcription initiation by negative-strand RNA viruses

Core concept introduced: cap snatching in negative-sense RNA viruses

• Negative-sense RNA viruses lack mRNA capping activity and rely on:
  • cap snatching: stealing cap fragments from host mRNAs to initiate viral RNA transcription.
• Mechanism:
  • Viral L protein contains multiple essential enzymatic and binding domains.
  • L binds host mRNA cap fragments, then an endonuclease cleaves downstream.
  • The cap fragment is incorporated into viral transcripts, producing chimeric viral mRNAs that recruit host translation machinery.

Target system and biological background

• Focus virus: SFTSV (severe fever with thrombocytopenia syndrome virus).
• Genome organization:
  • tri-segment genome: S / M / L
  • L segment encodes the L protein, including RNA-dependent RNA polymerase activity.
• L protein functional architecture:
  • N-terminal endonuclease
  • central RNA-dependent RNA polymerase (RdRP) core
  • C-terminal cap-binding domain (CBD)

Methodology / structural biology approach

• Primary technique: single-particle cryo-electron microscopy (cryo-EM).
• General cryo-EM workflow (conceptual):
  • express and purify L protein
  • incubate with designed RNA substrates/primers
  • collect grids (collaborators in Hamburg mentioned)
  • acquire micrographs
  • computational processing:
    • particle picking
    • 2D classification
    • 3D reconstruction
    • fit atomic models into density maps
• Stepwise structural strategy to capture initiation vs elongation:
  • no NTPs during initiation/priming state to prevent elongation
  • use short RNAs corresponding to genome ends:
    • a short 5’ RNA and 3’ RNA (template/primer elements)
    • include cap RNA fragments and relevant primer components
• Two principal cryo-EM targets:
  1. Transcription initiation / priming state
     • L protein + RNA components (5’ and 3’ ends + cap fragment/primer elements)
     • structure at ~3.3–3.4 Å reported
  2. Early elongation state
     • include NTPs with a non-hydrolyzable UTP analog to stall after incorporation
     • structure resolves early product with stalling

Key structural findings

Initiation / priming state

• L protein binds the 5’ RNA at a defined site via a distinctive “hook-like” corkscrew duplex architecture involving the template.
• Cap RNA binding occurs at two locations:
  • the 5’ cap end binds the cap-binding domain (CBD)
  • the other end base-pairs with the template near the RdRP core
• RNA recognition logic:
  • interactions appear largely phosphate-based rather than strongly sequence-specific in observed segments
  • many coordinating residues come from regions of L outside the CBD
  • implication:
    • studying isolated CBD constructs may miss crucial coordination residues

Priming → early elongation changes

• Comparative observation:
  • overall coarse architecture may appear similar
  • but domain positions shift substantially as transcription progresses
• Two-step conformational model proposed:
  1. CBD folds outward from the RdRP core region (to a different position seen in later-stage structure)
  2. endonuclease folds inward toward the CBD position
• Late-stage elongation structures from prior work help anchor domain transition modeling.

Conclusions and outlook (Ari)

• Cap-snatching coordination requires multi-domain rearrangement, not only localized cap binding.
• Apparent lack of strict sequence specificity at the CBD suggests:
  • selection may be driven more by downstream base pairing with template/RdRP
  • rather than CBD sequence preference.
• Next structural/biological goal:
  • understand assembly of L, RdRP, and nucleoprotein (NP) into viral ribonucleoprotein complexes (vRNPs) in infected cells
  • proposed methods:
    • single-particle cryo-EM of complexes
    • cryo-electron tomography for within-cell visualization

Drug discovery relevance mentioned

• Structural data can clarify:
  • the precise binding modality/location for cap RNA
  • RNA-dependent conformational states
  • protein-protein interfaces unique to transcription/replication
• Potential intervention strategies:
  • target RNA-binding pockets
  • or target state-specific conformational interfaces required for cap snatching/transcription

---

Question-and-answer themes (high-level)

• Cara (RNA ADP-ribosylation):

  • differences between transcriptional (NAD cap) vs post-transcriptional (ADP-ribose cap) introduction
  • quantitative abundance: likely low abundance (few percent tops), with some transcripts showing higher modification
  • mono vs poly ADP-ribosylation: observed so far primarily mono (poly not fully assessed)
  • next steps: identify modified RNAs, link phenotypes to RNA vs protein ADP-ribosylation, and assess disease relevance (cancer/neuro/autoimmune)

• Ari (cap snatching):

  • whether cap snatching depends on host factors: structural data supports cap requirement more than strict host targeting; selection may rely on base pairing and processivity vs CBD sequence specificity
  • spread across viruses: likely widespread among negative-sense RNA viruses lacking capping activity, but may differ mechanistically by family
  • drug targeting challenge: structural mechanisms are needed given limited virus-specific drugs
  • host receptor specificity for entry: not addressed definitively; would consult colleagues
  • assay differences for ADP-ribosylation signals: suggested importance of substrate features such as single-stranded RNA overhangs

---

Speakers / sources featured (as named in the subtitles)

Speakers (main presenters)

• Dr. Mikel T (host/moderator; webinar introduction)
• Cara F. (first invited speaker; RWTH Aachen University)
• Ari Williams (second invited speaker; Bernhard Nocht Institute for Tropical Medicine / Center for Structural Systems Biology)

Cited / referenced external groups or papers (mentioned in subtitles)

• Schuman lab (2018 paper on yeast enzyme activity transferring ADP-ribose to RNA/DNA)
• Arap and Ox (2019 paper(s) on human PARP enzymes modifying RNA)
• Mention of HeLa cells for RNA extraction
• Mention of biological stimuli: EBSS, interferon alpha
• Mention of enzyme/cap analog context: m7G cap, NAD cap, PARP/PARG-like hydrolases, macro-domain hydrolases
• Mention of additional virus context: influenza and other negative-strand RNA viruses (referenced indirectly via colleagues)
• Named collaborators/people credited at the end of Ari’s talk:
  • Carola
  • Dominic
  • Morin
  • Sigy
• Mentioned institution in Ari’s workflow:
  • CSSB in Hamburg (collaborators for cryo-EM grid/data collection)