Alternative polyadenylation drives isoform-dependent m6A remodeling during Zika virus infection

This study reveals that Zika virus infection reshapes the host epitranscriptome by inducing noncanonical alternative polyadenylation to generate new RNA isoforms that are preferentially targeted by METTL3, thereby driving isoform-specific m6A remodeling and altering immune-related signaling pathways.

The Gist

Imagine your body's cells are like a massive, bustling library. Inside this library, the books are your genes (DNA), and the stories being read aloud to run the cell are the messages (RNA).

For a long time, scientists thought the library's rules were fixed: once a book was written, the story was set. But recently, we discovered a system of "sticky notes" that can be stuck onto these RNA stories. These sticky notes are called m⁶A. They don't change the words, but they tell the cell how to handle the story: read this faster, throw this away, or keep this safe.

This paper is about what happens when a virus (Zika) invades this library. The virus doesn't just steal books; it hires a team of "editors" to rewrite the sticky notes, changing how the library operates to help the virus survive.

Here is the story of how they figured it out, using some simple analogies:

1. The Virus's Sneaky Trick: Cutting the Books Short

Usually, when a cell makes an RNA message, it adds a long tail to the end (like a bookmark) to decide where the story stops. This is called polyadenylation.

The researchers found that when Zika infects a cell, it tricks the cell's machinery into cutting the stories short. Instead of finishing the whole chapter, the cell stops early.

• The Analogy: Imagine you are reading a novel. The virus grabs the scissors and cuts the book right in the middle, so the story ends abruptly.
• The Result: These "shortened" versions of the stories (isoforms) are different from the original long ones. They have different endings, which means they have different "sticky note" zones available.

2. The Editor's New Favorite Spot

The cell has a specific editor named METTL3. Its job is to walk along the RNA and stick the m⁶A notes on specific spots.

• Before the virus: METTL3 was busy sticking notes on the long, full-length stories.
• After the virus: Because the virus forced the cell to make these new, shortened stories, the "sticky note zones" on the long stories disappeared. But the new, short stories had fresh, empty spots where METTL3 could stick notes.
• The Discovery: The researchers found that METTL3 started flocking to these new, shortened versions of the stories. It was like the editor suddenly finding a whole new section of the library that was previously empty, and now it was covered in sticky notes.

3. The "Scissors" Team (CSTF2 and CSTF2T)

Who told the cell to cut the books short? The researchers found two specific proteins, CSTF2 and CSTF2T, acting like the scissors.

• The Experiment: When the scientists "tied the hands" of these scissors (by blocking them with a drug), the virus couldn't make the short stories anymore.
• The Consequence: Without the short stories, the METTL3 editor had nowhere to go, and the sticky notes (m⁶A) didn't get added. This proved that the virus needs these scissors to change the sticky notes.

4. Why Does This Matter? (The Immune System)

Why would the virus want to change these sticky notes?

• The Target: The researchers found that the stories getting the most new sticky notes were related to the immune system (specifically the JAK/STAT and TGF-β pathways). These are the "alarm systems" that tell the body to fight back.
• The Strategy: By changing the sticky notes on these alarm stories, the virus might be subtly turning the volume down on the alarm or changing how the alarm is interpreted. It's a way of hacking the library's security system without breaking the locks.

The Big Picture

Think of the cell as a house.

1. The Virus breaks in.
2. It doesn't just smash the furniture; it hires a renovation crew (CSTF2/T) to knock down walls and build new, smaller rooms (shortened RNA).
3. Because the rooms are different, the security guards (METTL3) have to stand in different places.
4. This changes how the house functions, specifically making it harder for the house's security system (the immune response) to do its job.

In summary: This paper reveals that Zika virus doesn't just attack the cell directly; it rewrites the cell's instruction manual by forcing it to make "short versions" of its messages. This forces the cell's chemical editors to put new labels on these messages, which helps the virus hide from the immune system. It's a sophisticated game of "cut and paste" that the virus plays to win.

Technical Summary

1. Problem Statement

Viral infections, particularly by Flaviviridae (e.g., Zika virus/ZIKV), reprogram host gene expression to facilitate replication and evade immunity. A key layer of this regulation is N6-methyladenosine (m⁶A), the most abundant internal mRNA modification. While previous studies established that ZIKV infection alters m⁶A levels on specific host transcripts (e.g., RIOK3, CIRBP), the mechanisms driving these transcript-specific changes remained unclear.

• Limitations of prior work: Previous studies relied on meRIP-seq, which lacks single-nucleotide resolution and cannot distinguish between different transcript isoforms of the same gene. Consequently, it was unknown whether m⁶A changes were driven by global shifts in the m⁶A machinery (writers/erasers) or by specific changes in RNA processing (e.g., splicing, polyadenylation) that alter the accessibility of m⁶A sites.
• Core Question: How does ZIKV infection induce selective, site-specific m⁶A remodeling, and what is the role of transcript isoform architecture in this process?

2. Methodology

The authors employed an integrated multi-omics approach to map m⁶A dynamics and METTL3 (the catalytic subunit of the m⁶A writer complex) binding at high resolution.

• Cell Model: Huh7 human hepatoma cells infected with ZIKV (MOI 1, 48 hours post-infection).
• GLORI-seq (Glyoxal and Nitrite-mediated deamination of unmethylated adenosine sequencing):
  • Used to generate a site-specific, stoichiometric map of m⁶A.
  • Allows for the quantification of absolute m⁶A levels at single-nucleotide resolution, distinguishing between methylated and unmethylated adenosines.
  • Compared mock-infected vs. ZIKV-infected samples (n=3 replicates).
• METTL3 RNA Immunoprecipitation Sequencing (RIP-seq):
  • Performed native RIP using an anti-METTL3 antibody to map METTL3 binding across the transcriptome.
  • Dual Analysis: Analyzed data at both the gene-level (aggregating all isoforms) and the transcript-level (isoform-specific) using Salmon and DESeq2. This was crucial for detecting changes in low-abundance isoforms masked by dominant isoforms in gene-level analysis.
• Alternative Polyadenylation (APA) Analysis:
  • Used DaPars on input RNA-seq data to calculate the Percentage of Distal Poly(A) Site Usage Index (PDUI) to quantify 3'UTR shortening/lengthening.
• Functional Validation:
  • siRNA Knockdown: Depleted CSTF2 and CSTF2T (cleavage stimulation factors known to promote noncanonical polyadenylation) and WTAP (m⁶A writer component) to test their roles in isoform switching and m⁶A deposition.
  • RT-qPCR & meRIP-qPCR: Validated isoform expression levels and m⁶A enrichment on specific targets (ALCAM, NHSL1, RRAGD, MXD1).
  • Cross-Virus Analysis: Tested DENV, WNV, and HCV to determine conservation of the mechanism.

3. Key Contributions

1. High-Resolution m⁶A Map: Generated the first comprehensive, site-specific stoichiometric map of m⁶A dynamics during ZIKV infection using GLORI-seq, identifying over 2,000 dynamically regulated sites.
2. Isoform-Specific Mechanism Discovery: Demonstrated that many m⁶A changes are isoform-specific and would be missed by conventional gene-level analyses.
3. Mechanistic Link to APA: Established a causal link between ZIKV-induced alternative polyadenylation (APA) and m⁶A remodeling. Specifically, ZIKV promotes the usage of proximal poly(A) sites (3'UTR shortening) and intronic polyadenylation.
4. Role of CSTF2/T: Identified CSTF2 and CSTF2T as the critical host factors driving the generation of these short, METTL3-targeted isoforms.
5. Immune Pathway Specificity: Revealed that m⁶A remodeling preferentially targets genes involved in JAK/STAT and TGF-β signaling pathways, suggesting a viral strategy to fine-tune antiviral responses.

4. Key Results

• Selective m⁶A Remodeling:

  • ZIKV infection altered ~2% of m⁶A sites (2,004 upregulated, 502 downregulated).
  • Upregulated m⁶A sites were enriched in genes involved in JAK/STAT and TGF-β signaling, as well as the Cul3-RING ubiquitin ligase complex (which ZIKV hijacks to degrade STAT2).
  • Motif analysis (DRACH) showed no significant change in motif preference, suggesting the changes are driven by structural accessibility rather than altered enzyme specificity.

• Isoform-Dependent METTL3 Targeting:

  • Transcript-level METTL3 RIP-seq revealed that 85% of upregulated METTL3 binding events were detected only at the transcript level, not the gene level.
  • Case Studies:
    • NHSL1: ZIKV induces a short isoform where the 4th exon becomes the terminal exon. METTL3 binds strongly to this new 3'UTR but not the long isoform.
    • ALCAM, RRAGD, MXD1: All showed ZIKV-induced shifts toward proximal poly(A) sites (3'UTR shortening), creating new METTL3 binding landscapes.
  • Mechanism: The removal of exon-exon junctions (via intronic polyadenylation) relieves the suppression of m⁶A deposition normally caused by Exon Junction Complexes (EJC), allowing METTL3 to access new DRACH sites.

• CSTF2/T Drives the Phenotype:

  • Depletion of CSTF2 and CSTF2T (but not WTAP) abolished the ZIKV-induced expression of short isoforms (ALCAM-S, NHSL1-S, etc.).
  • Crucially, CSTF2/T double knockdown prevented the increase in m⁶A levels on these transcripts, proving that the m⁶A increase is a consequence of the isoform switch, not a direct effect of viral infection on the methyltransferase.
  • This mechanism is conserved across DENV and WNV infections.

• Functional Implications:

  • The study suggests that ZIKV uses APA to generate short isoforms that are preferentially methylated. These m⁶A marks may subsequently influence the stability or translation of these specific immune-regulatory transcripts, potentially aiding viral evasion of the interferon response.

5. Significance

• Paradigm Shift in Epitranscriptomics: This work moves beyond the "gene-level" view of m⁶A, establishing that transcript architecture (specifically 3'UTR length and exon composition) is a primary determinant of m⁶A deposition.
• Viral Subversion Strategy: It reveals a novel mechanism by which flaviviruses rewire host gene expression: by hijacking host RNA processing factors (CSTF2/T) to generate specific isoforms that are then chemically modified (m⁶A) to alter their fate.
• Therapeutic Insight: Understanding that CSTF2/T activity is central to this viral remodeling suggests that targeting these factors or the APA machinery could disrupt the viral strategy of immune evasion without necessarily targeting the virus directly.
• Methodological Benchmark: The study serves as a proof-of-concept for integrating GLORI-seq (quantitative m⁶A) with transcript-level RIP-seq to resolve complex regulatory layers that single-method approaches miss.