Visual summary (plots reproduce key claims from the paper)

Data plotted are taken directly from the preprint’s overlap analysis (HeLa RNA-seq + combined RelA/RelB/cRel siRNA comparisons) and illustrate the central quantitative claim: NF-κB dependency is larger for hypoxia-repressed genes than for induced genes (Primary preprint overlap stats)

Paper reports RelA depletion produced the largest transcriptional change in hypoxia, but RelB accounted for the highest proportion of NF-κB-dependent hypoxia-regulated genes across contexts (authors highlight RelB importance is underappreciated). See validation qPCR panels in multiple cell lines (Rel-subunit contribution & qPCR)

Concise critical appraisal (evidence-weighted)

• Strength — scope and data integration: The study combines new HeLa RNA-seq (siRelA/RelB/cRel) with re-analysis of multiple public hypoxia RNA-seq datasets across cell types, supported by qPCR and protein/ROS assays; that multi-dataset view increases external validity beyond a single cell line (Integrated RNA-seq + validations)
• Limit — replication and sample size: RNA-seq reported n=2 biological replicates for the HeLa sequencing; this is minimal for robust differential-expression inference and increases false-discovery risk even with DESeq2; independent replication or increased n would strengthen claims (RNA-seq sample size note)
• Limit — direct mechanistic proof: The paper uses ChIP-Atlas overlaps and motif enrichment to suggest direct NF-κB and HIF co-regulation, but does not present NF-κB ChIP/CUT&RUN performed under hypoxia in the same cells. Therefore direct binding under hypoxia remains inferred rather than demonstrated (ChIP-Atlas inference)
• Context and consistency with prior mechanistic work: Prior mechanistic studies show IKK/TAK1-dependent NF-κB activation in hypoxia and roles for sumoylated IκBα and PHD enzymes; the present transcriptome-level result is consistent with NF-κB contributing to hypoxia responses but expands the influence to repression of OxPhos genes and ROS regulation (Culver et al. mechanism)
• Functional link to mitochondria/ROS — plausible but needs deeper mechanism: Authors show NF-κB depletion reduces hypoxia-induced ROS (CellROX) and reverses OxPhos protein changes — consistent with literature connecting NF-κB to mitochondrial function and OxPhos, but causality (e.g., direct transcriptional repression vs secondary effects via HIF or metabolic shifts) remains to be fully mapped (ROS & OxPhos evidence)

Where the paper could mislead / blindspots

1. n=2 RNA-seq increases chance of both false positives and unstable fold-change estimates; many claims are validated by qPCR which reduces, but does not eliminate, this concern.
2. Cell-line panel (HeLa,A549,U87,MCF7) covers diverse cancers but remains an in vitro cancer-cell-centric sample; stromal, immune or in vivo tissues may behave differently.
3. Use of ChIP-Atlas implies historical ChIP experiments (from varied labs/conditions) — enrichment is informative but not a substitute for same-condition NF-κB hypoxia ChIP/CUT&RUN in matched samples.
4. Interpretation of repression: NF-κB can repress directly (p50 homodimers/HDAC recruitment) or indirectly (sustaining HIF levels or altering metabolism); distinguishing these routes requires targeted experiments (see suggested experiments below).

Recommended experiments to strengthen/falsify claims (concise)

• Hypoxia NF-κB CUT&RUN/ChIP-seq (RelA/RelB/p50): performed in HeLa (and one other validated cell line) ± hypoxia ± siRel to map direct binding under study conditions; would prove direct binding at candidate promoters (discriminates direct vs indirect regulation) (Authors).
• Rescue experiments: re-express NF-κB subunits (siRNA-resistant) to show reversal of RNA and protein phenotypes; use p50 homodimer/RelA mutants to test repression vs activation roles.
• Mitochondrial functional assays: measure OCR/ECAR (Seahorse) in NF-κB depleted vs control under hypoxia to demonstrate functional OxPhos changes beyond protein levels, and measure mitochondrial ROS by ratiometric probes (MitoSOX + careful controls) (OxPhos/ROS limitations).

Conclusions & confidence

Conclusion: The paper provides persuasive, multi-modal evidence that NF-κB contributes substantially to the transcriptional response to hypoxia — particularly to repression of OxPhos/ROS-related genes — and highlights a larger role for RelB than commonly appreciated. The strongest claims (percentages of NF-κB-dependent DEGs) rest on modest RNA-seq replication but are supported by cross-dataset overlap and targeted qPCR/protein/ROS validation. Confidence is moderate: conclusions are plausible and supported but would be materially strengthened by additional RNA-seq replication, hypoxia-specific NF-κB chromatin mapping, and mitochondrial functional assays (Primary preprint summarizing support & limits).

Quick reproducibility checklist for a replication team

1. Obtain GEO GSE260616 raw fastq; re-run STAR alignment (hg38), featureCounts, DESeq2 with stated thresholds (|log2FC|>0.58, FDR<0.05) — verify DEG counts reported (Methods and data access).
2. Reproduce qPCR panels (use primers in Supplementary Table S1) and immunoblots (same antibodies) in HeLa and one other validated cell line.
3. Perform CUT&RUN for RelA/RelB under normoxia vs 1% O2 to identify direct NF-κB hypoxia targets.

Selected literature context (annotated)

• Culver et al. (2010) mechanism
• Rius et al. (2008) NF-κB→HIF-1α
• Primary Rocha lab preprint

Practical evaluation metrics (paper-scored)

How to improve this review / next steps

To evolve this analysis: re-run HeLa GEO dataset (GSE260616) differential-expression with increased replicates (if available), and run hypoxia-specific NF-κB CUT&RUN in the same samples; request raw counts and QC metrics from authors to recalculate DEG stability.

Buttons — Author Reviews