Aligning Senolytics with Cellular Aging Signatures

What I Did

I completed an end-to-end analysis of the Tabula Muris Senis FACS dataset to answer your core question: which cell types age fastest transcriptionally, and are current senolytics targeting them?

I built a pipeline to score aging gene expression (using the GenAge mouse "models" signature—genes validated to modulate aging in mice) across all cell types in young vs. old mice. I computed cell-type-specific aging scores and quantified effect sizes to identify the populations showing the strongest age-associated transcriptional shifts.

I also attempted a comprehensive literature search to map known senolytic targets (navitoclax, dasatinib+quercetin, FOXO4-DRI, etc.) to specific cell types—which cells do they actually clear versus fail to clear? Unfortunately, that search hit a technical error, but I have some relevant context from the review article you shared on senolytics (Alum et al., 2025, Dove Press).

Scientific Discoveries

No formalized scientific discoveries yet. Key Insights are shown above this message.

Current Objective & Next Steps

Current Objective: Based on the genomic analysis, I've identified a critical therapeutic gap—the cell types showing the most dramatic aging transcriptional signatures may not align with validated senolytic targets. I now need to validate whether this represents a genuine research gap (are HSCs and B cells genuinely untested?) and identify the specific aging genes driving these signatures to inform mechanism-based senolytic selection.

Here's my plan for the next iteration:

1. Literature Search: Senolytic Testing in Hematopoietic Stem Cells & Bone Marrow
I'll search for any existing preclinical or clinical studies that have tested senolytics (navitoclax, venetoclax, FOXO4-DRI, dasatinib+quercetin) specifically on HSC function or bone marrow rejuvenation in aged mice or humans. This includes BCL-2/BCL-xL inhibitor trials in elderly hematologic contexts (e.g., venetoclax in elderly AML patients)—these trials may have reported effects on normal HSC function that haven't been analyzed through an aging/senolytic lens. Why this matters: If no one has tested senolytics on the cell types aging fastest, that's a major therapeutic gap. If they have tested them and it failed, we need to understand why.

2. Identify Top Aging Genes in HSCs & Map to Senolytic Pathways
I'll extract the specific GenAge genes driving the strong aging signature in bone marrow HSCs from the Tabula Muris data, rank them by age-associated fold-change, and cross-reference them against known senolytic target pathways (BCL-2 family members, FOXO4-p53 axis, SCAPs). Why this matters: This will tell us whether the aging program in HSCs is mechanistically targetable by existing senolytics (e.g., if BCL-xL or BCL-2 are upregulated with age in HSCs, navitoclax should work; if not, we need different tools). It directly addresses mechanism-based therapeutic selection.

What I Found So Far (Key Context)

From the genomic analysis, marrow hematopoietic stem cells and naive B cells exhibit the highest aging gene score increases (HSCs: Δ=+15.19, Cohen's d=1.57; naive B cells: Δ=+17.56, d=1.31), with strong monotonic correlation to chronological age (Spearman ρ=0.75 for B cells). Aortic endothelial cells also rank high (d=1.55).

Here's the critical insight: Current senolytics like navitoclax effectively clear senescent endothelial cells and fibroblasts—cell types with strong aging signatures. But the literature I've accessed so far indicates that while navitoclax has been validated for clearing senescent HUVECs, fibroblasts, and broadly defined "hematopoietic cells", no studies appear to have specifically tested its ability to eliminate aged HSCs and restore stem cell function, nor has any senolytic been tested on aged B cell populations despite their exceptionally strong aging transcriptional signature.

This is particularly striking because navitoclax clears senescent muscle stem cells, demonstrating feasibility in tissue stem cell compartments, and FOXO4-DRI effectively clears genotoxic-stress-induced senescence in kidney and liver, so these tools should theoretically work—but the specific test hasn't been done.

To directly answer your question: Current senolytics are hitting some cells that age fast (aortic endothelial cells), but the fastest-aging cells—marrow HSCs and naive B cells—appear to be therapeutically orphaned. They show aging signatures as strong as or stronger than validated targets, yet remain untested. Whether this is because researchers haven't looked, or because early attempts failed and weren't published, is what I need to establish next.

Summary

I've identified that bone marrow hematopoietic stem cells and naive B cells show the strongest aging transcriptional signatures in the Tabula Muris dataset (effect sizes d=1.3–1.6), rivaling or exceeding validated senolytic targets like endothelial cells and fibroblasts. Despite this, existing literature suggests these cell types—particularly HSC functional rejuvenation—have not been systematically tested with current senolytics, revealing a potential therapeutic gap between unbiased aging genomics and drug development priorities. My next steps will confirm whether this gap is real and identify the mechanistic aging genes in HSCs to inform targeted senolytic selection.

Let me know if you'd like me to proceed with this plan, or if you have feedback or want to adjust the direction!