To prioritize genetic effects on immune responses, the authors analyzed surface marker data generated by CITE-seq from peripheral blood mononuclear cells. The dataset comprised 1,055,857 PBMCs collected from 356 individuals. From those multimodal single-cell profiles, they defined genetic associations to 148 surface proteins across eight immune cell types. The approach focused the genetic search space on measurable cell-surface phenotypes relevant to immunity and autoimmune disease.
Among the protein quantitative trait loci (pQTL) discovered, a signal in the CD40 locus stood out because of prior implication in rheumatoid arthritis (RA) and other autoimmune conditions. The reported RA risk variants at this locus were associated with an approximately 20% increase in CD40 protein expression on B cells. Notably, this change in protein abundance occurred with minimal corresponding effects on CD40 mRNA levels, indicating a discordance between the protein-level association and steady-state transcript abundance in the sampled PBMCs.
To move from association to causality and to resolve the specific causal allele, the authors deployed base-resolution genome editing combined with single-cell multimodal readout, an approach they refer to as CRAFT-seq. CRAFT-seq captures the genomic DNA sequence at the edited site and measures multimodal cellular phenotypes at single-cell resolution. Using this platform, they defined a single causal allele within the CD40 locus: rs1883832, which lies in the Kozak motif upstream of the CD40 coding sequence.
The team edited the identified causal allele, rs1883832, in primary B cells and performed CRAFT-seq profiling. This experiment linked the precise genomic change to downstream cellular phenotypes by both capturing the edited genomic sequence and profiling multimodal phenotypes in the same single cells. The editing experiments in primary B cells enabled direct testing of downstream effects that might not be resolvable from population-scale observational cohorts alone.
When rs1883832 was edited in primary B cells and assayed with CRAFT-seq, the authors observed trans-effects on the expression of more than 200 genes. Crucially, these trans-effects were context-specific: they occurred only in a light zone germinal center–like state of B cells. The same broad trans-effects were not detected in population-scale cohorts of unstimulated B cells, emphasizing that the downstream consequences of the causal allele depended on cellular activation or differentiation context. This finding demonstrates that causal genetic variation can produce sizable and specific trans-regulatory effects that are invisible in resting or heterogeneous population samples.
The study outlines a workflow from large-scale multimodal single-cell genetic mapping to base-resolution editing and single-cell multimodal phenotyping. By combining CITE-seq association mapping with CRAFT-seq editing and readout, the authors moved from locus-level association to a single causal base change (rs1883832) and revealed its downstream, context-dependent trans-effects in B cells. The work illustrates how defining causal genetic variation experimentally can reveal regulatory consequences that population studies of unstimulated cells may miss.
This report is a preprint and has not been peer reviewed. Funders declared include NIH Common Fund grants listed in the source. Competing interest disclosures were reported for specific authors in the original preprint.
Large-scale single-cell CITE-seq enabled discovery of a CD40 pQTL associated with RA risk and increased CD40 protein on B cells.
CRAFT-seq genome editing established rs1883832 in the Kozak motif as the causal allele at the CD40 locus.
Editing rs1883832 in primary B cells produced context-specific trans-effects on >200 genes, restricted to a light zone germinal center–like state and not evident in unstimulated population datasets.
The combined mapping-and-editing framework demonstrates a path to identify causal alleles and their context-dependent regulatory networks, highlighting the importance of cellular state in revealing genetic effects.