Single-cell CRISPR activation screens in primary B cells discover gene regulatory mechanisms for hundreds of autoimmune risk loci.

This study employs single-cell CRISPR activation screens in primary human B cells to systematically link hundreds of non-coding autoimmune risk loci to their specific target genes, revealing shared regulatory mechanisms and a gain-of-function variant that drives autoimmunity through the transcription factor cREL.

The Gist

Imagine your body's immune system as a highly trained security force. Its job is to spot and eliminate invaders like viruses and bacteria. But sometimes, this security force gets confused and starts attacking the body's own buildings (your tissues). This is what we call autoimmune disease (like Lupus, Crohn's, or Rheumatoid Arthritis).

For years, scientists have known that genetics play a huge role in who gets these diseases. They found thousands of "suspect" spots in our DNA (called GWAS loci) that are linked to these illnesses. However, there was a massive problem: 90% of these suspects were hiding in the "dark matter" of our DNA.

Think of the human genome as a massive library.

• The genes are the actual instruction manuals (books) that tell cells how to build proteins.
• The non-coding regions are the empty shelves, the sticky notes, and the wiring diagrams between the books.

Scientists knew the "bad guys" were hiding in the empty shelves (non-coding regions), but they had no idea which specific book (gene) those shelves were supposed to control. It was like finding a broken light switch in a hallway but having no idea which room's light it turned on.

The Big Breakthrough: SCANDAL

This paper introduces a new tool called SCANDAL (Single Cell Analysis of Non-coding Distal Autoimmune Loci). Think of SCANDAL as a super-powered, high-tech flashlight that can instantly tell you which light switch controls which light, even if the switch is 100 rooms away from the light.

Here is how they did it, broken down into simple steps:

1. The Challenge: The "Uncooperative" Cell

The researchers wanted to test these DNA switches in B-cells (a type of white blood cell that is often the culprit in autoimmune diseases). But B-cells are notoriously difficult to work with in a lab; they are like shy guests who refuse to enter the party (they die or stop working when you try to put new tools inside them).

The Solution: The team built a custom "delivery truck" (using a modified virus) to sneak their tools into these shy cells without scaring them away. They used a system called CRISPR-SAM, which acts like a volume knob. Instead of breaking a gene (which is hard to see if the gene is already quiet), they turned the volume up on the DNA switches to see what happens.

2. The Experiment: The "Massive Switchboard"

They created a library of 763 different DNA switches (the autoimmune risk spots) and turned them on one by one in thousands of individual B-cells. They then watched to see which "lights" (genes) flickered on.

The Results:

• They found 524 connections. They successfully linked 378 of those mysterious DNA switches to specific genes they control.
• Long-distance relationships: They discovered that a switch could be hundreds of thousands of letters away from the gene it controls, skipping over the genes right next to it. It's like a switch in the kitchen turning on the TV in the living room, while ignoring the lamp in the hallway.
• The "Whispering" Genes: Many of the genes they found were ones that usually whisper (are expressed very lowly) or are silent in healthy cells. Previous methods missed these, but SCANDAL's "volume knob" made them loud enough to hear. These included important immune messengers (cytokines) and regulators.

3. The "Domino Effect" (Pleiotropy)

One of the most exciting discoveries was how one small change can cause a chain reaction.

They found a specific genetic variant (a typo in the DNA) associated with Lupus (SLE). This typo acts like a volume booster for a master switch called REL.

• The Analogy: Imagine REL is the "Chief of Security."
• The Lupus typo makes the Chief of Security work overtime (too much REL protein).
• Because the Chief is overactive, he goes around and flips dozens of other switches in the house, many of which are linked to different diseases (like Rheumatoid Arthritis or Type 1 Diabetes).
• The Lesson: This explains why people with one autoimmune disease often have a higher risk of others. It's not just one broken part; it's a master controller that got stuck in the "ON" position, messing up the whole house.

Why This Matters

Before this study, we had a map of "suspect locations" but no idea what they did.

• Old Way: "We know this spot is bad, but we don't know why."
• New Way (SCANDAL): "We know this spot is a switch, and we know it controls this specific gene, which leads to this specific disease mechanism."

This is a huge step forward for medicine. By understanding exactly how these genetic switches work, scientists can now design drugs that don't just treat the symptoms, but fix the specific broken switch or the overactive "Chief of Security" (REL) causing the chaos. It moves us from guessing to knowing, paving the way for better treatments for millions of people with autoimmune diseases.

Technical Summary

1. The Problem

Genome-wide association studies (GWAS) have identified thousands of genetic variants linked to autoimmune diseases. However, over 90% of these risk loci reside in non-coding genomic regions, making it difficult to determine their target genes and underlying molecular mechanisms.

• Key Challenges:
  • Cell Type Specificity: Most functional genomic screens have been performed in immortalized cancer cell lines, which may not reflect the biology of primary immune cells. Primary B cells, a critical cell type in autoimmunity, have historically been refractory to high-throughput viral delivery and CRISPR screening.
  • Detection Limits: Existing CRISPR interference (CRISPRi) screens rely on measuring gene downregulation. This limits the detection of regulatory effects on lowly expressed genes (e.g., cytokines, transcription factors) or genes that are silent in the specific cell type being screened.
  • Complexity: It remains unclear how non-coding variants regulate gene expression over long genomic distances or how they contribute to the pleiotropy (shared risk) observed across diverse autoimmune diseases.

2. Methodology

The authors developed a novel framework called SCANDAL (Single Cell Analysis of Non-coding Distal Autoimmune Loci) to address these challenges.

• Resource Curation: They curated 5,655 fine-mapped risk variants from 31 autoimmune traits (including SLE, Crohn's disease, MS, RA). They prioritized 818 variants overlapping open chromatin regions in Germinal Center (GC) B cells.
• CRISPR-SAM System in Primary B Cells:
  • Overcame the difficulty of transducing primary B cells by engineering a tripartite viral delivery system using RD114-pseudotyped retroviruses (for dCas9-VP64 and MCP-p65-HSF1) and GaLV-pseudotyped lentiviruses (for sgRNAs fused to MS2 loops).
  • Used a CRISPR-Synergistic Activation Mediator (CRISPR-SAM) system to activate (upregulate) transcription at targeted loci, rather than repressing them. This allows for the detection of positive regulatory effects on low-abundance transcripts.
• Single-Cell Multi-Omic Screen:
  • Transduced tonsil-derived GC B cells with a library of 2,403 sgRNAs targeting 763 autoimmune risk loci.
  • Performed single-cell RNA sequencing (scRNA-seq) coupled with antibody-derived tags (ADT) for 137 surface proteins.
  • Analyzed ~201,000 cells to identify cis-regulatory target genes within 500 kb of the perturbed loci using the SCEPTRE statistical framework.
• Validation and Mechanistic Follow-up:
  • Micro-Capture-C (MCC): Validated physical chromatin interactions between risk loci and target promoters.
  • Massively Parallel Reporter Assays (MPRA): Quantified allele-specific effects of risk variants on transcriptional activity in primary B cells.
  • Prime Editing: Used in situ prime editing (PE7) to introduce specific risk alleles into primary B cells to confirm gain-of-function effects on endogenous gene expression.
  • SCAFE: Analyzed transcriptional activity of the enhancers themselves (transcribed CREs or tCREs).

3. Key Contributions

• First High-Throughput CRISPRa Screen in Primary B Cells: Successfully established a protocol for viral delivery and single-cell screening in primary human GC B cells, a cell type previously considered "refractory" to such screens.
• SCANDAL Resource: Generated a comprehensive map linking 378 non-coding risk loci to 524 cis-regulatory target gene effects.
• Discovery of "Gain-of-Function" Mechanisms: Demonstrated that CRISPR activation (CRISPRa) is superior to CRISPRi for detecting regulatory relationships, particularly for lowly expressed genes and genes not normally active in B cells but relevant in disease tissues.
• Elucidation of Pleiotropy: Provided a mechanistic model for how a single risk variant can drive multiple autoimmune diseases through transcription factor networks.

4. Key Results

• Mapping Cis-Regulatory Targets:
  • Identified significant cis-regulatory effects for 49.5% (378/763) of the targeted loci, mapping 524 locus-gene pairs.
  • Successfully detected regulation of lowly expressed genes (e.g., IL10, IL27, EGR2, KLF4) and genes not typically expressed in B cells but relevant in disease tissues (e.g., IGF2 in Type 1 Diabetes).
  • Observed high concordance between mRNA changes and surface protein levels (ADT).
• Long-Range Regulation:
  • Found that many risk loci regulate genes over large distances (>100 kb).
  • Demonstrated "gene skipping," where a risk locus regulates a distal gene while ignoring the nearest gene (e.g., rs67289879 regulating CCND3 98 kb away, skipping TAF8 11 kb away).
  • Validated these long-range interactions using Micro-Capture-C.
• MHC/HLA Regulation:
  • Mapped non-coding risk loci that regulate the expression of MHC class I and II genes (HLA-A, HLA-B, HLA-C, HLA-DRB1), suggesting non-coding variants modulate MHC levels in autoimmunity.
• Co-Regulatory Landscapes:
  • Discovered CRE-CRE co-regulation: Activating one risk locus often increased the transcriptional activity of nearby intergenic enhancers, suggesting cooperative regulatory hubs.
• Mechanism of Pleiotropy (The REL Example):
  • Identified rs1432296 (associated with SLE) as a gain-of-function variant that increases expression of the transcription factor REL (c-REL).
  • Prime editing confirmed that the risk allele increases REL expression.
  • ChIP-seq analysis revealed that c-REL binds to dozens of other autoimmune risk loci and target genes associated with different diseases (e.g., RA, IBD).
  • Conclusion: Dysregulation of a single transcription factor by a non-coding variant can create a downstream network effect, explaining genetic pleiotropy across diverse autoimmune conditions.

5. Significance

• Translational Impact: The study bridges the gap between GWAS statistical associations and biological mechanisms, providing a "dictionary" of non-coding risk variants and their target genes.
• Methodological Advance: Proves that primary human immune cells can be used for large-scale single-cell CRISPR screens, moving beyond cancer cell lines to more physiologically relevant models.
• Conceptual Shift: Highlights that CRISPR activation is a more powerful tool than CRISPRi for mapping enhancer function, especially for low-abundance transcripts.
• Understanding Autoimmunity: Offers a unified model for autoimmune pleiotropy, suggesting that shared risk arises from non-coding variants that dysregulate key transcription factors (like c-REL), which then broadly alter the regulatory landscape of the immune system.

This work provides a foundational resource for the field, enabling the systematic interpretation of non-coding genetic variation in human autoimmunity.