Shared genetic factors between lung function and asthma by age at onset

This study reveals that the shared genetic architecture between asthma and lung function varies by age at onset, with earlier-onset asthma showing stronger genetic correlations and mediated effects on lung function through specific loci on chromosomes 5 and 6 involving genes like SLC22A5 and C5orf56.

The Gist

Imagine your body's immune system is like a massive, complex orchestra. Sometimes, this orchestra gets a little out of tune, leading to asthma (where the airways tighten and inflame) and affecting lung function (how well you can breathe).

For a long time, scientists tried to study asthma as if it were one single song. But this paper suggests that asthma is actually a whole album with different tracks, depending on when the music started playing. Did the symptoms start in childhood (the "early track") or in adulthood (the "late track")?

Here is the story of what this research team discovered, broken down into simple concepts:

1. The "Genetic Playlist" Problem

Scientists knew that asthma runs in families, but they were confused. Why do some people get it as kids and others as adults? And how does this relate to how well their lungs work?

The researchers decided to stop guessing and start using a genetic "sorting machine." Instead of just asking people, "When did you get asthma?", they looked at the DNA of thousands of people to see who naturally grouped together based on their genetic code.

They found that asthma naturally splits into three distinct "genres" or playlists:

• The Early Starters (LT20): People who got asthma before age 20.
• The Middle Starters (20-40): People who got it in their 20s or 30s.
• The Late Starters (GT40): People who got it after age 40.

2. The Shared "DNA Threads"

The team asked: Do these different playlists share any of the same genetic "threads" that also affect how well our lungs work?

Think of your DNA like a giant instruction manual. Some pages in the manual tell your body how to build lungs, and other pages tell your body how to fight allergies. Sometimes, the same page controls both!

They found four specific pages (regions on chromosomes 5, 6, 12, and 17) where the instructions for asthma and lung function were tangled together.

3. The "Early Bird" Effect (The Big Discovery)

Here is the most interesting part: The earlier you get asthma, the more your genetics seem to be the boss of your lung health.

• The Analogy: Imagine two houses.
  • House A (Early Asthma): The foundation was built with a specific type of brick (genetics) that is slightly weak. Because the house was built young, that weak brick is the main reason the roof leaks later in life.
  • House B (Late Asthma): The house was built strong, but over 40 years, the rain and wind (environmental factors like pollution or smoking) wore it down.

The study found that for the "Early Starters," the genetic "weak bricks" were the main reason their lung function dropped. For the "Late Starters," the genetics played a smaller role, and outside factors (like the weather) played a bigger role.

4. The Culprits: The "Trouble-Makers"

The researchers zoomed in on the specific genes causing these issues, like finding the specific musicians in the orchestra who are playing the wrong notes.

• On Chromosome 5: They found genes like SLC22A5 and C5orf56. Think of these as the "fuel managers" and "signal messengers" for your cells. When they are glitchy, they seem to mess up the energy and inflammation processes in the lungs, especially in people who started having asthma as kids.
• On Chromosome 6: They found a gene called BACH2. This is like the "traffic controller" for your immune system. If this controller is confused, it sends the immune system into a panic, causing inflammation that hurts the lungs.

5. Why Does This Matter?

This study is like a detective solving a mystery by realizing that "Asthma" isn't just one thing.

• For Doctors: It suggests that treating a 7-year-old with asthma might require a different genetic approach than treating a 50-year-old. The 7-year-old's problem is deeply rooted in their DNA, while the 50-year-old's might be more about their environment.
• For the Future: By understanding when the asthma started, we can better predict how a person's lungs will function in the future and maybe even find new drugs that fix those specific "glitchy pages" in the instruction manual.

In a nutshell: This paper used a giant genetic map to show that when you get asthma matters. If you get it young, your genes are the main reason your lungs might struggle later. If you get it older, your environment plays a bigger role. It's a crucial step toward personalized medicine, where treatment is tailored to your specific "genetic playlist."

Technical Summary

1. Problem Statement

Asthma is a heterogeneous disease with varying clinical presentations and underlying mechanisms, heavily influenced by the Age-at-Onset (AAO). While previous studies have identified distinct genetic architectures for childhood-onset versus adult-onset asthma, there is no universally accepted, genetically validated categorization for AAO subtypes. Furthermore, the shared genetic etiology between asthma and lung function (specifically FEV1/FVC) remains poorly understood, particularly regarding how this relationship varies across different AAO groups. The study aims to:

• Define genetically homogeneous asthma AAO subgroups using an agnostic approach.
• Investigate the shared genetic architecture between these specific AAO groups and lung function.
• Identify causal variants and biological mechanisms driving the pleiotropic effects of asthma on lung function.

2. Methodology

The study utilized data from the UK Biobank (Discovery: ~366k White British; Replication: ~44k non-British White Europeans).

• AAO Grouping Strategy:

  • Instead of using arbitrary age cutoffs, the authors used genetic correlation ($r_g$) to cluster asthma cases.
  • Initial GWAS were performed for 10-year AAO intervals (≤10, 10–20, 20–30, 30–40, >40).
  • Pairwise genetic correlations were calculated using LD Score Regression (LDSC).
  • An iterative clustering algorithm merged groups with $r_g > 0.8$, resulting in three genetically distinct groups: LT20 (≤20 years), 20–40, and GT40 (>40 years).

• Genetic Analysis Pipeline:

  • GWAS: Performed using REGENIE (linear mixed model approximation) for the three AAO groups and lung function (FEV1/FVC).
  • Pleiotropy Detection: Identified regions significant in at least one AAO group and lung function (excluding the HLA region).
  • Fine-mapping: Used SuSiE (Sum of Single Effects) for univariate fine-mapping and mvSuSiE (multivariate) to identify shared causal variants across phenotypes. Credible sets (CS) with a conditional local false sign rate (clfsr) < 0.01 were generated.
  • Mediation Analysis: Decomposed the effect of lead variants on lung function into direct effects and indirect effects mediated through asthma status, using the R package mediation.
  • Functional Annotation: Conducted eQTL/sQTL analysis using GTEx v8 data (European subset) in relevant tissues (lung, blood, skin, etc.) to link variants to gene expression and splicing.

3. Key Contributions

• Novel AAO Stratification: Demonstrated that genetic correlation analysis is a robust method for defining asthma subtypes, revealing three distinct groups (LT20, 20–40, GT40) with different heritability profiles ($h^2$: 0.397, 0.181, and 0.156, respectively).
• Differential Genetic Architecture: Showed that while genetic correlations between asthma and lung function are similar across AAO groups (~-0.30), the number of shared significant variants and genes decreases as the age of onset increases. This suggests later-onset asthma shares more weak signals or is driven more by non-genetic factors.
• Mechanistic Insight via Mediation: Provided evidence that the impact of shared genetic variants on lung function is more strongly driven by the asthma phenotype in early-onset cases (LT20) compared to later-onset cases.
• Identification of Specific Loci: Pinpointed specific chromosomal regions (Chr5 and Chr6) where colocalization and mediation effects are strongest, identifying specific genes and pathways involved.

4. Key Results

• Genetic Correlations:
  • $r_g$ between LT20 and 20–40 was 0.737; between 20–40 and GT40 was 0.747.
  • All three AAO groups showed negative genetic correlations with lung function (LT20: -0.297; 20–40: -0.302; GT40: -0.323), indicating that genetic risk for asthma is associated with reduced lung function.
• Shared Variants:
  • Out of 18 replicated regions, four independent regions (Chr5, Chr6, Chr12, Chr17) were identified.
  • Chr5 and Chr6 showed strong evidence of colocalization across phenotypes in multivariate fine-mapping.
  • Chr12 and Chr17 showed no evidence of colocalization between AAO groups and lung function.
• Mediation Analysis (The "Driver" Effect):
  • For variants on Chr5 (rs7713065) and Chr6 (rs6925032), the proportion of the effect on lung function mediated by asthma was highest for the LT20 group and decreased for 20–40 and GT40.
  • Example (rs7713065): Proportion mediated was 12.2% for LT20, 8.9% for 20–40, and 6.4% for GT40. This implies that for early-onset asthma, the genetic risk acts more directly through the asthma pathway to reduce lung function.
• Functional Genomics (QTLs):
  • Chr5 (rs7713065): Acts as a cis-eQTL for SLC22A5 and cis-sQTL for C5orf56 and P4HA2 in lung and blood tissues. SLC22A5 is involved in fatty acid metabolism and energy production; P4HA2 is involved in collagen biosynthesis.
  • Chr6 (rs6925032): Acts as a cis-eQTL for BACH2 in the pancreas. BACH2 is a transcription factor critical for T- and B-lymphocyte differentiation and Th2-mediated inflammation.

5. Significance and Conclusion

• Pathophysiological Insight: The study suggests that the genetic link between asthma and reduced lung function is most potent in early-onset asthma, likely driven by immune-mediated mechanisms (e.g., BACH2, SLC22A5). In contrast, later-onset asthma may share fewer strong genetic signals with lung function, potentially indicating a greater role for environmental factors (e.g., pollution, occupational exposures) in the latter.
• Methodological Advancement: The use of genetic correlation to define AAO subgroups offers a more biologically grounded approach than arbitrary age cutoffs, which can be applied to other complex traits with variable onset ages (e.g., Type 2 Diabetes).
• Clinical Implications: Identifying that early-onset asthma is more genetically driven in its impact on lung function supports the need for early intervention and genetic screening in pediatric populations to prevent long-term lung function decline.
• Limitations: The study is limited to individuals of European ancestry (UK Biobank), relies on self-reported AAO (which may be subject to recall bias), and QTL analysis was restricted to cis-variants due to sample size constraints in GTEx.

In summary, this paper provides a high-resolution map of the shared genetics between asthma and lung function, revealing that early-onset asthma is the primary genetic driver of the comorbidity between asthma and reduced lung function, mediated through specific immune and metabolic pathways.