Fine Structural Features of Complex InDels and NHEJ Repair at Naturally Occurring Damage Sites in Normal Human Colon Crypts

This study utilizes a novel whole-genome sequencing method on single human colon crypts to analyze complex insertion-deletion events repaired by non-homologous end joining, thereby providing insights into the mechanisms of naturally occurring DNA damage repair within a physiologically relevant context.

The Gist

The Big Picture: Fixing the "Typos" in Our Body's Instruction Manual

Imagine your body is a massive library, and every single cell is a copy of the same instruction manual (your DNA). Over time, these manuals get damaged. Sunlight, pollution, and just the natural aging process can cause "typos" or torn pages in the text.

Usually, cells have a repair crew that fixes these tears. The most common repair method in our cells is called NHEJ (Non-Homologous End Joining). Think of NHEJ as a "duct tape" repair. When a page is ripped, the repair crew doesn't always have the perfect matching piece of paper to glue back in. Instead, they just tape the two torn edges together. Sometimes, this leaves a little bit of extra tape (an insertion) or cuts off a tiny bit of text (a deletion). These messy repairs are called complex indels.

The Problem: Finding the Needle in a Haystack

For a long time, scientists could study how this "duct tape" repair worked in test tubes or in petri dishes with cancer cells. But they couldn't easily see how it worked in healthy, normal people inside their actual bodies.

Why? Because a normal tissue sample (like a piece of colon) is like a giant crowd of people. If one person in the crowd has a typo in their manual, but 1,000 others don't, standard DNA tests just see the "average" and miss the typo entirely. It's like trying to hear one person whisper in a stadium full of cheering fans.

The Breakthrough: Zooming In on Single Families

The researchers in this paper used a clever trick. They looked at colon crypts. Imagine the lining of your colon as a field of tiny, independent villages. Each village (crypt) is populated by a single family of cells that all came from one ancestor. Because they are a "clone family," they all share the same history.

By sequencing the DNA of these single "villages" (crypts) one by one, the scientists could finally hear the whispers. They found the specific "typos" that happened naturally in healthy people as they aged.

What They Found

1. Aging is like a slow leak.
As people get older, the number of these "duct tape" repairs (NHEJ events) in their healthy colon cells slowly increases. It's like the instruction manuals are getting more and more worn out over time. The study found that for every year of life, a healthy colon cell accumulates about 0.19 of these repair scars.

2. Radiation is like a hurricane.
The study looked at patients who had cancer treatment.

• Chemotherapy alone was like a light rain; it caused a few more typos than usual, but not a huge amount.
• Radiation combined with chemo was like a hurricane. It caused a massive spike in DNA damage. The healthy colon cells of these patients had about 5 times more repair scars than people who hadn't had treatment.
• Crucial finding: Even though radiation caused more damage, it didn't change how the cells fixed it. The cells still used the same "duct tape" method; they just had a lot more tears to fix.

3. The "Glue" has a favorite flavor.
When the cells add new letters to fix the tear, they don't pick them randomly. The study found that the repair enzymes have a strong preference for adding Adenine (A) and Thymine (T) letters. It's as if the repair crew always reaches for the red and blue duct tape, ignoring the green and yellow, even when the green and yellow would fit better.

4. They rarely use "matching patterns."
Sometimes, if two torn edges have a tiny matching pattern (microhomology), the repair crew uses it to line things up perfectly. But the study found that in healthy human cells, the crew rarely waits for a match. They just tape the ends together immediately, even if the patterns don't match. This confirms that the "duct tape" method is the dominant repair strategy in our bodies, not the more precise "matching" method.

Why This Matters

This paper is a big deal because it moves DNA repair research out of the artificial lab and into real human biology.

• Before: We knew how repair worked in test tubes.
• Now: We know how it actually works in a 70-year-old human being.

This helps us understand how aging happens at a molecular level and gives us a baseline to measure how dangerous things like radiation really are for normal, healthy tissue. It tells us that while our bodies are good at fixing damage, the scars pile up over a lifetime, and radiation can accelerate that process significantly.

Technical Summary

1. Problem Statement

• Limitation of Existing Models: Current understanding of DNA repair mechanisms (specifically Non-Homologous End Joining, or NHEJ) is largely derived from in vitro biochemical assays or cell culture systems using transfected DNA. These models often rely on artificial DNA ends or specific selection pressures that may not reflect the physiological reality of DNA repair in intact human tissues.
• Detection Barrier: In normal multicellular organisms, cellular heterogeneity makes it difficult to detect replication-independent DNA damage sites. Standard Whole Genome Sequencing (WGS) of bulk tissue yields a consensus sequence, masking rare somatic mutations present in only a subset of cells.
• Algorithmic Blind Spots: Standard somatic variant callers (e.g., Mutect2, Strelka) are optimized for simple Single Nucleotide Variants (SNVs) and simple indels. They frequently fail to identify complex indels (simultaneous deletions and insertions with misincorporations) characteristic of NHEJ repair, often splitting them into multiple calls or missing them entirely.

2. Methodology

The authors employed a novel approach combining high-resolution tissue sampling with customized bioinformatics to analyze naturally occurring DNA damage in human colon crypts.

• Sample Collection & Sequencing:
  • Source: Normal colon tissue from 21 individuals (ages 10 months to 90 years).
  • Unit of Analysis: Single colon crypts (harboring natural cell clones), avoiding the need for whole-genome amplification (WGA) which introduces artifacts.
  • Scale: 106 single crypt libraries and 21 bulk control libraries sequenced using Illumina NovaSeq 6000 (high-depth WGS).
• Variant Calling Strategy:
  • Tool Selection: The authors tested multiple callers (Mutect2, Strelka, HaplotypeCaller, Freebayes) using a simulated dataset of complex indels. Freebayes was identified as the superior tool for calling complex indels as a single variant event.
  • Custom Filtering Algorithm: A multi-step algorithm was developed to filter Freebayes VCFs:
    1. Removal of germline variants (RS SNPs) and low-quality calls.
    2. Exclusion of simple indels, slippage events (mononucleotide/dinucleotide repeats), and variants present in bulk controls.
    3. Filtering for variants where the deletion length exceeds the insertion length (a hallmark of NHEJ).
    4. Removal of "hidden inter-sharing" variants (artifacts appearing in multiple individuals due to low-quality reads).
• Manual Validation:
  • 896 unique candidate variants were manually inspected using the Integrated Genome Viewer (IGV).
  • Criteria for NHEJ: Presence of deletions, non-templated nucleotide insertions, nearby misincorporations, absence of repetitive sequences, and high-quality alternate allele reads (MAPQ > 20) in both KGP and DRAGEN alignments.
  • Minimum Repair Zone (MRZ): Defined as the shortest region of sequence alteration required to derive the mutant allele from the reference, encompassing deletions, insertions, and substitutions.

3. Key Contributions

• First In Vivo Characterization: Provides the first comprehensive analysis of complex indels and NHEJ repair signatures in normal human somatic cells within their native chromatin environment, free from the artifacts of cell culture or artificial DNA substrates.
• Bioinformatics Advancement: Demonstrates that standard variant callers miss the majority of complex indels and establishes a workflow using Freebayes + custom filtering + manual IGV validation to accurately capture these events.
• Quantification of Age-Related Damage: Establishes a baseline rate of spontaneous DNA double-strand break (DSB) accumulation in normal human colon crypts.
• Impact of Therapy: Quantifies the specific impact of chemotherapy and radiation on DNA damage burden in normal (non-tumor) tissue adjacent to treated areas.

4. Key Results

• Event Identification: From 106 crypts, the study identified 385 high-confidence NHEJ events (complex indels) after manual validation.
• Structural Features:
  • MRZ Length: The Minimum Repair Zone is typically small. 60% of events have an MRZ of 2–3 nucleotides; 73% are ≤5 nucleotides. The longest observed was 43 nucleotides.
  • Nucleotide Insertion Bias:
    • Non-templated additions (86.5% of events) showed a strong bias toward A/T nucleotides (72.8% A/T vs. 27.2% C/G).
    • Templated additions (10.9%) showed no such bias (approx. 50/50 A/T vs. C/G).
  • Microhomology (MH): MH usage was rare. 75.1% of events showed no microhomology. Apparent 1–3 bp MH was often statistically indistinguishable from random chance, confirming that NHEJ in vivo does not strictly rely on MH.
• Age and Treatment Correlation:
  • Age: NHEJ event counts increase with age in untreated individuals (Rate: ~0.19 events/crypt/year).
  • Chemotherapy: Patients treated with chemotherapy alone showed a ~1.5-fold increase in NHEJ events compared to age-matched untreated controls.
  • Radiation + Chemotherapy: Patients receiving combined therapy showed a massive increase (~5-fold over untreated, ~3-fold over chemo-only).
  • Repair Mechanism Stability: Despite the massive increase in the number of breaks caused by radiation, the MRZ size and repair signatures remained unchanged compared to untreated individuals. This suggests radiation increases the frequency of breaks but does not alter the fundamental mechanism of NHEJ repair.

5. Significance

• Physiological Relevance: The study validates that NHEJ is the dominant DSB repair pathway in normal human colon epithelium (where Pol $\theta$ expression is low, limiting alternative end joining).
• Radiation Safety: The findings support the "Linear No-Threshold" model for radiation risk by demonstrating that even in normal tissue, radiation significantly increases DSB frequency. Crucially, the lack of change in repair mechanism suggests the cell's repair capacity is overwhelmed rather than fundamentally altered by the damage type.
• Methodological Benchmark: The study provides a critical dataset and methodology for distinguishing true somatic complex indels from sequencing artifacts, setting a new standard for analyzing natural genomic instability in human tissues.
• Clinical Implications: Understanding the baseline mutation load and the specific impact of adjuvant therapies on normal tissue is vital for assessing long-term cancer risks and secondary malignancies in patients undergoing cancer treatment.