Novel KIF22 Variants Disrupt Mitosis in Human Chondrocytes and Expand SEMDJL2 Mechanisms

This study characterizes novel KIF22 variants in human chondrocytes, revealing that heterozygous mutations cause dominant mitotic defects through constitutive activation while the recessive R49Q variant leads to milder, partially penetrant segregation errors via mixed-state dysregulation, thereby expanding the known genotypic and mechanistic landscape of SEMDJL2.

The Gist

The Big Picture: A Broken Construction Crew

Imagine your body is a massive city being built. The "construction workers" are your cells, and they need to divide perfectly to build new rooms (tissues) and expand the city (growth).

One specific type of worker, a protein called KIF22, acts like a specialized crane operator inside the cell. Its job is to push the "blueprints" (chromosomes) into the perfect center of the room before the cell splits in two.

If this crane operator gets confused or breaks, the blueprints get scattered. The cell can't divide correctly, leading to a construction disaster. In humans, this specific disaster causes a rare bone condition called SEMDJL2, which results in short stature, loose joints, and skeletal deformities.

What This Paper Did

Scientists already knew that some "typos" (mutations) in the KIF22 instructions caused this disease. However, they didn't fully understand how the different typos broke the crane.

In this study, the researchers looked at four different typos:

1. Two that were already known to be bad (the "Hotspot" variants).
2. Two brand new ones they found in patients (P144T and E222Q).
3. One rare "recessive" typo (R49Q) that only causes disease if a person has two copies of it.

They grew human cartilage cells (chondrocytes) in a lab, introduced these broken instructions, and watched the cells divide under a microscope to see exactly what went wrong.

The Two Types of Breakdowns

The researchers discovered that the broken cranes fail in two very different ways. Think of it like a car's gas pedal and brake system.

1. The "Stuck Gas Pedal" (The Dominant Variants)

The Variants: P144T, E222Q, and the known R149Q.
The Problem: These variants are like a gas pedal that gets stuck in the "ON" position.

• How it works: Normally, the KIF22 crane pushes the blueprints to the center, and then stops pushing when it's time to split the cell.
• The Glitch: These broken variants refuse to stop. They keep pushing even after the cell should be splitting.
• The Result: The blueprints get pushed too hard and get stuck in the middle. The cell tries to pull apart, but the crane is still shoving the blueprints together. The cell ends up with a messy, lumpy nucleus (the cell's brain) or fails to divide entirely.
• The Takeaway: You only need one bad copy of the instruction for this to happen. It's a "dominant" problem because the stuck pedal overrides the good instructions.

2. The "Weak Engine with a Sticky Brake" (The Recessive Variant)

The Variant: R49Q.
The Problem: This is a different kind of mess. It's like a car with a weak engine that also has a brake that won't fully release.

• How it works: This variant is a "loss of function." It doesn't push as hard as a normal crane.
• The Glitch: However, it also doesn't turn off completely. It's in a confused "halfway" state.
• The Result: Because the engine is weak, it doesn't do its job well. But because it doesn't turn off, it still interferes with the splitting process, just not as violently as the "stuck gas pedal."
• The Takeaway: You need two bad copies (one from mom, one from dad) to see the disease. If you have one good copy, that good copy is strong enough to do the job, and the weak one doesn't cause enough trouble to break the system.

Why This Matters

1. New Diagnoses: The scientists found two new mutations (P144T and E222Q) in patients who had the classic symptoms of the bone disease. They proved these mutations are definitely the cause, not just random noise. This helps doctors diagnose patients faster.
2. Understanding the "Why": By showing that some variants act like a "stuck gas" and others like a "weak engine," they created a new map for understanding the disease.
3. Future Treatments: If we know how the machine is broken, we can design better tools to fix it.
   • For the "Stuck Gas" variants, we might look for drugs that force the crane to turn off.
   • For the "Weak Engine" variants, we might look for ways to boost the activity of the remaining good copies.

The Bottom Line

This paper is like a mechanic opening the hood of a broken car and saying, "Ah, I see the problem. In some cars, the accelerator is jammed. In others, the engine is weak but the brakes are sticky."

By understanding these specific mechanical failures in the cells that build our bones, we get one step closer to understanding why people with SEMDJL2 have short stature and joint problems, and perhaps one day, how to fix it.

Technical Summary

1. Problem Statement

Spondyloepimetaphyseal dysplasia with joint laxity, type 2 (SEMDJL2) is a rare skeletal disorder caused by pathogenic variants in the KIF22 gene. KIF22 is a mitotic chromokinesin (Kid) responsible for generating polar ejection forces (PEF) that align chromosomes and facilitate their segregation.

While previous research established that specific "hotspot" heterozygous variants (e.g., P148L, R149Q) cause mitotic defects in epithelial cells by failing to inactivate KIF22 at the metaphase-to-anaphase transition, several critical gaps remained:

• The mechanism of how these variants affect human chondrocytes (the primary cell type in skeletal dysplasia) was unknown.
• The functional impact of novel heterozygous variants (P144T, E222Q) identified in patients was uncharacterized.
• The mechanism of the recently reported recessive homozygous variant (R49Q) was unclear, specifically whether it acts via loss-of-function or a distinct dysregulation mechanism.
• There was a need to reclassify Variants of Uncertain Significance (VUS) based on functional data to expand the genotypic landscape of SEMDJL2.

2. Methodology

The authors employed a combination of clinical genetics, cell biology, and live-cell imaging to characterize the variants.

• Clinical Cohort: Two new patients were identified with classic SEMDJL2 phenotypes carrying heterozygous KIF22 variants: P144T and E222Q. These were previously classified as VUS.
• Cell Model: The study utilized CHON-001 human immortalized chondrocytes.
  • Inducible Expression: Researchers generated cell lines with doxycycline-inducible expression of GFP-tagged wild-type (WT) KIF22 and variants (R149Q, P144T, E222Q, R49Q).
  • siRNA Knockdown: To isolate the effects of specific variants, endogenous KIF22 was knocked down using siRNA. The GFP-tagged constructs contained silent mutations to confer resistance to the siRNA, allowing simultaneous depletion of endogenous protein and expression of the variant.
• Assays:
  • Polar Ejection Force (PEF) Assay: Cells were treated with monastrol (a KIF11 inhibitor) to collapse spindles into monopoles. The distance between chromosomes and the spindle pole was measured via immunofluorescence (DAPI/Centrin) to quantify PEF generation.
  • Live-Cell Imaging: Time-lapse microscopy (SiR-tubulin for microtubules, Spy-DNA or KIF22-GFP for chromosomes) was used to track anaphase chromosome segregation and spindle pole separation.
  • Nuclear Morphology Analysis: Post-mitotic interphase nuclei were stained with DAPI, and nuclear solidity (ratio of area to convex hull area) was measured to quantify nuclear shape abnormalities.

3. Key Contributions

• Reclassification of VUS: Provided functional evidence to reclassify P144T and E222Q from VUS to likely pathogenic, expanding the known mutational spectrum of SEMDJL2.
• Mechanistic Dichotomy: Defined two distinct mechanistic classes of KIF22 dysregulation:
  1. Constitutive Activation: Heterozygous variants (P144T, E222Q, R149Q) that retain motor function but fail to down-regulate activity at anaphase onset.
  2. Mixed-State Dysregulation: The recessive R49Q variant, which exhibits partial loss of motor function coupled with incomplete inactivation.
• Chondrocyte-Specific Validation: Confirmed that the mitotic defects observed in epithelial cells (HeLa/RPE-1) are recapitulated in human chondrocytes, validating the relevance of the model for skeletal dysplasia.

4. Key Results

A. Polar Ejection Force (PEF) Activity

• Heterozygous Variants (P144T, E222Q): Retained full PEF-generating activity comparable to Wild-Type (WT) KIF22. Even when endogenous KIF22 was knocked down, these variants could push chromosomes away from the pole effectively.
• Recessive Variant (R49Q): Displayed reduced PEF activity. In the absence of endogenous KIF22, R49Q produced significantly less force than WT, consistent with a partial loss-of-function allele.

B. Anaphase Defects (Live Imaging)

• Heterozygous Variants: All heterozygous variants (P144T, E222Q, R149Q) caused dominant defects. They failed to down-regulate KIF22 at the metaphase-to-anaphase transition.
  • Consequence: Persistent PEF during anaphase B impeded chromosome segregation and spindle pole separation.
  • Severity: E222Q caused the most severe chromosome separation defects (max distance 5.0 µm vs. ~14.8 µm in WT), followed by R149Q (4.7 µm) and P144T (~8.6 µm).
• Recessive Variant (R49Q):
  • In the presence of endogenous KIF22, R49Q caused a modest reduction in chromosome separation.
  • When endogenous KIF22 was knocked down (simulating homozygosity), the defect worsened but remained milder than the heterozygous variants.
  • Unique Finding: Unlike the heterozygous variants, R49Q did not significantly impair spindle pole separation, suggesting a different mechanism of anaphase disruption.

C. Nuclear Morphology

• Expression of all variants led to a significant increase in abnormal nuclear shapes (low solidity values) in daughter cells.
• P144T showed the highest percentage of abnormal nuclei (approx. 76% with KD), while E222Q showed a lower percentage (approx. 55% with KD), despite E222Q causing more severe segregation defects. This suggests that nuclear morphology defects are not strictly linearly correlated with the severity of chromosome segregation errors.

5. Significance

• Mechanistic Insight: The study resolves the molecular pathology of SEMDJL2 by demonstrating that the disease is driven by anaphase dysregulation rather than a simple loss of motor function. The "constitutive activation" model explains why heterozygous variants are dominant: the failure to turn off the motor creates a "traffic jam" of forces that prevents chromosome separation.
• Genetic Expansion: By validating P144T and E222Q, the study broadens the diagnostic criteria for SEMDJL2, allowing for the identification of more patients who were previously undiagnosed or misdiagnosed.
• Therapeutic Implications: Understanding that the pathology arises from failure to inactivate KIF22 rather than total loss of function suggests that therapeutic strategies might focus on modulating the regulatory phosphorylation pathways or the autoinhibitory mechanisms of KIF22, rather than simply replacing the protein.
• Cell-Type Specificity: The findings highlight that while the core mitotic defect is conserved across cell types, the specific phenotypic outcomes (e.g., severity of nuclear defects vs. segregation defects) may vary, emphasizing the need for disease-relevant cell models (chondrocytes) in studying skeletal dysplasias.