Assessing positive selection in centromere-associated kinetochore proteins across Metazoan groups.

This study investigates positive selection in outer kinetochore and condensin proteins across diverse metazoan groups, finding sporadic signatures of selection in parasitic wasps, ray-finned fishes, and primates, but none in the asexual Amazon molly, which is exempt from centromere drive.

The Gist

The Centromere Arms Race: A Story of Selfish DNA and Its Bodyguards

Imagine your body is a massive construction site. Inside every cell, there are 46 blueprints (chromosomes) that need to be copied and split perfectly in half every time a cell divides. If even one blueprint gets lost or torn, the cell dies or becomes cancerous.

The "zipper" that holds these blueprints together and allows them to be pulled apart is called the centromere.

The Problem: The Selfish Zipper

Here's the twist: The centromere is made of repetitive DNA that acts a bit like a selfish teenager. It wants to be the "favorite child" and get passed down to the next generation more often than it should.

In female cells, there's a competition. The centromere that is "bigger" or "stickier" gets pulled into the egg cell, while the smaller one gets discarded. This is called Centromere Drive. The selfish centromere keeps growing bigger and stickier just to win this competition.

The Reaction: The Bodyguards Fight Back

But there's a problem. The centromere is just a piece of DNA; it can't pull itself. It needs a team of protein "bodyguards" (called kinetochore proteins) to grab onto it and pull the chromosome apart.

If the centromere changes its shape to be "stickier," the bodyguards' hands no longer fit. They have to evolve new hands to grab the new shape. This creates a never-ending evolutionary arms race:

1. The centromere changes to cheat the system.
2. The bodyguards change to catch the cheat.
3. The centromere changes again.

Scientists call this the Red Queen Effect (from Alice in Wonderland): "It takes all the running you can do, to keep in the same place."

The Study: Checking the Bodyguards

For a long time, scientists only looked at the "inner" bodyguards (the ones touching the DNA directly) and found they were constantly changing (evolving rapidly) in many animals, like fruit flies.

But this new study asked a bigger question: Does this arms race spread to the "outer" bodyguards? These are the proteins further out in the chain that help organize the chromosomes (like the Condensin and Mis12 complexes). Do they also have to run as fast as the inner bodyguards?

The researchers looked at three different groups of animals:

1. Wasps (Insects)
2. Fish (including a special asexual fish called the Amazon Molly)
3. Primates (Monkeys and Apes, including humans)

The Findings: A Mixed Bag

1. The Amazon Molly (The Control Group)
The researchers included a special fish, the Amazon Molly, which reproduces without males (asexually). Because it doesn't have the "competition" of male vs. female chromosomes, it shouldn't have this selfish centromere drive.

• Result: As predicted, the Amazon Molly's bodyguards were calm and stable. They weren't changing much. This confirmed that the "arms race" is indeed driven by sexual competition.

2. The Sexual Animals (Wasps, Fish, Primates)
In the animals that reproduce sexually, the researchers found that the arms race does happen, but it's not as chaotic as they expected.

• The Inner Circle: The "inner bodyguards" (CENP-A and CENP-C) were definitely changing fast, just like in fruit flies.
• The Outer Circle: The "outer bodyguards" (Condensin and Mis12) showed some signs of changing, but it was sporadic. It wasn't a constant, wild race across every single protein in every single animal. Sometimes a protein changed, sometimes it didn't.

The Takeaway

Think of it like a game of telephone.

• In fruit flies, the message (the centromere change) was so loud that everyone in the line (all the proteins) had to shout back and change their message constantly.
• In these other animals (wasps, fish, monkeys), the message is still being passed, but only some people in the line are shouting back. The "outer" bodyguards are mostly stable, only occasionally needing to update their "gloves" to catch the changing centromere.

Why does this matter?
It tells us that evolution is messy and specific. Even though the "selfish centromere" is a universal problem, different animals solve it in different ways. Some have a full-blown war involving all their proteins, while others only have skirmishes with a few specific ones.

The study also highlights that we still have a lot to learn about how these proteins talk to each other, but we now know that the "arms race" is real, it's happening in many animals, and it stops completely when the competition (sexual reproduction) stops.

Technical Summary

1. Problem Statement and Hypothesis

Centromeres are essential chromosomal regions that evolve rapidly and often "selfishly" through a mechanism known as centromere drive. In this process, larger centromeres bias their transmission into oocytes during female meiosis. To maintain functional compatibility with these rapidly evolving centromeres, interacting kinetochore proteins must co-evolve in a "Red Queen" scenario, often driven by positive selection.

While positive selection in inner kinetochore proteins (specifically CENP-A and CENP-C) is well-documented in model organisms like Drosophila, plants, and mammals, the extent of this selection in outer kinetochore and condensin protein complexes across diverse Metazoan groups remains underexplored.

Hypothesis: The authors hypothesized that signatures of positive selection would be pervasive not only in inner kinetochore genes but also in genes encoding components of the Condensin I and Mis12 protein complexes across diverse animal groups. They further hypothesized that an asexual diploid species (the Amazon molly, Poecilia formosa), which reproduces via gynogenesis and lacks meiotic drive, would show no such signatures of positive selection.

2. Methodology

The study employed a comparative phylogenetic approach across three major animal groups:

• Insects: Parasitic wasps (Family Braconidae) and Drosophila (for context).
• Fish: Two families of ray-finned fishes: Cichlidae and Poeciliidae (including the asexual P. formosa).
• Primates: Two families: Hominidae (great apes/humans) and Cercopithecidae (Old World monkeys).

Target Genes:

• Inner Kinetochore: cenp-a and cenp-c.
• Condensin I Complex: SMC2, SMC4, CAP-G (NCAP-G), CAP-D2 (NCAP-D2), CAP-H (NCAP-H).
• Mis12 Complex: MIS12, DSN1, PMF1, NSL1.

Data Processing:

• Sequences were retrieved from Ensembl (v98), NCBI, and unpublished genome assemblies.
• Orthologs were identified via BLAST and verified for exon-intron structure.
• Multiple sequence alignments were performed using MAFFT, MUSCLE, and Seaview, followed by manual curation to remove gaps and stop codons.
• Note: cenp-c was absent in fish; dsn1 was absent in fish; smc4 was absent in wasps; and cenp-a was excluded from wasps due to ambiguous orthology.

Statistical Analysis (PAML Suite v4.9):

1. Site Models (codeml): Compared a neutral model (M7) against a positive selection model (M8) using Likelihood Ratio Tests (LRT). Significant results were validated using Bayes Empirical Bayes (BEB) to identify specific sites under selection.
2. Branch Models (Free Ratio): Estimated the non-synonymous to synonymous substitution rate ratio ($\omega = dN/dS$) for each branch independently. Branches with $\omega > 1$ were flagged as candidates for positive selection.
3. Branch-Site Models: Used to confirm positive selection on specific foreground branches identified by the Free Ratio model.
4. False Positive Filtering: Rigorous screening was applied to remove artifacts where high $\omega$ values resulted from low sequence divergence (e.g., $dS=0$) rather than true adaptive evolution.

3. Key Results

A. Site Model Tests (M7 vs. M8):
Significant evidence of positive selection (M8 fit significantly better than M7) was found in a subset of genes across all groups, but not universally:

• Braconidae: ncap-d2, ncap-h, pmf1.
• Poeciliidae: cenp-a, smc2.
• Cichlidae: ncap-d2.
• Hominidae: ncap-g, nsl1.
• Cercopithecidae: mis12.

B. Branch-Specific Analysis ($\omega > 1$):
Widespread elevation of $\omega > 1$ was observed, though it was sporadic rather than pervasive:

• Inner Kinetochore: Confirmed positive selection in cenp-a (fish and primates) and cenp-c (primates only), consistent with prior literature.
• Condensin I & Mis12: Multiple lineages showed $\omega > 1$ in genes such as ncap-g, ncap-d2, smc4, nsl1, pmf1, mis12, and dsn1 across all groups except the asexual molly.
• The Amazon Molly (P. formosa): Crucially, no gene in P. formosa exhibited $\omega > 1$. This supports the hypothesis that centromere drive is absent in asexual diploids, halting the cascade of compensatory positive selection in kinetochore proteins.

C. Confirmation via Branch-Site Models:
While the Free Ratio model suggested many candidates, the more stringent Branch-Site model confirmed positive selection in only two cases:

1. ncap-d2 in the cichlid Astatotilapia calliptera.
2. cenp-a in the primate Rhinopithecus bieti.

4. Key Contributions

• Broad Taxonomic Scope: This is one of the first studies to systematically assess positive selection in outer kinetochore and condensin complexes across insects, fish, and primates, moving beyond the Drosophila model.
• Validation of the "Centromere Drive" Cascade: The study provides empirical support that centromere drive drives positive selection in interacting protein complexes (Condensin I and Mis12), not just the inner kinetochore.
• The Asexual Control: The inclusion of the Amazon molly serves as a critical negative control, demonstrating that without meiotic drive, the selective pressure on kinetochore proteins is absent.
• Methodological Caution: The authors highlight the discrepancy between "suggestive" signals (Free Ratio $\omega > 1$) and "confirmed" signals (Branch-Site models), suggesting that previous studies relying solely on $\omega > 1$ may have overestimated the prevalence of positive selection.

5. Significance and Limitations

Significance:
The findings suggest that while positive selection is a real phenomenon in centromere-associated proteins across Metazoa, it is sporadic and lineage-specific rather than a universal, pervasive pattern as seen in Drosophila. The lack of pervasive selection in the Amazon molly reinforces the link between sexual reproduction/meiotic drive and the rapid evolution of cell division machinery.

Limitations:

• Data Resolution: The study relied on single sequences per species (no population-level polymorphism data). This prevented the use of more powerful tests for long-term positive selection (e.g., McDonald-Kreitman test) or selective sweeps.
• Confirmation Rate: The low confirmation rate of the Branch-Site model suggests that many signals of $\omega > 1$ may be noise or driven by specific, short-term events rather than sustained adaptive evolution.
• Structural Knowledge Gaps: The authors note that a lack of detailed knowledge regarding specific protein-protein interaction domains across diverse taxa limits the ability to pinpoint exactly where compensatory co-evolution is occurring.

Conclusion:
The study confirms that centromere drive drives positive selection in outer kinetochore and condensin proteins, but this evolutionary dynamic is highly variable across the animal kingdom. The absence of these signatures in the asexual Amazon molly provides strong evidence that meiotic drive is the primary engine for this rapid co-evolution.