📄 evolutionary biology

Bacterial strain structure shapes the trajectory of antibiotic resistance genes from plasmid to chromosome

This study proposes that bacterial strain diversity acts as a barrier to the chromosomal integration of antibiotic resistance genes, allowing plasmid-borne resistance to transiently dominate populations for decades and explaining why mobile resistance remains prevalent despite theoretical predictions favoring stable chromosomal inheritance.

Original authors: Guillemet, M., Lehtinen, S.

Published 2026-04-15

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Guillemet, M., Lehtinen, S.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Idea: Why Superbugs Keep Carrying "Backpacks"

Imagine bacteria are like hikers in a vast forest. Sometimes, these hikers need a special tool (an antibiotic resistance gene) to survive a storm (antibiotics). They can carry this tool in two ways:

The Backpack (Plasmid): A detachable bag that is easy to share with other hikers but is heavy, uncomfortable, and can easily fall off.
The Tattoo (Chromosome): A permanent mark on their skin. It's lightweight, never falls off, and is very stable, but you can't share it with others easily.

The Puzzle: Evolutionary theory suggests that once a hiker gets the tool, they should eventually switch from the heavy backpack to the permanent tattoo. It makes more sense in the long run. But in the real world, we see bacteria still carrying these heavy backpacks everywhere. Why haven't they switched yet?

The Discovery: The "Neighborhood" Effect

The authors of this paper discovered that the reason bacteria haven't switched to the "tattoo" yet is because of bacterial diversity.

Think of the bacterial population not as a single, uniform crowd, but as a city made of many different neighborhoods (strains). Each neighborhood has its own unique culture, language, and social rules.

The Backpack (Plasmid) is a Social Butterfly: It can jump from one hiker to another, even across neighborhood lines. It's great at spreading quickly through the whole city.
The Tattoo (Chromosome) is a Hereditary Secret: It can only be passed down to children (vertical transmission). It cannot jump between neighbors.

The Analogy of the "Wall":
Imagine a new, better way of carrying a tool (the Tattoo) appears in Neighborhood A. Because it's lighter, the hikers in Neighborhood A start winning and taking over that specific neighborhood.

However, because the Tattoo cannot jump to Neighborhood B, C, or D, those other neighborhoods are still stuck with the heavy Backpacks. The "Tattoo" hikers are stuck in their own little bubbles. To take over the whole city, the Tattoo hikers have to wait for a slow, random process to happen in every single neighborhood one by one.

Meanwhile, the Backpacks keep jumping between neighborhoods, keeping the "Backpack" style alive in the city for a very long time.

The Timeline: It's a Marathon, Not a Sprint

The paper uses math to show that this process takes decades.

In a simple world (one big neighborhood), the switch from Backpack to Tattoo would happen in a few years.
In the real world (many diverse neighborhoods), the switch takes 10 to 50 years.

This explains why we still see so many superbugs carrying resistance genes on plasmids (backpacks). We aren't looking at a "finished" state of evolution; we are looking at a snapshot of a very slow, ongoing transition. The bacteria are stuck in a "transient" phase, like a traffic jam that has been stuck for decades.

The Twist: The "Super-Backpack"

The paper also looked at what happens when bacteria face multiple antibiotics at once (Multi-Drug Resistance).

Imagine a hiker needs a hammer and a saw.

Getting a tattoo for both is hard and takes a long time.
But, a Super-Backpack (a plasmid carrying both genes) can be shared instantly.

In this scenario, the "Super-Backpack" becomes so useful that it actually pushes the "Tattoo" back. The bacteria temporarily go back to carrying the heavy backpacks because the backpack is the only way to carry both tools and share them quickly. This explains why some bacteria seem to be moving away from the chromosome and back to the plasmid.

The Real-World Proof

The authors didn't just guess; they looked at real data from three common gut bacteria (E. coli, Salmonella, and Klebsiella) over the last few decades.

The Result: They found that, on average, resistance genes are slowly moving from the "backpack" to the "tattoo," just as their model predicted.
The Speed: The change is incredibly slow (moving only a few percent per year), confirming that the "neighborhood walls" are indeed slowing down evolution.
The Exceptions: In some bacteria (like Klebsiella), the trend is mixed or even reversed, likely because they are dealing with complex "Super-Backpacks" containing multiple drugs.

The Takeaway

The current crisis of antibiotic resistance isn't a static problem; it's a slow-motion movie.

The reason we see so many plasmids today isn't because they are the "best" long-term solution, but because the diverse structure of bacterial populations acts like a brake, slowing down the switch to the more stable chromosomal location. This gives us a window of time: the "backpack" era is a prolonged, temporary state, and understanding this helps us predict how resistance will evolve in the coming decades.

1. Problem Statement

The evolutionary dynamics governing whether antimicrobial resistance (AMR) genes reside on mobile plasmids or the stable bacterial chromosome are critical for understanding the spread of resistance.

The Paradox: Theoretical models often predict that beneficial traits should eventually integrate into the chromosome due to its stability and lower fitness costs compared to plasmids. However, empirical genomic data frequently shows a strong enrichment of resistance genes on plasmids.
The Gap: Existing literature often focuses on equilibrium states (long-term outcomes). The authors argue that contemporary resistance patterns may not represent a stable equilibrium but rather a transient state in an ongoing evolutionary process.
The Hypothesis: The authors propose that intra-species strain diversity (population structure maintained by balancing selection) acts as a barrier to the selective sweep of vertically transmitted chromosomal genes. While plasmids can cross strain boundaries via horizontal gene transfer (HGT), chromosomal genes are largely limited to vertical transmission, slowing their population-wide fixation.

2. Methodology

The study combines stochastic mathematical modeling with longitudinal genomic data analysis.

A. Mathematical Modeling

Framework: A stochastic multi-strain model implemented as a Continuous Time Markov Chain (CTMC) using a $\tau$ -leaping approach.
Population Structure: The model simulates $n$ competing bacterial strains maintained by balancing selection (negative frequency-dependent selection), preventing any single strain from dominating.
Genotypes: Within each strain, four genotypes are tracked:
1. Sensitive ( $S_i$ )
2. Plasmid-borne resistance ( $P_i$ )
3. Chromosome-borne resistance ( $C_i$ )
4. Double-resistant ( $B_i$ , carrying both)
Dynamics & Transitions:
- Growth: Logistic growth with strain-specific balancing selection.
- HGT: Plasmid conjugation (high rate, $\gamma_p$ ) allows cross-strain transfer; chromosomal HGT (negligible rate, $\gamma_c$ ) is rare.
- Transposition: "Copy-paste" mechanism ( $q_T$ ) moves the gene from plasmid to chromosome within a cell.
- Loss: Plasmid segregation loss ( $q_L$ ) occurs during replication; chromosomal loss is negligible.
Scenarios:
1. Base Model: Resistance introduced on a plasmid; eventual equilibrium favors chromosomal integration.
2. MDR Extension: Introduction of a Multi-Drug Resistance (MDR) plasmid carrying linked resistance genes to test for "linked selection" effects that might reverse the trajectory.

B. Empirical Data Analysis

Data Source: NCBI Pathogen Detection MicroBIGG-E database (accessed Dec 2025).
Species: Escherichia coli, Klebsiella pneumoniae, and Salmonella enterica.
Criteria: Only "Complete Genomes" were used to accurately assign genes to plasmids or chromosomes.
Statistical Approach: Binomial logistic regression was used to model the log-odds of a gene being plasmid-borne as a function of time (Year).
- Slope ( $\beta_1$ ): Represents the annual change in the probability of plasmid carriage.
- Significance: Assessed via Wald Z-tests ( $P < 0.05$ ).

3. Key Results

A. Strain Diversity Delays Chromosomal Fixation

Single Strain ( $n=1$ ): In a homogeneous population, plasmid-borne resistance invades quickly, followed by a rapid selective sweep of the fitter chromosomal version. The transition is fast.
Structured Population ( $n > 1$ ): In a diverse population, the trajectory remains the same (plasmid $\to$ $\to$ chromosome), but the timescale is drastically extended.
- With 10 strains, the time for chromosomal resistance to reach 99% fixation increased by a factor of 13.2 (from ~5,000 hours to ~66,000 hours).
- The period of plasmid dominance (where plasmid frequency > chromosomal frequency) extended by a factor of 3.4.
Mechanism: The delay occurs because the plasmid must invade each naive strain via HGT, and the chromosomal integration must occur independently within each strain. The population structure prevents a single "sweep" from clearing the entire population at once.
Timescale Estimate: For moderate diversity, the transition takes on the order of $10^5$ hours (approx. 10 years).

B. Multi-Drug Resistance (MDR) Can Reverse Trajectories

In scenarios where a population has fixed chromosomal resistance to Drug A, the introduction of an MDR plasmid (carrying resistance to Drug A and Drug B) can cause a transient reversal.
The MDR plasmid spreads rapidly due to the benefit of Drug B resistance and its high conjugation rate. This temporarily increases the proportion of Drug A resistance residing on plasmids, even if the chromosomal version is more stable for Drug A alone.
This suggests that complex selective pressures (multiple antibiotics) can maintain plasmid-borne resistance longer than single-drug models predict.

C. Empirical Validation

General Trend: Across all three species, the median regression slope ( $\beta_1$ ) was negative, indicating a long-term shift of resistance genes from plasmids to chromosomes.
Species Heterogeneity:
- S. enterica: Strong, uniform shift toward the chromosome (13 significant genes, 0 shifting to plasmids).
- E. coli: Intermediate profile (9 shifting to chromosome, 5 shifting to plasmids).
- K. pneumoniae: Distinct behavior; despite a slightly negative median, the subset of significant genes showed a strong trend toward increased plasmid association (12 vs 8).
Rate of Change: The observed slopes ( $|\beta_1| \approx 0.1–0.5$ ) correspond to a slow, gradual decline in plasmid association (e.g., dropping from 50% to ~38% in one year), supporting the model's prediction of a decade-long transient process rather than an instantaneous sweep.

4. Key Contributions

Non-Equilibrium Perspective: The paper shifts the paradigm from viewing AMR gene location as a static equilibrium to viewing it as a prolonged transient state driven by population structure.
Role of Strain Structure: It identifies intra-species diversity as a fundamental "brake" on the selective sweep of vertically transmitted beneficial genes, a factor often overlooked in single-strain models.
Timescale Quantification: It provides a semi-quantitative estimate that the transition from plasmid to chromosome dominance can take decades, explaining why plasmid-mediated resistance remains prevalent in modern clinical data.
MDR Complexity: It demonstrates how linked selection on MDR plasmids can create complex, non-monotonic evolutionary trajectories, potentially explaining species-specific anomalies (like in K. pneumoniae).

5. Significance

Public Health Implications: The persistence of plasmid-borne resistance is not necessarily a failure of evolution to optimize, but a consequence of the timescale of evolution in diverse populations. This suggests that plasmid-mediated resistance may remain a dominant threat for decades even if chromosomal integration is theoretically superior.
Evolutionary Theory: The findings highlight that population genetic structure is a critical determinant of the evolutionary trajectory of beneficial genes, applicable beyond AMR to any recently introduced adaptive trait.
Data Interpretation: The study provides a framework for interpreting genomic surveillance data, suggesting that current distributions of AMR genes reflect an "ongoing process" rather than a final evolutionary endpoint.

In summary, the authors demonstrate that bacterial population structure acts as a barrier to the rapid fixation of chromosomal resistance, extending the window of plasmid dominance to human-relevant timescales (decades). This explains the discrepancy between theoretical predictions of chromosomal stability and the observed prevalence of plasmid-mediated resistance in modern pathogens.