Loss of Mycobacterium marinum ESX-1 genes increase transcription of ESX-6 genes

This study reveals that the loss of ESX-1 genes in the smooth *Mycobacterium marinum* 1218S strain leads to increased transcription of ESX-6 and LOS genes, suggesting a regulatory link between these secretion systems, lipooligosaccharide synthesis, and colony morphology.

The Gist

The Big Picture: A Tale of Two Bacterial Cousins

Imagine two brothers, 1218R and 1218S, who are both strains of a bacterium called Mycobacterium marinum. This bacterium is a close cousin of the one that causes tuberculosis in humans, so scientists study it to understand how these germs work.

• Brother 1218R (The "Rough" Brother): He forms "rough" colonies (like a bumpy rock) when grown in a lab. He is tough, aggressive, and very good at infecting fish. He is the "bad boy" of the family.
• Brother 1218S (The "Smooth" Brother): He is a mutant version of his brother. He forms "smooth" colonies (like a polished pebble). He is much weaker, less aggressive, and not very good at making fish sick.

The scientists wanted to know: Why is the Smooth brother so weak, and why does he look so different?

The Missing Toolkit (The ESX-1 System)

Think of the bacteria as a factory. To survive and attack, they need a special secret weapon system called ESX-1. This system is like a high-tech injection gun that shoots harmful proteins into host cells.

• The Discovery: When the scientists looked at the Smooth brother's (1218S) instruction manual (his genome), they found a huge chunk of the manual was missing. Specifically, he was missing several key parts of his "injection gun" (the ESX-1 genes).
• The Consequence: Because he lost his main weapon, he is much less dangerous. This explains why he can't infect fish as well as his Rough brother.

The "Backup Generator" Effect (ESX-6)

Here is where it gets interesting. The bacteria have a backup system called ESX-6. It's like a spare tire or a backup generator. In the Rough brother, this backup system is barely used; it sits idle in the garage.

• The Twist: In the Smooth brother, because his main weapon (ESX-1) is broken/missing, the factory goes into panic mode. It screams, "We need power!" and turns the backup generator (ESX-6) up to maximum volume.
• The Result: The Smooth brother is actually over-producing the backup system to try to compensate for the loss of the main system. It's like a car with a broken engine revving its backup battery so loudly it shakes the whole car.

The Smooth Skin Mystery (LOS Genes)

Why does one brother look like a bumpy rock (Rough) and the other like a smooth pebble (Smooth)?

• The Surface: Bacteria have an outer skin. The Rough brother has a bumpy, complex skin. The Smooth brother has a slick, slippery skin. This skin is made of special fats called LOS (Lipooligosaccharides).
• The Clue: The scientists checked the Smooth brother's manual again. Surprisingly, he has the exact same instructions for building his skin as his Rough brother. He didn't lose the genes; he just has them turned on differently.
• The "Volume Knob": It turns out that in the Smooth brother, the genes that build the smooth skin are turned up much louder, especially when the bacteria are "tired" (in the stationary phase of growth).
• The Regulator: There is a "manager" protein called Lsr2 that usually tells the bacteria to keep the skin genes quiet. The Smooth brother has two copies of this manager instead of one, and he also has a different "volume control" (a non-coding RNA called Ms1) that seems to crank up the volume on the smooth skin genes. This extra volume makes the skin smooth and slippery.

The Biofilm Connection (The Bacterial City)

Bacteria often build communities called biofilms (like a slime city) to protect themselves.

• The Rough Brother: Builds a very strong, thick, and robust city.
• The Smooth Brother: Builds a weak, flimsy city.
• The Experiment: The scientists tried to give the Smooth brother the missing "injection gun" parts back on a plasmid (a tiny delivery truck). They hoped this would fix his weak city.
• The Result: It didn't work. The Smooth brother still couldn't build a strong city. This tells us that the missing weapon parts aren't just about the gun; they are part of a complex network that affects how the bacteria stick together and build their homes.

The "Glitch" in the System (RNase J)

Finally, the scientists found a small but important glitch. The Smooth brother has a broken piece of machinery called RNase J. Think of this machine as a trash compactor for old, broken instructions (RNA).

• Because the trash compactor is broken in the Smooth brother, old instructions pile up. This might be why he has so much "noise" in his system, leading to the over-production of the backup generator (ESX-6) and the extra smooth skin genes.

The Bottom Line

This paper tells the story of how a small genetic accident (losing a chunk of DNA) caused a chain reaction:

1. The bacteria lost its main weapon.
2. It panicked and over-activated its backup system.
3. It accidentally turned up the volume on its "smooth skin" genes.
4. It lost its ability to build strong protective cities (biofilms).

All of this combined makes the "Smooth" brother a weakling compared to his "Rough" brother. It shows that in the microscopic world, losing a part can change the entire personality and behavior of a germ.

Technical Summary

1. Problem Statement

Mycobacterium marinum (Mmar) serves as a critical model for studying Mycobacterium tuberculosis pathogenicity. Mmar strains exhibit two distinct colony morphologies: Rough (R) and Smooth (S). The R morphotype (e.g., strain 1218R) is highly virulent, while the S morphotype (e.g., strain 1218S) is significantly less virulent.

• The Gap: Previous genomic analysis suggested that strain 1218S lacks several genes in the ESX-1 type VII secretion system, which are known virulence factors. However, the draft genome of 1218S was incomplete, leaving uncertainty regarding the exact extent of these deletions and the presence of truncated genes.
• The Question: How does the loss of ESX-1 genes in the smooth variant (1218S) affect the transcription of remaining ESX genes (specifically the duplicated ESX-6 system), lipooligosaccharide (LOS) synthesis genes (responsible for colony morphology), and overall biofilm formation? Furthermore, what regulatory mechanisms drive the smooth phenotype despite the absence of genomic deletions in the LOS locus?

2. Methodology

The study employed a multi-omics approach combining comparative genomics, transcriptomics, and phenotypic assays:

• Genome Sequencing & Assembly: The authors generated a complete, single-scaffold genome for the smooth strain 1218S using PacBio long-read sequencing (HGAP3 assembly). This was compared against the complete genomes of 1218R and three other Mmar strains (CCUG 20998, M, and ATCC 927T).
• Comparative Genomics:
  • Used RAST and VFanalyzer (VFDB) for functional classification and virulence factor prediction.
  • Performed "all-vs-all" BLAST and PanOCT analysis to identify core and unique genes across the five strains.
  • Manually curated annotations to correct for SNPs, frameshifts, and assembly artifacts found in the draft genome.
• Transcriptomics (RNA-Seq):
  • Isolated total RNA from exponential and stationary growth phases for strains 1218R, 1218S, CCUG, and M.
  • Mapped reads to the complete genomes using Bowtie2 and Tophat.
  • Analyzed differential expression using DESeq2 (TPM values and fold changes).
  • Focused on ESX systems (1, 3, 4, 5, 6), LOS biosynthesis genes, and regulators (WhiB6, EspR, EspM, PhoPR, Lsr2, WhiB4).
• Functional Manipulation:
  • Overexpression: Cloned the Ms1 RNA gene (a non-coding RNA) into a tetracycline-inducible plasmid (pBS401) and introduced it into strain CCUG to assess its effect on LOS and ESX gene transcription.
  • Complementation: Cloned the missing ESX-1 genes (espF_2, espG1_2, espH) from 1218R into a plasmid and transformed 1218S to test for phenotypic rescue.
• Phenotypic Assays: Quantified biofilm formation in 96-well plates over 26 days for strains with and without the complementation plasmid.

3. Key Contributions & Results

A. Genomic Refinement of Strain 1218S

• Genome Size: The complete 1218S genome is 6,410,821 bp, approximately 56 kb smaller than 1218R.
• ESX-1 Deletion: Confirmed that 1218S lacks a large segment of the ESX-1 locus (nucleotides 6,408,265–6,418,995 in 1218R).
• Truncated Genes: Contrary to the draft genome, the complete assembly revealed that espF_2 and esxB_3 are not entirely absent but are truncated.
  • espF_2 retains its N-terminus.
  • esxB_3 (CFP-10 homolog) is truncated, resulting in a frameshift. The authors predict a novel 98-amino acid peptide is produced from a fusion of the remaining espF_2 and the truncated esxB_3, which likely cannot form the functional EsxA-EsxB heterodimer required for secretion.
• ESX-6 Duplication: The ESX-6 region (a partial duplication of ESX-1) contains a 1,371 bp deletion in 1218S, resulting in a C-terminal truncation of a PE family protein.

B. Transcriptional Regulation of ESX Systems

• Compensatory Upregulation: The loss of ESX-1 genes in 1218S triggers a significant increase in the transcription of ESX-6 genes.
  • In 1218R: ESX-1 transcripts are high; ESX-6 transcripts are low.
  • In 1218S: ESX-1 transcripts (from remaining genes) are reduced, but ESX-6 transcripts are significantly elevated in both exponential and stationary phases.
• Growth Phase Effects: ESX-1 and ESX-5 transcripts generally increase in stationary phase, while ESX-3 transcripts decrease.
• Regulatory Impact: The deletion of the "espF_2 – esxB_3" region in 1218S leads to reduced transcript levels of the regulators whiB6 and espM, suggesting a feedback loop where the integrity of the ESX-1 secretion machinery influences its own regulatory network.

C. LOS Genes and Colony Morphology

• Genomic Identity: The LOS gene cluster (responsible for smooth/rough morphology) is identical in sequence and synteny between 1218R and 1218S, with one exception: lsr2 is duplicated in 1218S.
• Transcriptional Divergence: Despite identical gene sequences, LOS gene transcripts are significantly higher in stationary-phase 1218S cells compared to 1218R.
• Role of Ms1 RNA:
  • Ms1 RNA (a non-coding RNA) levels are higher in 1218S than 1218R.
  • Overexpression of Ms1 RNA in strain CCUG increased the transcription of key LOS genes (e.g., pimF/losA, papA1_2, rffA, pglD).
  • This suggests Ms1 RNA acts as a global regulator promoting LOS synthesis, potentially driving the smooth phenotype in 1218S.
• RNase J Mutation: A partial deletion in the rnj (RNase J) gene in 1218S was identified. This may alter mRNA stability/degradation rates, contributing to the accumulation of LOS transcripts.

D. Biofilm Formation

• Phenotype: 1218R forms robust biofilms; 1218S forms weak biofilms.
• Complementation Failure: Introducing the missing ESX-1 genes (espF_2, espG1_2, espH) via plasmid into 1218S did not restore biofilm formation.
• Negative Regulation: Interestingly, introducing these genes into 1218R reduced its biofilm formation, suggesting that an excess of these specific ESX-1 components negatively impacts biofilm stability.

4. Significance

• Mechanism of Virulence Attenuation: The study confirms that the loss of the ESX-1 secretion system in 1218S is a primary driver of its reduced virulence, likely due to the inability to secrete key effectors (like the EsxA-EsxB complex).
• Transcriptional Compensation: It reveals a novel regulatory mechanism where the loss of a primary secretion system (ESX-1) leads to the compensatory upregulation of its paralog (ESX-6), highlighting the plasticity of mycobacterial gene regulation.
• Colony Morphology Regulation: The findings decouple colony morphology from simple gene presence/absence. Instead, they demonstrate that transcriptional regulation (driven by Ms1 RNA, lsr2 duplication, and mRNA stability via RNase J) dictates the smooth/rough phenotype.
• Biofilm Complexity: The results indicate that biofilm formation in Mmar is not simply restored by adding back missing ESX genes, implying a complex interplay between secretion machinery, cell wall composition, and global regulators.
• Therapeutic Implications: Understanding how LOS synthesis and ESX systems are regulated provides new targets for disrupting mycobacterial virulence and biofilm persistence, which are critical for treating chronic infections.

In summary, this paper provides a comprehensive genomic and transcriptomic dissection of the transition from a virulent rough to an avirulent smooth M. marinum strain, identifying specific gene losses, compensatory transcriptional shifts, and non-coding RNA-mediated regulation as key determinants of pathogenicity and morphology.