Characterization of the biofilm landscape of Bacillus subtilis by spatial microproteomics

This study utilizes a combined approach of top-down proteomics and mass spectrometry imaging to map the spatial proteomic landscape of *Bacillus subtilis* biofilms, successfully demonstrating the detection of differentiated subpopulations through proteoforms and revealing the localization of post-translational modifications at the microscale.

The Gist

Imagine a bustling city built by a single species of bacteria called Bacillus subtilis. To the naked eye, this bacterial city (a biofilm) looks like a simple, flat blob of slime. But inside, it's actually a complex metropolis with different neighborhoods, each with its own jobs, culture, and even its own "fashion sense" (the specific proteins the cells are wearing).

For a long time, scientists could only take a "smoothie" of this city. They would scrape the whole thing up, mix it all together, and analyze the average. This told them what was in the city, but not where it was or who was doing what. It's like knowing a city has police, firefighters, and bakers, but not knowing if the firefighters are actually putting out fires or just hanging out at the bakery.

This paper introduces a new, super-powerful way to look at this bacterial city without destroying it. Here is how they did it, explained simply:

1. The "Freeze-Slice" Technique

Instead of making a smoothie, the scientists treated the bacterial city like a loaf of bread. They flash-froze it solid and then used a microscopic knife to slice it into incredibly thin layers (20 micrometers thick—thinner than a human hair). This allowed them to look at the city layer by layer, preserving the exact layout of the neighborhoods.

2. The "Super-Microscope" (Mass Spectrometry Imaging)

Usually, looking at proteins in a slice is like trying to read a book in the dark. The scientists used a special camera called MALDI-MSI (Mass Spectrometry Imaging).

• The Analogy: Imagine shining a laser on a tiny spot of the bacterial city. The laser doesn't just take a picture; it gently zaps the proteins there, turning them into tiny, charged messengers that fly into a super-sensitive detector.
• The Result: This detector acts like a high-tech barcode scanner. It reads the "weight" and "shape" of every protein in that tiny spot. Because the machine is so precise, it can tell the difference between two proteins that are almost identical, like telling the difference between a twin wearing a red hat and a twin wearing a blue hat.

3. The "Library" Match

To know what they were seeing, the scientists needed a dictionary. They created a custom "library" of all the possible proteins this bacteria could have, including all its tiny variations (called proteoforms).

• The Analogy: Think of a protein as a basic T-shirt. A "proteoform" is that same T-shirt with a specific stain, a missing button, or a rolled-up sleeve. The scientists matched the signals from their laser scanner against their library to identify exactly which "stained T-shirt" was in which part of the city.

What They Discovered: The City's Neighborhoods

By mapping these proteins, they found that the bacterial city is divided into distinct zones, much like a human city:

• The Inner Core (The Cannibal Zone): In the center of the colony, they found proteins associated with "cannibalism." These are toxic peptides that kill off older, weaker bacteria. It's like the city recycling its own resources to keep the strong survivors alive.
• The Outer Edge (The Retirement Home): On the very outside rim, they found proteins associated with sporulation. This is the bacteria's way of making "survival pods" (spores) to wait out bad times. It's like the city building bunkers on the perimeter because it senses a storm is coming.
• The Middle Ground: In between, they found ribosomal proteins (the machines that build other proteins) and other workers keeping the city running.

Why This Matters

Before this, we knew the bacteria could do these things. Now, we can see exactly where they are doing them and when.

• The Big Picture: This is like going from a blurry, black-and-white photo of a city to a high-definition, color 3D map where you can see the fire department actively fighting a fire in one district while the bakery is baking bread in another.
• The Future: This technology helps scientists understand how bacteria communicate, how they fight infections, and how they build communities. It could lead to better ways to stop bad bacteria (like those causing infections) or help us use good bacteria to build better materials.

In short: The scientists didn't just take a snapshot of the bacteria; they took a high-resolution, color-coded map of the bacterial city's soul, revealing that even a simple blob of slime is actually a complex, organized society with distinct neighborhoods and specialized workers.

Technical Summary

1. Problem Statement

While bulk proteomics can differentiate bacterial subpopulations, it lacks spatial resolution, failing to capture the heterogeneity inherent in microbial biofilms. Bacillus subtilis biofilms contain distinct microenvironments (e.g., central core vs. outer periphery) governed by differential gene expression, leading to varied phenotypes such as sporulation, colonization, and cannibalism.

• The Gap: Mass Spectrometry Imaging (MSI) has been successfully used for metabolites and lipids, but applying it to intact proteins (top-down proteomics) in microbes is challenging.
• Specific Challenges:
  • Theoretical proteomes of even model organisms like B. subtilis are only partially mapped regarding proteoforms (proteins with post-translational modifications [PTMs] and truncations).
  • Identifying detected peptides/proteins in complex microbial samples is difficult without specific libraries.
  • There is a lack of methods to visualize the spatial distribution of intact proteoforms to understand functional microenvironments.

2. Methodology

The authors developed a workflow combining Top-Down Proteomics (TDP) with Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging (MALDI-MSI) to map the spatial proteomic landscape of B. subtilis biofilms.

• Sample Preparation:

  • B. subtilis was cultured on MSgg agar to form biofilms.
  • Biofilms were embedded in agarose, snap-frozen, and cryosectioned into 20 µm thin slices on Indium Tin Oxide (ITO) slides.
  • A mild ethanol fixation and washing protocol was applied to preserve protein integrity without the need for extensive desalting (unlike mammalian tissue protocols).
  • Matrix deposition (2,5-DHA) was performed using an automated sprayer with specific parameters to enhance proteoform signal.

• Top-Down Proteomics (LC-MS/MS):

  • Bulk samples were prepared via bead beating, organic extraction (chloroform/methanol), and size-exclusion filtration.
  • Analysis was performed on Exploris 480 and Lumos Tribrid Orbitrap mass spectrometers.
  • Data was processed using TopPIC Suite for deconvolution and identification, generating a sample-specific experimental library of proteoforms.

• MALDI-MSI:

  • Imaging was conducted on a Q Exactive HF Orbitrap equipped with an Ultra-High Mass Range (UHMR) detector and an elevated pressure (EP) MALDI source.
  • Parameters: 20 µm spatial resolution, m/z range of 2,000–20,000, and a resolving power of ~45k at m/z 6,725.
  • Data Analysis: Spectra were averaged and baseline-corrected using Mozaic. Proteoform identification utilized IsoMatchMS for isotopic distribution matching against the TDP-derived library, requiring <5 ppm mass error and Pearson correlation ≥0.7.

3. Key Contributions

• First Spatial Proteoform Map: Demonstrated the feasibility of detecting and imaging intact proteins and proteoforms (up to ~13.5 kDa in the biofilm, though the instrument supports up to 20 kDa) directly from microbial biofilms.
• Sample-Specific Library Integration: Proved that coupling MALDI-MSI with a custom, sample-specific TDP library is essential for high-confidence annotation of PTMs and truncations in complex microbial samples.
• Differentiation of Subpopulations: Successfully visualized distinct spatial distributions of proteins corresponding to specific physiological states (cannibalism vs. sporulation) within a single biofilm colony.
• Methodological Pipeline: Established a robust pipeline for "spatial biotyping," where microbial phenotypes can be identified based on proteomic signatures rather than just taxonomy.

4. Key Results

• Detection Scale: Over 285 unique isotopic envelopes (proteoforms) were detected and annotated.
• Spatial Heterogeneity:
  • Central Core: Enriched with cannibalistic toxins (e.g., Sporulation Delaying Protein, SDP) and ribosomal proteins. This suggests active recycling of older generations to delay sporulation.
  • Outer Periphery: Dominated by sporulation-associated proteins, including Small Acid-Soluble Spore Proteins (SASP J and K) and Stage V sporulation protein S. This indicates the biofilm is in the early phases of sporulation at the edges.
• Proteoform Resolution:
  • Detected specific PTMs, such as initiator methionine truncations on ribosomal proteins and oxidation events.
  • Resolved isotopic signatures separated by only a few hundred milli-Daltons (mDa), highlighting the necessity of high mass-resolving power.
  • Identified a 7.1 kDa protein with a disulfide bond (Cys37↔Cys87), potentially annotated as uncharacterized protein YhjA or regulatory protein DegR, demonstrating the ability to detect structural features like disulfide bridges.
• Validation: The spatial distribution of SDP (core) vs. SASPs (periphery) aligns with known biological mechanisms where cannibalism delays sporulation, confirming the biological relevance of the spatial data.

5. Significance

• Mechanistic Insight: This study moves beyond simple metabolite mapping to provide a direct window into the proteomic mechanisms driving biofilm development, differentiation, and survival strategies.
• Universal Applicability: While demonstrated on B. subtilis, the methodology is applicable to any microbial system culturable on solid media, including co-cultures, mutants, and bioengineered strains.
• Future Directions: The work lays the groundwork for integrating spatial LCM-TDP (Laser Capture Microdissection) with MSI to analyze larger proteoforms and validate findings with fluorescence microscopy.
• Uncharacterized Proteins: The approach offers a pathway to functionally characterize unannotated proteins by correlating their spatial localization with known biological processes (e.g., linking a 7.1 kDa protein to protease activity or motility regulation).

In summary, this paper establishes proteoform-informed spatial proteomics as a powerful tool for resolving the biochemical heterogeneity of microbial communities, offering a new dimension to understanding how bacterial populations organize and differentiate within biofilms.