Protein without farms: What comparative genomics reveals about ''Power-to-Food'' microbes

This study utilizes comparative genomics to characterize the distinct genomic architectures, metabolic capabilities, and safety profiles of two hydrogen-oxidizing bacteria, *Cupriavidus necator* H16 and *Xanthobacter* sp. SoF1, thereby establishing a scientific foundation for selecting and engineering robust, food-grade microbes for scalable, land-independent protein production on Earth and in space.

The Gist

Imagine the future of food not as a field of wheat or a herd of cows, but as a giant, high-tech factory where the ingredients are simply air, water, and electricity. This is the promise of "Power-to-Food."

This paper is a deep-dive comparison between two microscopic "chefs" (bacteria) that are best suited to run these factories. The goal? To figure out which one is the better candidate to turn carbon dioxide (CO₂) and hydrogen gas (H₂) into nutritious protein for humans, without needing any farmland.

Here is the story of the two contenders, explained simply:

The Two Contenders: The "Modular Giant" vs. The "Streamlined Pro"

The researchers compared two specific bacteria:

1. Cupriavidus necator H16 (H16): The veteran.
2. Xanthobacter sp. SoF1 (SoF1): The modern industrial star (used to make "Solein," a protein already approved for sale in Singapore).

Think of them like two different types of cars designed for the same race.

1. The Blueprint (Genome Architecture)

• H16 is like a massive, modular mansion. It has a huge genome (the instruction manual) split across three different "books" (two chromosomes and a giant plasmid). It's packed with extra tools, backup systems, and complex wiring. It's very flexible and can handle weird conditions, but it's complicated to manage.
• SoF1 is like a sleek, single-story smart home. Its instruction manual is much smaller and fits on a single "book" (one chromosome). It's streamlined, efficient, and does exactly what it needs to do without the extra clutter.

The Takeaway: If you want a machine that is easy to control and prove is safe, the "smart home" (SoF1) is easier to inspect. If you want a machine with built-in backup systems for extreme chaos, the "mansion" (H16) has more redundancy.

2. The Fuel and Food (Metabolism)

Both bacteria are amazing because they can eat "air." They take CO₂ (exhaled by humans or factories) and H₂ (made from renewable electricity) to build protein.

• The Nitrogen Problem: To build protein, you need nitrogen.
  • H16 is like a guest who forgets to bring a dish to a potluck. It cannot make its own nitrogen; it needs someone else to give it ammonia or nitrate (fertilizer) in the food mix.
  • SoF1 is the guest who brings their own dish. It has a special "nitrogen-fixing" tool kit (genes called nif) that lets it grab nitrogen directly from the air. This is a huge advantage for space missions or closed-loop systems where you can't easily bring in fertilizer.

3. The Safety Check (Is it safe to eat?)

Before you can eat a new food, you have to prove it won't kill you. The researchers ran a "background check" on both bacteria's DNA to look for:

• Weapons: Genes that make toxins (poisons).
• Armor: Genes that make them resistant to antibiotics (so if you take medicine, the bacteria doesn't become a super-bug).
• Spyware: Genes that help them invade human cells.

The Verdict: Both bacteria passed the test with flying colors. Neither has the "weapons" or "spyware" that would make them dangerous. They are essentially "clean" bacteria. However, H16 has a few more "grey areas" (like some genetic tools that look like they could be dangerous but probably aren't in this context), whereas SoF1 is cleaner and simpler.

4. The "Secret Sauce" (Secondary Metabolism)

Bacteria sometimes make weird chemicals to fight off other bacteria or survive stress.

• H16 has a big toolbox of these secret chemicals. While this helps it survive in the wild, it means scientists have to be careful to ensure none of these weird chemicals end up in your food.
• SoF1 has a smaller, more focused toolbox. It makes fewer weird chemicals, which makes it easier to guarantee the final food product is pure and consistent.

Why Does This Matter?

This paper is like a mechanic comparing two engines to see which one is best for a new type of car.

• For Earth: We need to feed 10 billion people without destroying the planet. These bacteria can make protein without using land, water, or fertilizer. SoF1 looks like the winner for immediate food production because it's simpler, safer, and can even grab nitrogen from the air.
• For Space: Astronauts need to recycle everything. They breathe out CO₂, and they can make H₂ from water. SoF1 is the perfect "space chef" because it can turn that recycled air and water into food, even grabbing nitrogen from the air, closing the loop completely.

The Bottom Line

The researchers concluded that while H16 is a brilliant, complex model for scientists to study how these bacteria work, SoF1 is the better choice for actually making food. It's the "streamlined production model" that is ready for the factory floor, offering a safer, more efficient path to "protein without farms."

In short: We are moving from farming the land to farming the air, and these bacteria are the tractors of the future.

Technical Summary

1. Problem Statement

Conventional agriculture faces critical planetary boundaries, including excessive land/water use, greenhouse gas emissions, and nitrogen cycle disruption. To meet the protein demands of a growing global population (~9.7 billion by 2050) without expanding arable land, "Power-to-Food" technologies are emerging. These utilize Hydrogen-Oxidizing Bacteria (HOB) to convert renewable $H_2$, $CO_2$, and minerals into protein-rich biomass via gas fermentation.

However, the selection of optimal HOB chassis for industrial food production and closed-loop space life support systems lacks a rigorous, genome-level comparative framework. Two leading strains, Cupriavidus necator H16 (a well-characterized model organism) and Xanthobacter sp. SoF1 (the production strain for Solar Foods' Solein®), have not been systematically compared regarding their metabolic capabilities, genome architecture, and safety profiles for food-grade applications.

2. Methodology

The authors employed a comprehensive comparative genomics pipeline involving:

• Data Sources: Re-annotation of the complete multipartite genome of H16 (RefSeq) and the single-replicon draft genome of SoF1 (provided by Solar Foods) using Prokka and Bakta.
• Pan-Genome Analysis: Utilization of PEPPAN to classify genes into core (shared) and accessory (strain-specific) sets, followed by functional annotation via eggNOG/COG categories.
• Metabolic Reconstruction: Multi-layer pathway analysis using KEGG Orthology (KOfamScan), Pathway Tools (PathoLogic) to generate strain-specific PGDBs (H16Cyc, SoF1Cyc), and ModelSEED. This allowed for the comparison of $CO_2$ fixation, $H_2$ oxidation, and nitrogen metabolism pathways.
• Safety Screening: A tiered, strict-threshold screening for:
  • Antimicrobial Resistance (AMR): Using RGI (CARD) and AMRFinderPlus (≥90% identity/coverage).
  • Virulence Factors: Screening against VFDB and checking for secretion systems (T3SS, T4SS, T6SS) via MacSyFinder.
  • Toxins: Large-scale homology searches against DBETH and Toxinome using DIAMOND.
  • Mobility: Analysis of integrons and plasmid-like elements via IntegronFinder and MOB-suite.
  • Secondary Metabolism: Prediction of Biosynthetic Gene Clusters (BGCs) using antiSMASH.
• Visualization: Genome architecture mapped via Circos and MUMmer for synteny; pan-genome distribution visualized via custom Python scripts.

3. Key Results

A. Genome Architecture and Composition

• H16: Possesses a large, multipartite genome (~7.41 Mb) consisting of two chromosomes and a megaplasmid (pHG1). It has a lower GC content (66.3%) and higher gene count (6,853 genes).
• SoF1: Features a compact, single-replicon genome (~4.91 Mb) with a slightly higher GC content (67.8%) and fewer genes (4,520 genes).
• Core vs. Accessory: Despite the size difference, both share a conserved "core" autotrophic backbone (~9–11% of genes). However, H16 has a significantly larger accessory genome (56.9% strain-specific genes) compared to SoF1 (32.5%), indicating H16 carries more specialized, potentially environmental adaptation traits.

B. Metabolic Capabilities

• Carbon & Hydrogen: Both strains possess the complete Calvin–Benson–Bassham (CBB) cycle for $CO_2$ fixation and multiple [NiFe]-hydrogenase systems for $H_2$ oxidation. H16 shows duplicated CBB operons (one on pHG1, one on Chr2), suggesting regulatory redundancy.
• Nitrogen Economy (Critical Difference):
  • SoF1: Encodes a complete nitrogen fixation module (nifHDK) and nitrate assimilation genes, enabling growth on atmospheric $N_2$ or nitrate without external fixed nitrogen.
  • H16: Lacks nif genes. It relies on external ammonia/ammonium and possesses genes for denitrification (reducing nitrate to $N_2$) rather than assimilation, offering flexibility in oxygen-limited conditions but requiring fixed nitrogen inputs.
• Secondary Metabolism: Both strains have limited BGCs. SoF1 encodes clusters for osmoprotection (NAGGN) and PQQ. H16 has a broader repertoire including siderophores and $\beta$-lactones, likely for environmental competition rather than toxin production.

C. Safety Profile

Under strict regulatory thresholds (≥90% identity/coverage):

• AMR: No acquired, clinically relevant AMR genes were detected in either strain.
• Virulence: No Type III or IV secretion systems (T3SS/T4SS) were found. No high-confidence virulence factors or canonical foodborne exotoxins were identified.
• Toxins: Zero matches to major toxin families. H16 showed weak hits for RTX-like and Tc-like proteins, but these lacked the necessary genomic context (e.g., secretion modules) to be considered active toxins.
• Conclusion: Both strains exhibit a "low-risk" genetic profile suitable for food-grade qualification.

4. Key Contributions

1. First Comparative Genomics of Food-Grade HOB: Provides the first head-to-head genomic comparison between the industrial workhorse H16 and the commercial production strain SoF1.
2. Strain Selection Framework: Establishes that while both strains share the core "power-to-food" engine, their genomic architectures dictate different engineering strategies:
   • SoF1 is identified as the superior candidate for closed-loop, resource-constrained systems (e.g., space life support) due to its compact genome and ability to fix atmospheric nitrogen.
   • H16 remains valuable as a modular chassis for mechanistic studies and engineering robustness due to its redundant regulatory networks and multipartite genome.
3. Safety-by-Design Validation: Demonstrates that rigorous in silico safety screening can support regulatory dossiers (GRAS/Novel Food) by confirming the absence of virulence and acquired resistance mechanisms.
4. Actionable Engineering Targets: Identifies specific metabolic levers for optimization, such as attenuating PHB (polyhydroxybutyrate) storage pathways to maximize protein yield and managing secondary metabolite expression to ensure compositional consistency.

5. Significance

This study bridges the gap between fundamental microbiology and industrial food biotechnology. It validates the safety and feasibility of HOB for "protein without farms," offering a blueprint for scalable, land-free protein production.

• Terrestrial Impact: Supports the transition to sustainable protein sources that decouple food production from climate variability and fertilizer dependency.
• Space Exploration: Provides a critical genetic blueprint for Bioregenerative Life Support Systems (BLSS), where SoF1's ability to recycle crew $CO_2$ and $N_2$ into edible biomass is essential for long-duration missions.
• Regulatory Pathway: The study outlines a "genomics-first" qualification workflow that can accelerate the approval of novel microbial food ingredients by proactively addressing safety concerns through genome sequencing and annotation.