Systematic assessment of the impact of targeted selection methods and environment-mimicking culture conditions on fungal natural product libraries

This study systematically evaluates how fungal selection strategies, phylogenetic classification, and environment-mimicking culture conditions influence chemical diversity in natural product libraries, providing practical guidelines to optimize library design for more efficient drug discovery.

The Gist

Imagine you are a treasure hunter looking for new, magical potions (which scientists call natural products) hidden inside a vast forest of mushrooms (fungi). These potions could become the next life-saving medicine.

For years, scientists have been trying to build the ultimate "library" of mushroom extracts to find these potions. But they've been facing two big problems:

1. The "Needle in a Haystack" Problem: As their libraries get bigger, they keep finding the same old potions over and over, wasting time and money.
2. The "How to Collect" Dilemma: They aren't sure how to pick the best mushrooms to grow. Should they pick mushrooms from every different state? Should they pick mushrooms that look very different from each other? Or should they just grab them randomly?

This paper is like a massive experiment to figure out the best way to build this library and how to grow the mushrooms to get the most magical potions out of them.

Part 1: How to Pick the Mushrooms (The Selection Strategy)

The researchers tested two popular "smart" ways to pick mushrooms, comparing them to just picking them at random.

1. The "World Traveler" Strategy (Geography)

• The Idea: "If we pick mushrooms from Alaska, Hawaii, and Florida, they will be so different because of their different climates that they will make totally different potions."
• The Reality Check: The scientists tried this, and it didn't work. They found that picking mushrooms from all over the map didn't give them more new potions than just picking them randomly. It turns out, mushrooms from different places often make the same stuff, and you can find a huge variety of potions just by looking closely in one specific neighborhood.
• The Analogy: Imagine trying to collect unique coins. You might think you need to travel to 50 different countries to get 50 unique coins. But this study says you could probably get just as many unique coins by digging deep into just one country's coin collection. Traveling far and wide is a lot of work for no extra reward.

2. The "Family Tree" Strategy (Phylogeny)

• The Idea: "If we pick mushrooms that are very distant cousins on the family tree, they will have very different DNA and therefore make different potions."
• The Reality Check: This also didn't work. Even when they picked mushrooms that were genetically very different from each other, they didn't get more new chemical recipes than if they had just picked them randomly.
• The Analogy: It's like thinking that because two people have very different eye colors or hair textures, they must speak completely different languages. The study found that even "distant cousins" in the fungal world often speak the same chemical language.

The Big Takeaway: Stop overthinking the selection! Randomly picking mushrooms is just as good as (and cheaper than) trying to be a "smart" geographer or geneticist.

---

Part 2: How to Grow the Mushrooms (The Culturing Strategy)

Once you have the mushrooms, you have to grow them in a lab to get the potions. Scientists have been trying to trick the mushrooms into making more or new potions by changing their environment. This is called the OSMAC approach (One Strain, Many Compounds).

The researchers tested three "tricks" to see if they could wake up the mushrooms' hidden talents:

1. The "Soil Soup" Trick

• The Idea: Mushrooms grow in dirt. If we add real dirt water to their food, they might think they are back home and start making cool stuff.
• The Result: It worked, but only for a short time. At the very beginning (7 days), the mushrooms made a burst of new potions. But by day 14, the effect wore off.
• The Analogy: It's like giving a plant a sudden burst of fertilizer. It grows fast for a week, but then it goes back to normal. It's a quick fix, not a long-term solution.

2. The "Bacteria Scare" Trick (LPS)

• The Idea: In nature, fungi often fight bacteria. If we add a piece of bacteria skin (LPS) to their food, maybe the fungus will get scared and make defensive weapons (potions).
• The Result: It backfired. Instead of making more potions, the mushrooms actually made fewer. They seemed stressed and shut down their production.
• The Analogy: It's like trying to get a musician to play a new song by yelling at them. Instead of playing a new song, they just stop playing entirely.

3. The "Secret Signal" Trick (Diketopiperazines)

• The Idea: Fungi talk to each other using chemical signals. If we add these signals (called DKPs) to their food, maybe they will think they are in a crowded party and start making more complex chemistry.
• The Result: This was the winner! Specifically, one type of signal (DKP 2C) caused the mushrooms to produce 49% more new chemical structures after two weeks.
• The Analogy: This is like putting on some cool music and turning on the lights at a party. The mushrooms (the guests) got excited, started dancing, and created a whole new vibe. It took a little longer to kick in (14 days), but the results were amazing.

---

The Final Verdict: What Should Scientists Do?

This paper is a huge "aha!" moment for the scientific community.

1. Stop wasting time on complex selection: You don't need to fly to the Amazon or sequence DNA to pick your mushrooms. Just grab a random bunch; you'll get just as much variety.
2. Focus on the "Party Trick": Instead of worrying about where the mushroom came from, focus on how you grow it. Adding specific chemical signals (like the DKP trick) is the best way to unlock new medicines.

In a nutshell: If you want to find new magic potions in a mushroom library, don't worry about the map or the family tree. Just pick your mushrooms randomly, feed them the right "party signal," and wait for the magic to happen!

Technical Summary

1. Problem Statement

Natural products (NPs) derived from fungi are a critical source of novel bioactive molecules for drug discovery. However, the field faces a "needle in a haystack" problem: as crude extract libraries grow, the rate of discovering novel chemical scaffolds diminishes due to the frequent rediscovery of known compounds.

• The Bottleneck: Current library curation relies heavily on "rational" selection strategies (e.g., maximizing geographic spread or phylogenetic diversity) and "One Strain Many Compounds" (OSMAC) culturing approaches.
• The Gap: There is a lack of systematic, large-scale data evaluating whether these rational strategies actually outperform random sampling in expanding chemical diversity. Furthermore, the generalizability of environmental mimicry (OSMAC) across diverse fungal taxa remains poorly characterized, with many studies reporting only isolated "success stories" rather than overall collection impacts.

2. Methodology

The authors conducted a comprehensive meta-analysis and experimental study using untargeted LC-MS/MS and Feature-Based Molecular Networking (FBMN) via GNPS.

A. Sample Selection Analysis (Geographic & Phylogenetic)

• Datasets:
  1. Trichoderma Collection: 835 samples covering all NOAA climate zones in the US.
  2. Multi-Genera Collection: 1,439 samples representing 141 different genera across the US.
• Approach: The authors simulated "rational" selection strategies (prioritizing maximum geographic spread or maximum phylogenetic distance) and compared the resulting scaffold accumulation curves against random selection baselines (25 iterations each).
• Metrics: Chemical diversity was quantified by counting unique molecular scaffolds (clusters in FBMN) and features.

B. Culturing Condition Experiments (OSMAC)

• Subjects: 84 fungal isolates from three genera: Chaetomium, Clonostachys, and Purpureocillium.
• Interventions: Cultures were subjected to three environmental mimicry strategies compared to a control:
  1. Soil Extract: Aqueous extract of local soil added to media.
  2. Lipopolysaccharide (LPS): Added at 0.06, 0.6, and 1.8 ng/mL to mimic bacterial presence.
  3. Diketopiperazines (DKPs): Five synthesized DKPs (quorum sensing molecules) added at 100 nM and 10 µM.
• Timepoints: Metabolomes were profiled at 7 days and 14 days.
• Analysis: Untargeted LC-MS/MS followed by data processing (MZmine), normalization, and statistical analysis (Wilcoxon rank-sum, Random Forest/Boruta for feature selection).

3. Key Contributions

• Quantitative Validation of Selection Strategies: Provided the first large-scale, systematic comparison showing that rational selection based on geography or phylogeny does not significantly outperform random sampling in terms of scaffold accumulation.
• OSMAC Efficacy Mapping: Systematically evaluated the impact of specific environmental cues (soil, LPS, DKPs) across multiple genera and timepoints, moving beyond anecdotal success stories to provide generalizable guidelines.
• Biomass Proxy Correlation: Demonstrated that total raw MS signal intensity strongly correlates with chemical diversity (feature count), suggesting biomass can serve as a proxy for chemical output in screening.

4. Key Results

A. Selection Strategies (Geography & Phylogeny)

• Geographic Spread: Prioritizing samples from distinct NOAA climate regions did not accelerate the accumulation of chemical diversity compared to random selection. While some southern regions showed slightly higher unique scaffold counts, the overall overlap of scaffolds across regions was substantial.
• Phylogenetic Distance: Selecting isolates to maximize phylogenetic distance (using ITS and tef sequencing) yielded accumulation curves nearly identical to, or in the multi-genera case, slightly worse than, random selection.
• Conclusion: Intuitive strategies of "covering more ground" (geographically or evolutionarily) do not guarantee faster discovery of novel chemotypes. Concentrated sampling in productive niches may be more efficient than broad, shallow sampling.

B. Culturing Conditions (OSMAC)

• Soil Extract:
  • Effect: Produced a significant, transient boost in scaffold diversity at 7 days (approx. +32% increase).
  • Duration: The effect largely disappeared by day 14, except in Purpureocillium.
• Lipopolysaccharide (LPS):
  • Effect: Generally suppressed scaffold diversity (approx. -39% at 7 days).
  • Mechanism: LPS triggered the production of some specific metabolites (e.g., fatty amides) but inhibited the production of others, resulting in a net loss of diversity.
• Diketopiperazines (DKPs):
  • Effect: Showed the most promising and tunable results.
  • Specifics: DKP 2C at 10 µM significantly increased scaffold diversity by 49% at 14 days.
  • Timing: Unlike soil/LPS, DKP effects were often more pronounced at the later timepoint (14 days) and were concentration-dependent.
  • Variability: Responses were genus-specific (e.g., Purpureocillium was highly sensitive; Clonostachys was less responsive).

5. Significance and Implications

• Resource Optimization: The study suggests that natural product discovery programs can save significant resources by abandoning complex, labor-intensive geographic or phylogenetic selection protocols in favor of random sampling, which yields comparable chemical diversity.
• Strategic Culturing: Rather than relying on a single "magic bullet" culture condition, the study highlights that DKP supplementation (specifically DKP 2C) is a highly effective, tunable strategy for enhancing scaffold diversity, particularly when applied at later timepoints (14 days).
• Future Directions: The findings advocate for a shift in library design: prioritize cultivation-based interventions (OSMAC) over selection-based strategies. Researchers should focus on optimizing culture conditions to unlock cryptic biosynthetic gene clusters rather than expending effort on broad, rational sample selection that offers diminishing returns.

In summary, this paper challenges long-held assumptions in fungal natural product discovery, providing data-driven evidence that how fungi are cultured (environmental mimicry) is a more powerful lever for chemical diversity than which fungi are selected (geography/phylogeny).