Metabolic modelling as a powerful tool to identify critical components of Pneumocystis growth medium

Olga A. Nev, Elena Zamaraeva, Romain De Oliveira, Ilia Ryzhkov, Lucian Duvenage, Wassim Abou-Jaoudé, Djomangan Adama Ouattara, Jennifer Claire Hoving, Ivana Gudelj, Alistair J. P. BrownPLOS Computational Biology 2024 · DOI · OpenReview

Abstract

Establishing suitable in vitro culture conditions for microorganisms is crucial for dissecting their biology and empowering potential applications. However, a significant number of bacterial and fungal species, including Pneumocystis jirovecii, remain unculturable, hampering research efforts. P. jirovecii is a deadly pathogen of humans that causes life-threatening pneumonia in immunocompromised individuals and transplant patients. Despite the major impact of Pneumocystis on human health, we remain ignorant about the pathobiology of this fungus. This is largely due to the fact that its experimental dissection has been constrained by the inability to culture the organism in vitro. We present a comprehensive in silico genome-scale metabolic model of Pneumocystis growth and metabolism, with a view to identifying metabolic requirements and imbalances that hinder growth in vitro. We utilize recently published genome data and available information in the literature as well as bioinformatics and software tools to develop and validate the model. In addition, we employ Flux Balance Analysis and Reinforcement Learning approaches to make predictions regarding metabolic fluxes and to identify critical components of the Pneumocystis growth medium. Our findings offer insights into the biology of Pneumocystis and provide a novel strategy to overcome the longstanding challenge of culturing this pathogen in vitro.

Introduction

Pneumocystis jirovecii represents one of the most significant challenges in clinical mycology. As an opportunistic fungal pathogen, it is a primary cause of life-threatening pneumonia in immunocompromised populations, including HIV/AIDS patients and organ transplant recipients. Despite its clinical importance, the fungus has remained notoriously "unculturable" in laboratory settings for over 30 years. This obligate parasitism creates a massive bottleneck: without a stable in vitro culture, researchers cannot easily test new antifungals, study drug resistance mechanisms, or explore the basic pathobiology of the organism.

To address this, researchers have turned to computational biology to create a "digital twin" of the organism's metabolism. By reconstructing a genome-scale metabolic model (GEM), this study aims to identify the specific nutritional gaps that prevent Pneumocystis from surviving outside its host.

Methodology

GEM Reconstruction and Flux Balance Analysis

The research team developed a comprehensive metabolic network for Pneumocystis murina, a mouse-specific proxy for the human pathogen. Using the Constellab platform, they integrated genomic and proteomic data (PRJNA70803) to identify 736 enzymes and 839 chemical reactions. This reconstruction was benchmarked against well-studied fungi like Schizosaccharomyces pombe and Saccharomyces cerevisiae.

To simulate growth, the authors employed Flux Balance Analysis (FBA). This mathematical approach calculates the flow of metabolites through the metabolic network. By defining a "biomass equation"—a mathematical representation of all the essential components required to produce a new cell—the model can predict which nutrient combinations maximize growth flux.

AI-Driven Medium Optimization

A novel aspect of this study is the application of Reinforcement Learning (RL) to simplify complex growth media. Starting with "Merali’s medium," a complex 155-component recipe that supports limited growth, the team used a Proximal Policy Optimization (PPO) algorithm. The RL agent was trained to systematically remove non-essential nutrients while maintaining the predicted biomass flux. This allowed the researchers to distill a bloated recipe into a streamlined, essential nutrient set.

Key Metabolic Insights

The Scavenger Lifestyle

The model confirms that Pneumocystis has undergone significant genome reduction, a hallmark of obligate parasites. The fungus has lost the ability to synthesize all 20 essential amino acids and lacks the glyoxylate cycle and gluconeogenesis pathway. Consequently, it is entirely dependent on scavenging these building blocks from the host's lung environment.

Lipid and Sterol Requirements

The simulations highlighted a critical dependency on external lipids. Specifically, cholesterol and fatty acids such as palmitic, stearic, and oleic acids were identified as essential for maximizing biomass. The model suggests that supplementing existing media with these specific lipids could significantly improve in vitro viability.

Resolving the SAM Dilemma

For years, the scientific community has debated whether Pneumocystis is an auxotroph for S-adenosyl-L-methionine (SAM). The metabolic model provides a nuanced answer: the fungus requires either SAM or L-methionine (or both) to sustain growth. This flexibility explains why previous experimental results regarding SAM requirements have been inconsistent.

Results and Validation

The final validated model consists of 774 reactions and 1,027 metabolites. Through the RL optimization process, the team successfully reduced the required nutrient set by 80%, identifying a core group of 32 essential nutrients. This streamlined medium includes:

  • 11 Amino acids
  • 4 Vitamins
  • 10 Lipids and sterols

The model's predictive accuracy was validated through in silico gene knockouts. For example, knocking out serine biosynthesis enzymes (SHMT and PSP) correctly predicted a total cessation of growth in the absence of external serine and glycine, matching known biological constraints.

Conclusion and Future Directions

This study provides the first comprehensive metabolic roadmap for Pneumocystis. By identifying a minimal set of 32 nutrients, the researchers have provided a concrete starting point for developing a functional in vitro culture medium.

While the steady-state assumptions of FBA do not account for dynamic protein signaling or gene expression changes, the model serves as a powerful predictive framework. Future work will focus on integrating transcriptomic and proteomic data to create a dynamic "multi-omics" model and, crucially, testing the predicted 32-nutrient medium in physical laboratory settings to finally break the barrier of Pneumocystis culturability.

Summary updated

Abstract

Establishing suitable in vitro culture conditions for microorganisms is crucial for dissecting their biology and empowering potential applications. However, a significant number of bacterial and fungal species, including Pneumocystis jirovecii, remain unculturable, hampering research efforts. P. jirovecii is a deadly pathogen of humans that causes life-threatening pneumonia in immunocompromised individuals and transplant patients. Despite the major impact of Pneumocystis on human health, limited progress has been made in dissecting the pathobiology of this fungus. This is largely due to the fact that its experimental dissection has been constrained by the inability to culture the organism in vitro. We present a comprehensive in silico genome-scale metabolic model of Pneumocystis growth and metabolism, to identify metabolic requirements and imbalances that hinder growth in vitro. We utilise recently published genome data and available information in the literature as well as bioinformatics and software tools to develop and validate the model. In addition, we employ relaxed Flux Balance Analysis and Reinforcement Learning approaches to make predictions regarding metabolic fluxes and to identify critical components of the Pneumocystis growth medium. Our findings offer insights into the biology of Pneumocystis and provide a novel strategy to overcome the longstanding challenge of culturing this pathogen in vitro.
BibTeX
@article{nev2024metabolic,
  author = {Olga A. Nev and Elena Zamaraeva and Romain De Oliveira and Ilia Ryzhkov and Lucian Duvenage and Wassim Abou-Jaoudé and Djomangan Adama Ouattara and Jennifer Claire Hoving and Ivana Gudelj and Alistair J. P. Brown},
  title = {{Metabolic modelling as a powerful tool to identify critical components of Pneumocystis growth medium}},
  journal = {PLOS Computational Biology},
  year = {2024},
  doi = {10.1371/journal.pcbi.1012545},
}
Figure