Creation of a genome-scale metabolic model for the fungal pathogen Cryptococcus neoformansSamantha Gonyea1, Dr. Amy Reese1, and Dr. Stephen Fong21Department of Biological Sciences, Cedar Crest College, Allentown, PA2Department of Chemical and Life Science Engineering, Virginia Commonwealth University, Richmond, VA
Why a model of C. neoformans?
Results and Conclusions
In recent years, Cryptococcus neoformans has become a growing fungal threat for individuals who are immunocompromised. The associated disease, cryptococcosis, affects the central nervous system and causes symptoms similar to meningitis. The pathogenicity of C. neoformans is mostly due to a complex polysaccharide capsule that evades the body’s immunological response. Many techniques are being used in order to better understand this unique capsule and the fungal biology of this organism. The research for this study is focused on a systematic and comprehensive evaluation of C. neoformans metabolism through the creation of a genome-based metabolic model. Constraint-based genome-scale metabolic modeling is a platform for tabulating biochemical information for an organism, and a completed model will allow researchers to perform in silico simulations to study gene and pathway usage in C. neoformans. Experimental testing of computational predictions may lead to further clues about the metabolic pathways of C. neoformans, specifically in the capsular synthesis and assembly pathways. The creation of an in silico metabolic model, in addition to other capsule-based research, may eventually aid in a greater understanding of how C. neoformans works. Current analysis of C. neoformans metabolism in comparison to that of a Saccharomyces cerevisiae model will be presented.
- Gap analysis shows that C. neoformans has a very complete central metabolism, specifically glycolysis and the TCA cycle (See Image 3).
- We are interested in obtaining a better understanding of capsular pathways, including capsule synthesis and cell wall attachment, as well as their connections to central fungal metabolism.
- This information may lead to an interest in cell wall genes associated with virulence that were not previously thought to be involved.
- New gene targets suggested by the in silico model can then be tested by experimental gene deletion studies to provide more information about capsule synthesis and cell wall attachment.
- Initial Analysis
- A partially annotated genome for Cryptococcus neoformans is located at The Institute for Genomic Research (TIGR) database.
- The known C. neoformans genes and their metabolic pathways were compared to the pre-existing genome-based model of Saccharomyces cerevisiae.
- From these comparisons, regions of incomplete annotation of the C. neoformans genome were identified for further gap analysis.
What is a metabolic model?
Image 3: The TCA cycle from the S. cerevisiae metabolic model’s central metabolism and the C. neoformans comparisons. C. neoformans pathways are highlighted in red.
- A quantitative in silico model representing all known metabolic pathways in a particular organism.
- These computer models are based on annotated genomic sequences and various other biochemical data.
- Model predictions simulate an organism’s metabolic behavior leading to hypotheses that can be tested experimentally.
- There are still some gaps present, such as transport mechanisms, that need to be verified experimentally before being included in the C. neoformans model.
- Gap Analysis
- To fill missing metabolic pathways in C. neoformans, known S. cerevisiae gene sequences were compared to the entire C. neoformans genome.
- The gene sequences corresponding to the gene proteins found in the S. cerevisiae model that were not in the C. neoformans database were retrieved from the Saccharomyces Genomic Database (http://db.yeastgenome.org/cgi-bin/seqTools).
- These S. cerevisiae sequences then were compared to the entire C. neoformans genome using a tblastx search on the TIGR BLAST website (http://tigrblast.tigr.org/er-blast/index.cgi?project=cna1) to locate possible equivalent genes.
- Found C. neoformans data was recorded in the excel worksheet of the S. cerevisiae metabolic model iND750 following the S. cerevisiae data (See Image 2).
- Those gene sequences with a smallest sum probability number of 1 x 10-3 or less were accepted as reactions in C. neoformans.
What are the next steps?
- To experimentally determine transport mechanisms present in C. neoformans.
- To use flux-balance analysis to simulate the metabolic behavior of C. neoformans.
- To study specific gene and pathway contributions to C. neoformans behaviors under diverse conditions, including polysaccharide capsule production.
Prior Studies Using Metabolic Models
- Escherichia coli
- The genome-scale E. coli model (iJE660) was used to predict growth behavior at the end of the adaptive evolution (Edwards et al. 2001, Ibarra et al. 2002, Fong et al. 2003, Fong & Palsson 2004).
- The genome-scale E. coli model correctly predicted two double gene deletions and one quadruple gene deletion strain as designs for increased output of lactic acid as a by-product of cellular growth (Fong et al. 2005).
Summary of Model Making Process
- Saccharomyces cerevisiae
- Growth patterns of S. cerevisiae predicted by the metabolic model were in agreement with experimental results 87.8% of the time (Famili et al. 2003).
- The metabolic model demonstrates eukaryotic compartmentalization to include extracellular space, cytosol, mitochondria, peroxisome, nucleus, Golgi apparatus, endoplasmic reticulum, and vacuoles (Duarte et al. 2004).
- S. cerevisiae, as a model yeast, is a good pre-existing model for research on another yeast, Cryptococcus neoformans.
C. neoformans annotated genome
Nutrients and Constraints
Image 4: A summary of the production of a metabolic model which was adapted from Figure 5 of Segre et al. (2003).
C. neoformans metabolic database
Flux Balance Analysis
In silico prediction
- Dr. Amy J. Reese and lab members, Cedar Crest College
- Dr. Stephen Fong, Virginia Commonwealth University
- The Bioinformatics and Bioengineering Summer Institute at Virginia Commonwealth University
What is Cryptococcus neoformans?
Duarte, N.C., M.J. Herrgard, and B.O. Palsson. 2004. Genome Research 14: 1298-1309.
Edwards, J.S., R.U. Ibarra, and B.O. Palsson. 2001. Nature Biotechnology 19: 125-130.
Famili, I., J. Forster, J. Nielsen, and B.O. Palsson. 2003. PNAS 100: 13134-13139.
Fong, S.S. and B.O. Palsson. 2004. Nature Genetics 36: 1056-1058.
Fong, S.S., A.P. Burgard, C.D. Herring, E.M. Knight, F.R. Blattner, C.D. Maranas, and B.O.
Palsson. 2005. Biotechnology and Bioengineering 91: 643-648.
Fong, S.S., J.Y. Marciniak, and B.O. Palsson. 2003. Journal of Bacteriology 185: 6400-6408.
Ibarra, R.U., J.S. Edwards, and B.O. Palsson. 2002. Nature 420: 186-189.
Segre, D., J. Zucker, J. Katz, X. Lin, P. D’Haeseleer, W.P. Rindone, P. Kharchenko, D.H. Nguyen, M.A. Wright, and G.M. Church. 2003. OMICS 7: 301-316.
- This pathogenic budding yeast is responsible for the disease cryptococcosis.
- The main virulence factor is a complex polysaccharide capsule (See Image 1).
- The metabolism of this fungus is less well understood compared to Saccharomyces cerevisiae.
Image 1: A scanning electron micrograph of budding C. neoformans provided by A. Casadevall. Visible spikes represent the polysaccharide capsule.
Image 2: An example of the iND750 excel worksheet using the pentose phosphate cycle metabolic pathway. Data includes abbreviations of the reaction, the reaction name, the actual reaction, enzyme classification (EC) number, subsystem, open reading frame, protein produced, C. neoformans TIGR annotation numbers, and smallest sum probability numbers for those sequences searched in BLAST.