Mathematical modeling of pathogenicity of Cryptococcus neoformans

Jacqueline Garcia¹, John Shea¹, Fernando Alvarez-Vasquez^1,2, Asfia Qureshi¹, Chiara Luberto¹, Eberhard O Voit³ & Maurizio Del Poeta^1,4,5

Article highlights

Synopsis

Extended Synopsis

Cryptococcus neoformans (Cn) is a fungal microbial pathogen that lives in the environment and in the gastrointestinal tract of several birds, pigeons in particular (Casadevall and Perfect, 1998). Upon inhalation of Cn spores or desiccated yeast cells, the fungus can grow in the extracellular space of the alveoli and in the intracellular environment of phagocytic cells, particularly alveolar macrophages (Levitz et al, 1999). Hence, Cn is considered a facultative intracellular pathogen. Thus, once in the lung, the fungus must adapt to two different environments: the extracellular space characterized by neutral/alkaline pH and the intracellular milieu of the phagolysosome characterized by acidic pH.

In recent years, we have found that a class of lipids, sphingolipids, represents a reservoir of molecules implicated in the regulation of fungal growth either in neutral/alkaline (Rittershaus et al, 2006; Kechichian et al, 2007) or acidic environments (Shea et al, 2006), in the modulation of Cn virulence factors, such as melanin production (Heung et al, 2004, 2005), and in the regulation of phagocytosis (Luberto et al, 2003; Mare et al, 2005; Tommasino et al, 2008). Interestingly, the shift of Cn cells from the extracellular to the intracellular compartment is particularly important because it changes the outcome of the infection in a severely immunocompromised host (Kechichian et al, 2007). Interestingly, the contribution of each Cn population (extracellular versus intracellular) to the outcome of infection is determined by the host immune status (Luberto et al, 2003). Therefore, the understanding of the mechanism(s) that regulate survival of Cn in both compartments (extracellular versus intracellular) may lead to novel therapeutic interventions based on the fine tune up of important Cn switches determined by the host immune status.

In this paper, a mathematical model representing the sphingolipid metabolic pathway of Cn was developed. It was tested to simulate sphingolipid adaptation to a shift from an alkaline to an acidic pH, to mimic the phagocytosis of Cn by macrophages (Figure 1). The model was designed and analyzed within the framework of the biochemical system theory, which uses power-law representations for all enzymatic and transport processes (Savageau, 1969a, b, 1976; Torres and Voit, 2002).

Figure 1

Model diagram of sphingolipid metabolism in Cn. Metabolites in boxes represent dependent variables that are defined through differential equations and are numbered from X₁ to X₁₉. Independent variables are numbered from X₁₀₀ to X₁₃₆. Solid arrows show flow of material. Plus signs associated with dotted arrows represent activation. The acylation state is coded as (1) C₂₆-CoA, (2) C₁₈-CoA, and (3) C₂₄-CoA; these are substrates for the DH-Cer synthase reaction or for the enzyme P-Cer synthase (see main text and Supplementary information for details). Dependent variables: Pal-CoA (X₁), palmitoyl-CoA; serine (X₂); KDHS (X₃), 3-ketodihydrosphingosine; DHS (X₄), dihydrosphingosine; dihydro-C₂₄ (X₅), dihydroceramide C₂₄; dihydro-C₂₆ (X₆), dihydroceramide C₂₆; dihydro-C₁₈ (X₇), dihydroceramide C₁₈; PHS (X₈), phytosphingosine; phyto-C₂₆ (X₉), phytoceramide C₂₆; phyto-C₂₄ (X₁₀), phytoceramide C₂₄; phyto-C₁₈ (X₁₁), phytoceramide C₁₈; Pma1 (X₁₂), newly synthesized Pma1; IPC-C₂₆ (X₁₃), inositol phosphorylceramide C₂₆; IPC-C₂₄ (X₁₄), inositol phosphorylceramide C₂₄; IPC-C₁₈ (X₁₅), inositol phosphorylceramide C₁₈; intracellular protons (X₁₆); ATP (X₁₇), adenosine-5'-triphosphate; palmitate (X₁₈); DAG (X₁₉), sn-1,2-diacylglycerol. Independent variables: palmitate ext (X₁₀₀), palmitate external; serine ext (X₁₀₁), serine external; palmitate transport (X₁₀₂); serine transport (X₁₀₃); Ac-CoA (X₁₀₄), acetyl CoA; C₂₆-CoA (X₁₀₅), very long-chain fatty acid (C₂₆-CoA); C₁₈-CoA (X₁₀₆), fatty acid (C₁₈-CoA); C₂₄-CoA (X₁₀₇), fatty acid (C₂₄-CoA); serine palmitoyltransferase (X₁₀₈); ADP, adenosine biphosphate (X₁₀₉); dihydro-CDase (X₁₁₀), dihydroceramide ceramidase; KDHS reductase (X₁₁₁), 3-ketodihydrosphingosine reductase; DH-Cer synthase (X₁₁₂), dihydroceramide synthase; phyto-CDase (X₁₁₃), phytoceramidase; hydroxylase (X₁₁₄); hydroxylase (X₁₁₅); P-Cer synthase (X₁₁₆), phytoceramide synthase; Pma1p (X₁₁₇), newly synthesized Pma1 in the ER; Sec61 (X₁₁₈), Sec61 as probable ER insertion protein; Isc1 (X₁₁₉), inositol phosphosphingolipid phospholipase C; PI (X₁₂₀), phosphatidylinositol; Ipc1 (X₁₂₁), inositol phosphorylceramide synthase; alternative respiration (X₁₂₂); NADHm (X₁₂₃), nicotinamide adenine dinucleotide; oxygen (X₁₂₄); Pma1-H⁺ATPase (X₁₂₅), synthesized plasma membrane H⁺-ATPase; H⁺ (X₁₂₆), protons external; ER–Golgi transport (X₁₂₇); H⁺transport (X₁₂₈), proton transport; SHMT (X₁₂₉) serine hydroxymethyl transferase; Golgi membrane (X₁₃₀); Pal-CoA synthase (X₁₃₁), palmitoyl-CoA synthase; ATP total (X₁₃₂); AMP (X₁₃₃), adenosine monophosphate; Golgi–ER transport (X₁₃₄); F₀F₁-ATPase (X₁₃₅), F₀F₁-ATP synthase; H⁺m (X₁₃₆), mitochondrial protons.

Full figure and legend (504K)Figures & Tables index

By coupling mathematical simulations using the model with experimental determinations, multiple factors were found to be required for adaptation of Cn to the acidic environment. In particular, experimental determinations showed that sphingolipid phytoceramide C₂₆ significantly increases when cells are shifted from alkaline to acidic pH (Table I), and this result was predicted by the model. Interestingly, production of phytoceramide C₂₆ is under the control of inositol phosphosphingolipid phospholipase C (Isc1) enzyme, because a isc1 mutant strain dramatically reduces the level of this sphingolipid. As expected, isc1 mutant has a growth defect at acidic pH (Supplementary Figure 1).

Table 1: Identification of ceramide species at 48 h of growth

Full tableFigures & Tables index

Moreover, the use of a different mutant of the sphingolipid pathway showed that phytoceramide C₂₆ is necessary but not sufficient for the adaptation process. In particular, downregulation of inositol phosphorylceramide synthase (Ipc1), which uses phytoceramide as a substrate, shows also a growth defect at low pH even though the levels of phytoceramide C₂₆ are not altered. Therefore, it is proposed that other lipids regulated by Ipc1, such as complex sphingolipids and/or diacylglycerol, may also be involved in this adaptation process.

Based on our simulations, the model suggests that the growth defect at low pH observed when Isc1 is deleted or Ipc1 downregulated is due to a decreased activity of the plasma membrane H⁺ATPase (Pma1), and the experimental findings (Table V) indeed support this prediction (Figure 1). We hypothesize that Isc1 regulates Pma1 through phytoceramide C₂₆, whereas Ipc1 regulates Pma1 through DAG and/or complex sphingolipids.

In conclusion, the mathematical model of sphingolipid metabolism helps to predict the adaptation of Cn in the host environments and contributes to a better understanding of its pathogenic traits.