13C-metabolic flux ratio and novel carbon path analyses confirmed that Trichoderma reesei uses primarily the respirative pathway also on the preferred carbon source glucose

Abstract

Background: The filamentous fungus Trichoderma reesei is an important host organism for industrial enzyme production. It is adapted to nutrient poor environments where it is capable of producing large amounts of hydrolytic enzymes. In its natural environment T. reesei is expected to benefit from high energy yield from utilization of respirative metabolic pathway. However, T. reesei lacks metabolic pathway reconstructions and the utilization of the respirative pathway has not been investigated on the level of in vivo fluxes.

Results: The biosynthetic pathways of amino acids in T. reesei supported by genome-level evidence were reconstructed with computational carbon path analysis. The pathway reconstructions were a prerequisite for analysis of in vivo fluxes. The distribution of in vivo fluxes in both wild type strain and cre1, a key regulator of carbon catabolite repression, deletion strain were quantitatively studied by performing 13C-labeling on both repressive carbon source glucose and non-repressive carbon source sorbitol. In addition, the 13C-labeling on sorbitol was performed both in the presence and absence of sophorose that induces the expression of cellulase genes. Carbon path analyses and the 13C-labeling patterns of proteinogenic amino acids indicated high similarity between biosynthetic pathways of amino acids in T. reesei and yeast Saccharomyces cerevisiae. In contrast to S. cerevisiae, however, mitochondrial rather than cytosolic biosynthesis of Asp was observed under all studied conditions. The relative anaplerotic flux to the TCA cycle was low and thus characteristic to respiratory metabolism in both strains and independent of the carbon source. Only minor differences were observed in the flux distributions of the wild type and cre1 deletion strain. Furthermore, the induction of the hydrolytic gene expression did not show altered flux distributions and did not affect the relative amino acid requirements or relative anabolic and respirative activities of the TCA cycle.

Conclusion: High similarity between the biosynthetic pathways of amino acids in T. reesei and yeast S. cerevisiae was concluded. In vivo flux distributions confirmed that T. reesei uses primarily the respirative pathway also when growing on the repressive carbon source glucose in contrast to Saccharomyces cerevisiae, which substantially diminishes the respirative pathway flux under glucose repression.

Figure 1

Origins of proteinogenic amino acids. The origins of the carbon backbones of the proteinogenic amino acids utilized in METAFoR analysis [26] and for which the biosynthetic pathways were reconstructed by computational pathway analysis method ReTrace [21]. The amino acids for which the biosynthetic pathway was not directly found by ReTrace are denoted in red italics. The amino acid carbons are denoted in the following way: a = α, b = β, g = γ, d = δ, e = ε, ksi = ξ.

Figure 2

Comparison of the fractions of corresponding intact bonds in amino acid precursors Oaa, Oga and Pep in T. reesei. The data is taken from all replicates of fractional [U-¹³C]glucose or [U-¹³C]sorbitol experiments performed (1-3 wild type on glucose, 4-6 Δcre1 on glucose, 7-9 wild type on sorbitol, 10-12 wild type on sorbitol (sophorose experiment control), 13-15 wild type on sorbitol (sophorose induction), 16-17 Δcre1 on sorbitol (sophorose experiment control), 18-19 Δcre1 on sorbitol (sophorose induction)). Oaa data was detected from Asp-Cα, -Cβ and Thr-Cα, Oga data from Glu-Cα, -Cβ and Pro-Cα, -Cβ and Pep data from Phe and Tyr-Cα. Fractions of intact bonds in Oaa, Oga and Pep were calculated from combinations of fragmentomers. A) OAA_x1x is the fraction of Oaa molecules with an intact bond at C2-C3, OGA_x1xx is the fraction of Oga molecules with an intact bond at C2-C3 and PEP_x1 is the fraction of Pep molecules with an intact bond at C2-C3. B) OAA_xx1 is the fraction of Oaa molecules with an intact bond at C3-C4 and OGA_1xxx is the fraction of Oga molecules with an intact bond at C1-C2. Error bars are ± SEMs. The carbon chain of Oaa_mit remains intact in the TCA cycle except that C1 is cleaved in the synthesis of Oga. Almost the entire labeling pattern of Oaa_mit can be assessed from the labeling pattern determined for Oga. If Asp and Thr synthesis originates from Oaa_mit, the fractions of corresponding Oaa and Oga intact fragments in the figures should match.

Figure 3

Effect of the reversible glycine cleavage pathway. Effect of the reversible glycine cleavage pathway in T. reesei wild type (wt) and Δcre1 strains on the ¹³C-labeling pattern of Ser. The fraction of Ser-Cα f(1) fragmentomer from the total pool of Ser, compared to the corresponding fraction in Pep (C2) observed from Phe and Tyr-Cα f(1) fragmentomers. Error bars are ± SEMs.

Figure 4

¹³C-His-Cβ centered contiguous ¹³C-fragments in T. reesei. Fractions of ¹³C-His-Cβ centered contiguous ¹³C-fragments in T. reesei in different genetic, wild type (wt) and Δcre1 mutant. Cultures were grown on glucose (glucose repressed) or sorbitol (derepressed). Sorbitol grown cultures were grown with or without induction of cellulase gene expression by the addition of sophorose to some cultures. His-Cβ f(1) denotes fragments with C-C bonds cleaved on both sides of Cβ, His-Cβ f(2) and f(2*) denote fragments with Cβ-Cα and Cβ-Cγ preserved, respectively, and His- Cβ f(3) denotes fragments were both bonds are intact. Error bars represent ± SEM.

Figure 5

T. reesei growth curves. Growth curves of T. reesei wild type (wt) and Δcre1 strains (A) on glucose and (B) on sorbitol. Error bars are standard deviations of three replicates.

Figure 6

Metabolic network model. Eukaryotic central carbon metabolism network model [26].