Functional characterization of thiolase-encoding genes from Xanthophyllomyces dendrorhous and their effects on carotenoid synthesis

Abstract

Background: The basidiomycetous yeast Xanthophyllomyces dendrorhous has been described as a potential biofactory for terpenoid-derived compounds due to its ability to synthesize astaxanthin. Functional knowledge of the genes involved in terpenoid synthesis would create opportunities to enhance carotenoid production. A thiolase enzyme catalyzes the first step in terpenoid synthesis.

Results: Two potential thiolase-encoding genes were found in the yeast genome; bioinformatically, one was identified as an acetyl-CoA C-acetyltransferase (ERG10), and the other was identified as a 3-ketoacyl Co-A thiolase (POT1). Heterologous complementation assays in Saccharomyces cerevisiae showed that the ERG10 gene from X. dendrorhous could complement the lack of the endogenous ERG10 gene in S. cerevisiae, thereby allowing cellular growth and sterol synthesis. X. dendrorhous heterozygous mutants for each gene were created, and a homozygous POT1 mutant was also obtained. This mutant exhibited changes in pigment composition and higher ERG10 transcript levels than the wild type strain.

Conclusions: The results support the notion that the ERG10 gene in X. dendrorhous is a functional acetyl-CoA C-acetyltransferase essential for the synthesis of mevalonate in yeast. The POT1 gene would encode a functional 3-ketoacyl Co-A thiolase that is non-essential for cell growth, but its mutation indirectly affects pigment production.

Keywords: Astaxanthin; Carotenoids; Functional complementation; Mevalonate; Sterols; Thiolase.

Fig. 1

Phylogenetic tree of ERG10A and ERG10B thiolases from X. dendrorhous compared to other organisms. The unrooted tree was created in MEGA 6.0 using the neighbor-joining method [20] with the following amino acid sequences. Mitochondrial thiolase/acetyl-CoA C-acetyltransferase: B. taurus [NP_001039540.1], H. sapiens [BAA01387.1], C. lupus familiaris [XP_546539.2], R. norvegicus [NP_058771.2], G. gallus [NP_001264708.1]. Cytoplasmic thiolase/acetyl-CoA C-acetyltransferase: S. pombe [Q9UQW6.1], B. graminis f. sp. tritici 96224 [EPQ61678.1], S. cerevisiae [P41338.3], N. tabacum [AAU95618.1], Z. mays [NP_001266315.1], A. thaliana [Q9FIK7.1]. Mitochondrial thiolase/3-ketoacyl-CoA thiolase: B. taurus [NP_001030419.1], M. musculus [NP_803421.1], R. norvegicus [NP_569117.1]. Peroxisomal thiolase/3-ketoacyl-CoA thiolase: S. cerevisiae [CAA37472.1], Y. lipolytica [Q05493.1], G. gallus [NP_001184217.1], M. musculus [NP_570934.1], E. caballus [XP_001488609.1], C. lupus familiaris [XP_534222.2]. B. taurus [NP_001029491.1], H. sapiens [NP_001598.1]. X. dendrorhous [ERG10A]: thiolase encoded by the ERG10A gene. X. dendrorhous [ERG10B]: thiolase encoded by the ERG10B gene. Accession numbers are given in parentheses. Numbers at each node indicate the percentage support for a specific node after 1,000-replica bootstrap analysis

Fig. 2

ERG10A and ERG10B relative transcript levels. a In strain 385-cyp61 ^(−/−), which does not produce ergosterol. Level of change was determined by comparison to wild type UCD 67–385 (control). Cultures were grown in YM liquid media for 24 h (blue) or 120 h (red). Each bar represents an average of three independent cultures. Black lines indicate standard deviation. *p < 0.05, **p < 0.01. b Effect of glucose addition. Gene expression kinetics and glucose concentration were quantified in strain UCD 67–385 after adding glucose (20 g/l final concentration). Error bars correspond to the standard deviation (n = 3). The negative values on the y-axis denote decrease relative to control

Fig. 3

PCR analyses of S. cerevisiae haploid strains. S. cerevisiae strains Sc-ERG10sc (carrying plasmid YEpNP-10sc) and Sc-cERG10xd (carrying plasmid YEpNP-c10xd) were analyzed by PCR to confirm the expected genotype. As controls, X. dendrorhous UCD 67–385 strain (Lane 1), S. cerevisiae strain S288c (Lane 2), S. cerevisiae diploid strain Sc-ERG10/erg10 (Lane 3) and a no-template control (Lane 6), were included. S. cerevisiae Sc-ERG10sc haploid strain (Lane 4) and S. cerevisiae Sc-cERG10xd haploid strain (Lane 5) were analyzed to assess: a absence of chromosomal ERG10 from S. cerevisiae (primers erg10scF and erg10scDWR); b presence of geneticin resistance cassette in S. cerevisiae (primers KanMXF2 and KanMXR2); c presence of ERG10 from S. cerevisiae (primers erg10scF and erg10scR); and d presence of ERG10 from X. dendrorhous (primers Thio2Fw and Thio2Rv). The molecular size markers Lambda DNA/HindIII (Lane M; 23.1, 9.4, 6.6, 4.4, 2.3, 2 and 0.6 kb) and GeneRuler 1 kb Plus (Lane 1kB, band size in kb is indicated) were used. On the right side of the picture, a schematic diagram of the amplification products is included; arrows represent primer sets with a letter indicating in which panel they were used. UP and DOWN (in blue) correspond to chromosomal regions located approximately 300 bp upstream and downstream of the S. cerevisiae ERG10 gene, respectively, KanMX4 corresponds to the geneticin (G418) resistance module, and pACT4 (in red) and tTDH3 (in green) correspond to promoter and terminator regions in the vector YEpNP, respectively

Fig. 4

Changes in transcript levels for POT1 mutant strains. Relative transcript levels of genes POT1, ERG10, crtS and crtR of X. dendrorhous were analyzed in samples obtained after 96 h of growth in liquid YM media for strains UCD 67–385, 385-pot1 ^(+/−) and 385-pot1 ^(−/−). The results were analyzed using the 2^-ΔΔCt method using actin as the normalizer gene and UCD 67–385 as the control strain compared to 385-pot1 ^(+/−) (blue) and 385-pot1 ^(−/−) (red). For POT1 transcript analysis in strain 385-pot1 ^(−/−), no transcript was detected. Each bar represents an average of three independent samples. Black lines indicate standard deviation. *p < 0.01