E. coli and S. cerevisiae are two model organisms that also serve as industrial cell factories for the production of a wide variety of compounds ranging from pharmaceutically active substances to food ingredients and biofuels.
In spite of the fact that X. dendrorhous has not been studied as extensively as E. coli or S. cerevisiae, it shows great potential to become a platform organism for terpene production . In order to assess the value of the red yeast as a cell factory, we expressed the Cop6 gene in the three X. dendrorhous mutants and compared the production of α-cuprenene with E. coli and S. cerevisiae strains expressing the same gene.
The cDNA from the Cop6 gene was expressed by all X. dendrorhous mutants, whereas the genomic version of the gene, when transferred to the red yeast, did not result in α-cuprenene accumulation (data not shown) indicating that X. dendrorhous cannot correctly splice the gene from C. cinereus.
When grown in rich medium, all X. dendrorhous strains, including the wild type, showed very similar growth rates: they reached OD600 values of 20 and produced a maximum of 12 grams of cell dry weight per liter of culture. Similarly, when looking at the α-cuprenene production, the three strains XdCop6, ΔE-Cop6 and ΔYB-Cop6 did not show big differences among each other, with levels of the sesquiterpene ranging from 70 to 80mg of compound per liter of medium.
Comparing Figures3 and 7, it is clear that, differently from the experiments in the rich medium, the three X. dendrorhous strains shared an altered growth behavior in the minimal medium. While XdCop6 and ΔYB-Cop6 reached a maximum cell dry mass of nearly 5g/L, similar to the one obtained with the wild type strain, ΔE-Cop6 could not produce more than 3.5g of dry cells per liter of medium. In 2008 Niklitschek and colleagues have reported the difficulty to isolate a X. dendrorhous strain in which both crtE alleles had been knocked out , suggesting an important role of this protein in the yeast growth. In the light of the results shown in this study, we can conclude that difference in growth between the ΔE-Cop6 mutant and all the other X. dendrorhous strains in the minimal medium could be explained by a lack, in this particular medium, of compounds important for the yeast growth produced directly or indirectly by the CrtE protein.
Concomitantly with the reduced growth in the minimal medium, the concentration of α-cuprenene in the dodecane was also affected, reaching values ranging from 15 to 21mg/L of culture. The decrease in sesquiterpene accumulation can partly be explained by the reduced cell mass and partly by the lower concentration of nutrients in the minimal medium which would induce the cells to minimize the energy consumption by shutting down unnecessary pathways.
When comparing cell mass accumulation and α-cuprenene production in all X. dendrorhous, E. coli and S. cerevisiae strains, the prokaryote showed the lowest values. The low biomass in the bacterium is most likely to be ascribed to a lack of glucose in its growth medium, while the limited sesquiterpene production is due to the lower terpene flux in E. coli compared to the two eukaryotes.
The differences in growth curves and dry weight between the X. dendrorhous and the S. cerevisiae strains seem to have morphological reasons. X. dendrorhous cells are on average bigger than S. cerevisiae ones  and at the same optical density S. cerevisiae cell counts are almost 10 times higher than in X. dendrorhous cultures, meaning that the same OD600 value corresponds to more S. cerevisiae cells than it does for X. dendrorhous. Since, at the beginning of the time course experiments, the initial OD600 for all the strains was set at 0.05, the number of cells initially transferred to the fresh medium was higher in ScCop6 than in all the X. dendrorhous mutants. This would explain the delay in growth we observed for XdCop6, ΔE-Cop6 and ΔYB-Cop6 compared to ScCop6. Additionally, the higher cell mass accumulation observed in the X. dendrorhous strains compared to S. cerevisiae may be due to the red yeast’s bigger sized cells rather than to a higher number of cells.
The highest α-cuprenene production levels were obtained with the X. dendrorhous strains both in the rich and in the minimal medium experiments. Remarkably, in the YPD medium the gap in sesquiterpene accumulation between the red yeast and the S. cerevisiae strain was far more pronounced. We assume that, since the complete medium does not allow selective pressure on ScCop6, which was isolated by its ability to grow on minimal medium lacking uracil, the strain might have undergone a reduction in plasmid copy number.
While ScCop6 mutants contain an average of 20 copies of Cop6, the X. dendrorhous white mutants possess just one copy of the gene since the recombination of the constructs can occur only once in the single crtE or crtYB genes. In order to obtain a mutant with a higher number of integrations of the gene in the genomic rDNA, we transformed the X. dendrorhous wild type strain with a higher concentration of the DNA fragment from the pPR-Cop6 vector and selected the transformants on YPD medium containing a concentration of geneticin 5 times higher than normal, hoping for gene amplification. Unfortunately, no colony grew after this transformation and we could not evaluate the effect of more gene copies on the α-cuprenene accumulation.
Nevertheless, we can safely assume that the concentration of the precursors is not a limiting factor in the sesquiterpene production in X. dendrorhous, since the strain XdCop6 can easily sustain the production of both α-cuprenene and astaxanthin, especially when grown in the YPD rich medium. The production of both terpene compounds in XdCop6 confirms the hypothesis that a higher gene copy number would positively influence the α-cuprenene production in X. dendrorhous.
In conclusion, X. dendrorhous shows great promise since it has the GRAS status, it grows at room temperature in minimal media, and it has already been used by industry for the production of astaxanthin. We discovered that it can produce at least three non-native sesquiterpenes, pentalenene , α-cuprenene and cubebol (data not shown). Furthermore, X. dendrorhous is the best microorganism, among the ones we have analyzed, to be used for the production of α-cuprenene. A better understanding of the molecular biology of this yeast will prove useful for the identification of stronger promoters for a higher gene expression.
In light of the aforementioned advantages and of the provided results, X. dendrorhous is an interesting candidate for being used as a cell factory for the production of terpenes.