Design

To evaluate the methanol-utilization potential of different strains, this study screened five marine-derived yeasts: TN3-1, P16, ZW03-4, DH177, and BZ. The strains were cultivated in YPM medium, and their growth was monitored by measuring OD600. Aureobasidium melanogenum P16 was selected as the best-performing strain based on its growth characteristics. The methanol utilization ability of P16 was further validated by comparing its growth in media containing methanol versus media without methanol. For comparative analysis, the non-methylotrophic yeast Saccharomyces cerevisiae and the natural methylotrophic yeast Komagataella pastoris X33 were included as controls to assess the methanol utilization level of P16. Ultimately, the melanin-producing A. melanogenum P16 was chosen as the chassis host.

Screening experiments demonstrated that P16 can utilize methanol as a carbon source and convert it into glucose (Glc). Although its methanol tolerance and metabolic efficiency are lower than those of the model strain P. pastoris, P16 exhibits unique adaptive advantages typical of marine microorganisms, including inherent stress-resistance mechanisms, diverse secondary metabolite biosynthetic pathways, and potential for gene readthrough regulation. These characteristics offer dual value for subsequent development: first, they provide a platform to explore a new biomanufacturing model based on methanol metabolism ("methanol → target products"); second, by combining the stress resistance of marine yeasts with their ability to utilize specialized marine substrates such as polysaccharides, P16 has the potential to serve as a next-generation cell factory that integrates high production efficiency with robust environmental adaptability. This provides a strong foundation for the biosynthesis of marine natural products and the development of stress-tolerant biomanufacturing systems.

To achieve efficient editing of target genes, this study designed a CRISPR-Cas9 system based on prior experience with fungal gene editing. First, the plasmid p414-TEF1p-Cas9-CYC1t (Addgene #43802) was obtained from Addgene. By analyzing restriction sites, SwaI and KpnI were selected as double-digestion sites to linearize the plasmid.

Next, using the laboratory-stored fl4a-nat-loxp plasmid as a template, primers were designed following the principles of seamless cloning to amplify the nourseothricin resistance gene (NAT). Primer R included a PacI restriction site, and the amplified fragment was inserted into the linearized vector to generate the Cas9-NAT plasmid. On this basis, a single PacI digestion of Cas9-NAT was performed to introduce the AMA1 autonomous replication sequence from the Aspergillus genome. Specifically, primers containing a NotI site were designed to amplify the AMA1 fragment, which was then seamlessly cloned into the Cas9-NAT vector to form the Cas9-NAT-AMA1 plasmid.

Finally, the Cas9-NAT-AMA1 plasmid was digested with NotI to enable insertion and assembly of a synthesized U6P+sgRNA-gRNA scaffold+U6T cassette, resulting in the final CRISPR-Cas9-NAT knockout vector. In practical applications, only the sgRNA region needs to be replaced to enable targeted editing of different genes.

A critical step in constructing and applying the CRISPR-Cas9-Am system is verifying its feasibility and editing efficiency in the host cells. To directly and accurately evaluate the functionality of the constructed CRISPR-Cas9-Am knockout vector, the Ade2 gene was chosen as the validation target. The Ade2 gene encodes phosphoribosyl pyrophosphate amidotransferase, a key enzyme in the de novo adenine biosynthesis pathway. Deletion of Ade2 disrupts adenine synthesis, leading to accumulation of metabolic intermediates that are oxidized into red pigments, resulting in colonies with a distinct red phenotype. This phenotypic change is clear, stable, and easy to observe, making it an ideal indicator for verifying the functionality of the CRISPR-Cas9 knockout system. Therefore, targeted deletion of Ade2 and observation of colony color provides a rapid and effective means to assess whether the constructed CRISPR-Cas9-Am vector is capable of precise gene editing.

Based on genome annotation and KEGG metabolic pathway comparison, key methanol metabolism-related genes were identified in the P16 genome, including AOX (alcohol oxidase), DAS (dihydroxyacetone synthase), and DAK (dihydroxyacetone kinase). Among these, AOX contains a typical peroxisomal targeting signal (PTS), whereas DAS and DAK lack conventional PTS motifs, suggesting that their localization mechanisms may differ from traditional patterns.

The spatial distribution of methanol metabolic enzymes can significantly influence metabolic efficiency. To determine the subcellular localization of key methanol-metabolizing enzymes, fluorescence labeling and co-localization experiments were designed.

Many peroxisomal enzymes in eukaryotic cells rely on short C-terminal peptide signals, such as the SKL tripeptide, for targeting. mCherry-SKL is a classical peroxisomal marker, whose fluorescent signal is stably localized to peroxisomes, making it suitable as a “localization reference.” When AOX is fused with GFP, significant overlap of GFP fluorescence with the mCherry-SKL signal indicates successful targeting of AOX to peroxisomes; lack of overlap suggests localization to other cellular compartments.

Using a double-digestion cloning strategy, AOX was fused with a GFP tag and co-expressed with the peroxisomal marker mCherry-SKL in P16 via electroporation. This approach allowed determination of the subcellular distribution of AOX and provided experimental evidence for the spatial regulation of methanol metabolism.

Freitag et al. reported that in fungi, some peroxisomal enzymes lack overt PTS signals; instead, the PTS sequence is encoded downstream of the first “conventional” stop codon. This indicates that the peroxisomal targeting of such proteins occurs only when ribosomal readthrough happens during translation, allowing translation to reach the hidden downstream PTS. Therefore, to verify whether a gene relies on the readthrough mechanism for peroxisomal targeting, it is necessary to construct different versions of the expression sequence that either retain or remove potential “cryptic PTS” elements.

Since the genetic background of melanin-producing Aureobasidium melanogenum is not well understood, the targeting mechanisms of its key enzymes remain unclear, posing challenges for chassis engineering. To address this, we designed experiments to study the localization mechanisms of methanol-metabolizing enzymes, providing a foundation for subsequent metabolic engineering.

Sequence analysis revealed potential PTS motifs downstream of the stop codons of DAK and DAS, suggesting that their peroxisomal localization may depend on ribosomal stop codon readthrough. To investigate this, we designed simulated readthrough experiments by constructing different plasmid versions to evaluate the role of this mechanism in regulating the localization of DAS and DAK. Three specific sequence fragments were constructed: Das, Das_cyt, and Dak_pex. Das includes the sequence up to the second stop codon. If readthrough occurs, the translated protein is expected to be targeted to peroxisomes; if readthrough efficiency is low, localization may primarily remain in the cytosol. Das_cyt includes the sequence only up to the first stop codon, removing the downstream PTS region. Even if readthrough occurs, the protein cannot carry an effective targeting signal. As a negative control, GFP fluorescence observed in the cytosol rather than the peroxisome would indicate that peroxisomal targeting indeed depends on the downstream extension beyond the first stop codon. Dak_pex includes the sequence up to the second stop codon but introduces a site-directed mutation converting the first stop codon TAG to TCG (coding for serine) to simulate ribosomal readthrough. If the resulting protein is stably localized to the peroxisome, this confirms that the downstream cryptic PTS is functional and that in natural conditions, localization depends on readthrough. The same principle was applied to DAK.

Since it remains unclear whether peroxisomal targeting of DAS depends on the ribosomal readthrough mechanism, this study further employed a heterologous overexpression strategy. By introducing the Das gene from P. pastoris, we aimed to enhance methanol metabolic flux in P16 and improve its growth performance in methanol-containing media, thereby evaluating the feasibility of functional optimization through overexpression.

The Das gene from P. pastoris was selected for heterologous expression rather than overexpressing the endogenous Das of A. melanogenum P16 for several reasons.

First, P. pastoris is a model methylotrophic yeast with a methanol assimilation pathway that has undergone long-term evolutionary optimization, forming a highly efficient metabolic module. Its DAS1/DAS2 genes exhibit superior catalytic kinetics, stability, and adaptation to methanol environments. Introducing these genes into P16 is therefore expected to increase methanol assimilation flux.

Second, sequence analysis indicates that the P16 Das gene contains potential cryptic peroxisomal targeting signals (PTS) downstream of the stop codon, whose localization relies on low-efficiency mechanisms such as stop codon readthrough. Even if transcription is upregulated, it is unlikely to significantly increase the amount of functional enzyme. In contrast, P. pastoris Das possesses a clear and functional PTS, effectively avoiding localization bottlenecks.

Finally, P. pastoris gene elements and regulatory mechanisms have been extensively studied and optimized; its promoters, terminators, and codon usage patterns facilitate efficient and stable heterologous expression. Introducing this gene into P16 under a strong promoter with proper targeting design can significantly enhance functional protein accumulation and peroxisomal localization efficiency. Therefore, heterologous expression of P. pastoris Das offers a more feasible and effective approach to increase methanol metabolic flux compared with overexpressing the native P16 Das.

To achieve efficient conversion of P16 into a methanol-based cell factory, this study implemented a metabolic pathway engineering strategy following the principle of “weakening consumption pathways while enhancing synthesis pathways.”

First, the yihX gene from Escherichia coli, encoding glucose-1-phosphatase, was heterologously introduced to enhance the conversion of metabolic intermediates into glucose. In the metabolic pathway, glucose-1-phosphate (G1P) is normally directed toward glycogen or UDP-glucose synthesis, causing partial carbon flux loss to pathways unrelated to methanol assimilation. The yihX-catalyzed reaction provides a shortcut by hydrolyzing G1P directly to free glucose, thereby creating an efficient “overflow route” for intermediate metabolites. This design is expected to promote glucose synthesis (Dumon et al., 2006; Nielsen & Keasling, 2016). However, construction of the heterologous expression strain and subsequent glucose quantification showed no significant increase, indicating the need for further engineering.

Next, we designed an experiment to delete the pfk gene, which encodes phosphofructokinase (PFK), to block glucose consumption in P16. PFK is a key rate-limiting enzyme in glycolysis, catalyzing the irreversible conversion of fructose-6-phosphate (F6P) to fructose-1,6-bisphosphate (F1,6BP), and is considered the committed step of glycolysis. Deletion of pfk effectively prevents P16 from utilizing glucose. This strategy has been widely validated in eukaryotic metabolic engineering as an effective approach to improve carbon utilization efficiency (Nielsen & Keasling, 2016; Curran et al., 2013).

1. [1] Kurtzman, C. P. & Fell, J. W. (2006). Yeast systematics and biotechnology. Springer.
2. [2] Gunde-Cimerman, N., et al. (2000). "Aureobasidium pullulans and related species: biology and applications." FEMS Microbiology Reviews, 24(4), 447–456.
3. [3] van der Klei, I. J., et al. (2006). "The significance of peroxisomes in methanol metabolism in methylotrophic yeast." Biochimica et Biophysica Acta, 1763(12), 1453–1462.
4. [4] Freitag, J., Ast, J. & Bölker, M. Cryptic peroxisomal targeting via alternative splicing and stop codon read-through in fungi. Nature 485, 522–525 (2012).
5. [5] Krainer, F. W., Dietzsch, C., Hajek, T., Herwig, C., Spadiut, O., & Glieder, A. (2012). Recombinant protein expression in Pichia pastoris strains with an engineered methanol utilization pathway. Microbial Cell Factories, 11:22.
6. [6] Nielsen, J., & Keasling, J. D. (2016). Engineering Cellular Metabolism. Cell, 164(6), 1185–1197.
7. [7] Curran, K. A., Leavitt, J. M., Karim, A. S., & Alper, H. S. (2013). Metabolic engineering of muconic acid production in Saccharomyces cerevisiae. Metabolic Engineering, 15, 55–66.