Bioproduction of target molecules in unicellular microbial chassis is constrained by metabolic burden, intermediate toxicity, and the integration of increasingly complex heterologous pathways. Distributing a biosynthetic pathway across several specialized strains within a synthetic consortium can overcome this constraint, but its exploitation depends on maintaining stable populations throughout cultivation. Among several control strategies, syntrophy is a passive mechanism based on complementary auxotrophies and the exchange of essential metabolites. This thesis aims to establish and characterize mono-species syntrophic consortia of the yeast Yarrowia lipolytica, and to demonstrate their applicability to the bioproduction of a pigment, betanin.

A Y. lipolytica collection of 19 auxotrophs was first generated from a rational selection of target genes. Growth-based screening of the 171 pairwise combinations identified 56 viable syntrophic consortia, the largest collection reported to date in a yeast.

Characterization of these consortia revealed distinct growth profiles, reflecting highly variable kinetics of syntrophy establishment. Subpopulation monitoring uncovered a key property of these consortia: the final consortium’s composition consistently converges toward a reproducible, pair-specific equilibrium regardless of the initial inoculation ratio, which affects only the kinetics of syntrophy establishment.

The most promising pairs were then applied to betanin production, a natural food colorant derived from tyrosine. In monoculture, expression of a three-gene heterologous cassette for betanin production (the betanin (B) cassette) is sufficient for pigment production, and reinforcing the precursor supply with a tyrosine (T) cassette further increased the titer. However, a ceiling was reached with as few as two copies of the betanin cassette, indicative of a metabolic burden. To address this, the pathway was partitioned between two strains at the tyrosine node, enabling production in co-culture. Several distributions of metabolic burden were tested by varying which strain carried the tyrosine (T) and betanin (B) cassettes. Syntrophic control via the ∆aro7-∆trp2 and ∆aro7-∆trp4 pairs stabilized these co-cultures, which otherwise exhibited population drifts in the absence of auxotrophies. The configuration in which both partners carried the betanin cassette performed best. The best performing pair (BB:∆aro7–BT:∆trp4) produced 104.4 mg/L of betanin in test tubes and 147.8 mg/L in microbioreactors, a 1.66-fold improvement over the corresponding monoculture. This work represents the first reported production of betanin in a syntrophic consortium.

Finally, factors limiting these consortia were investigated. Metabolite requirements proved highly heterogeneous across the auxotroph collection. Medium complexity enrichment reduced the syntrophic lag without compromising consortium stability. Strategies, including overproducing the exchanged metabolite between partners or overexpressing an amino acid transporter, showed improved growth and/or betanin production in the consortia.

Collectively, this work provides a broad repertoire of characterized syntrophic consortia in Y. lipolytica and demonstrates the value of metabolic partitioning when the metabolic burden becomes limiting in monoculture. This platform can be expanded to produce other tyrosine derivatives by leveraging the modularity of these synthetic consortia.