A. oryzae’s FFA productivity was enhanced by improving the metabolic pathways involved in the FFA biosynthesis. It is considered that FFA biosynthesis is performed via the metabolic reactions illustrated in Figure 1 . When A. oryzae is cultured in a liquid medium containing sufficient glucose (10%), citrate accumulates in the mitochondria via the catabolism of glucose during glycolysis and the TCA cycle. Consequently, citrate accumulated in the mitochondria is released into the cytosol and then converted to palmitic acid, a saturated C16 FFA, by a four-step enzyme reaction. Subsequently, palmitic acid is added to the CoA residue by an acyl-CoA synthetase and concomitantly transferred to the endoplasmic reticulum, where palmitic acid, as the CoA derivative, is modified by elongation and desaturation of the carbon chain. After modification, different fatty acyl-CoA molecules are fused to glycerol, leading to the generation of triacylglycerol. Triacylglycerol is released from the endoplasmic reticulum into the cytosol in the form of lipid droplets. When lipase interacts with triacylglycerol present in cytosol under conditions such as carbon-source starvation, triacylglycerol is degraded to FFA, followed by further degradation to acetyl-CoA via beta-oxidation.

Firstly, FFA productivity was increased by FFA accumulation, which interrupted its degradation. For this purpose, we attempted to remove acyl-CoA synthetase that converts FFA to fatty acyl-CoA ( Figure 2A ). Six genes in A. oryzae RIB40 strain’s genome were highly homologous to the acyl-CoA synthetase genes identified in the budding yeast Saccharomyces cerevisiae by a BLASTP homology search. These six genes were individually knocked out in A. oryzae by genetic engineering, followed by an evaluation of FFA productivity, which is the amount of FFAs produced per gram of dried hyphae. Each strain was cultured in 50 mL Czapek–Dox minimal medium at 30°C and 200 rpm for 120 h, followed by extraction of FFAs in chloroform after disrupting cells and the subsequent application to the FFA enzyme assay. Among the six knockout mutants tested, only the mutant of the acyl-CoA synthetase gene faaA (AO090011000642), which has high homology to the acyl-CoA synthetase gene FAA1 of S. cerevisiae, demonstrated a 9.2-fold increase in FFA productivity compared to the parental strain (Tamano et al., 2015).

Secondly, FFA productivity was further increased by strengthening FFA biosynthesis via metabolic improvement. NADPH is required as a cofactor for FFA biosynthesis by supplying a reducing force. Thus, more NADPH was considered necessary for A. oryzae to produce additional FFAs. Therefore, we focused on the pentose phosphate pathway in this study, which diverges from and merges to glycolysis like the bypass and consists of eight enzymatic reactions. This pathway provides pentose molecules as substrates for nucleic acids and generates NADPH. The genes responsible for the pentose phosphate pathway in A. oryzae were selected by referring to a study on the metabolic modeling of A. oryzae (Vongsangnak et al., 2008) and homology search results against the genes of pathways identified in other organisms. Ten genes were selected in total. Individual overexpression mutants of the selected genes were constructed using the A. oryzae faaA knockout mutant. Evaluation of the FFA productivity revealed that overexpression of the transketolase gene tkt (AO090023000345) in the faaA mutant increased FFA productivity 1.4-fold ( Figure 2A ) (Tamano and Miura, 2016). Taken together, combining faaA knockout and tkt overexpression increased FFA productivity 13-fold compared to the parental wild-type strain. Furthermore, the FFA production yield was attained at 2.7 g/L for a 5-day culture by complementing auxotrophy in the constructed mutant and enhancing nitrogen source concentration in the culture medium. This yield corresponded to a 90-fold increase from the original condition when the wild-type strain was cultured in a regular culture medium. Therefore, FFA yield was substantially increased via metabolic and culture improvements.

Under regular culture conditions, A. oryzae accumulates FFAs within the cells, requiring cell disruption for extraction of FFAs with the aim of their industrial utilization. However, disrupting cells is labor-intensive, not eco-friendly, and requires mechanical cell treatment such as bead beating or chemical cell treatment using organic solvents such as acetone. Cell disruption may become unnecessary if FFAs are released from the cells. Therefore, FFA extracellular release, namely FFA secretory production, was attempted in A. oryzae.

To induce FFA extracellular release from A. oryzae, it was hypothesized that FFAs would be released if the cell surfaces were injured by chemicals or materials. Under these assumptions, different culture conditions were tested for the A. oryzae faaA knockout mutant. When Triton X-100, a non-ionic surfactant, was added to the liquid medium at a final concentration of 1%, the faaA mutant grew healthily and sufficiently, and FFAs were released to culture supernatant (Tamano et al., 2017). The amount of released FFAs corresponded to over 80% of the total FFAs produced. Moreover, extracellular release was limited to FFA and the alkyl ester. Another major lipid, triacylglycerol, which is mainly contained in lipid droplets, was not released. Furthermore, release was confirmed in the wild-type parental strain of A. oryzae. Taken together, FFA extracellular release was likely due to the increased permeability of cell membranes and protection of cells from bursting with endogenous rigid cell wall structures. Moreover, triacylglycerol was not released possibly because its large macromolecular lipid droplets cannot pass through the pores of the membranes created by Triton X-100.

Additionally, the enhancement of the productivities of released FFAs was studied under the culture condition with 1% Triton X-100 supplementation. Herein, genes that demonstrated positive correlations between expression levels and released FFA amounts in the time-course shifts, and the identified or predicted functions relative to lipid metabolism were selected. Subsequently, the selected genes were comprehensively overexpressed in the faaA knockout mutant. Overexpression of a lipase gene (AO090701000644) involved in generating FFAs and overexpression of a diacylglycerol O-acyltransferase gene dgat (AO090011000863) involved in generating triacylglycerol, which is a lipase substrate, had almost the same effect on enhancing FFA secretory productivity ( Figure 2B ) (Wong et al., 2021). The faaA mutant, which individually overexpressed these genes, demonstrated an approximately 3-fold increase in FFA secretory productivity in the presence of 1% Triton X-100.

Polyunsaturation of FFAs was also attempted. The wild-type RIB40 strain of A. oryzae originally biosynthesizes four FFAs (palmitic acid, stearic acid, oleic acid, and linoleic acid). However, FFAs expected to be used as feedstock for pharmaceutical agents, and supplements are more polyunsaturated. Such polyunsaturated FFAs include docosahexaenoic, eicosapentaenoic, and arachidonic acids. Thus, FFA polyunsaturation produced in the A. oryzae faaA knockout mutant was attempted. For this purpose, Δ6-desaturase and Δ6-elongase genes derived from M. alpina were introduced into the A. oryzae faaA mutant in a heterologous expression manner ( Figure 2C ). The introduced strain successfully biosynthesized free dihomo-γ-linolenic acid. That is, Δ6-desaturase and Δ6-elongase functioned to convert linoleic acid to dihomo-γ-linolenic acid in A. oryzae (Tamano et al., 2019). The constructed strain was named DGLA1.

Subsequently, overexpression of the three enzyme genes encoding endogenous elongase (AO090003000572), Δ9-desaturase (AO090011000488), and Δ12-desaturase (AO090001000224) were applied to the DGLA1 strain for facilitating the conversion of palmitic acid to linoleic acid. The multiple overexpression mutant of these three genes was named DGLA3 ( Figure 2C ). The DGLA3 strain produced 1.8-fold higher levels of free dihomo-γ-linolenic acid than the DGLA1 strain. The production yield reached 284 mg/L in the liquid culture (Tamano et al., 2020).

Furthermore, the tkt gene overexpression as described in section 2.1.1 was introduced to the DGLA3 strain, which increased free dihomo-γ-linolenic acid productivity by 1.2-fold according to the increase in the total FFA productivity. Subsequently, knocking out the α-1,3-glucan synthase gene agsB (AO090003001500) involved in pellet formation of A. oryzae (Zhang et al., 2017b) further enhanced free dihomo-γ-linolenic acid productivity by 1.1-fold. This finally constructed strain called DGLA3_tktOE_ΔagsB produced 533 mg of free dihomo-γ-linolenic acid per liter of liquid culture (Tamano et al., 2023). Overexpression of tkt and knockout of agsB led to dispersed hyphae in the liquid culture. This improved efficiencies of glucose and oxygen uptakes compared to the original pellet form, thus enhancing productivity. Additionally, the production yield of free dihomo-γ-linolenic acid increased more than the productivity. This was attributed to the increase in biomass owing to the shift from pellet to dispersed hyphae. In other words, hyphal density increased by dispersion, contributing to an increase in production yield.