3. Results and Discussion
To prevent the degradation of benzoate by the engineered Pseudomonasstrain, the benABCD operon (PVLB_12215-12230) responsible for the conversion of benzoate to catechol (Figure 1) was deleted in a previously described phenylalanine-producing chassis [8] to yieldP. taiwanensis GRC3 Δ8ΔpykA -tap ΔbenABCD . Subsequently, the synthetic operon encoding the pathway from l-phenylalanine to benzoate was integrated at theattTn7 -site under the control of the constitutive promoterP14f [11]. The phenylalanine ammonia-lyase (PAL) deaminates l-phenylalanine to trans -cinnamate that is subsequently CoA-activated by the 4‑coumarate CoA-ligase (4CL) [10]. The resulting trans ‑cinnamoyl-CoA is converted into benzoate by the enzymes encoded by the phd cluster [12]. The Phd pathway from C. glutamicum was a key enabling factor for benzoate production because it accepts non-hydroxylated cinnamoyl-CoA as a substrate, unlike the ferulic acid pathway from P. putida,which only converts hydroxylated phenylpropanoids [5]. Shake flask cultivations were performed to characterize benzoate production from glucose and glycerol (Figure 2A,B).
The strain reached a final OD600 of ~3 while producing 1.9 ± 0.0 or 3.0 ± 0.0 mM benzoate from 20 mM glucose or 40 mM glycerol, respectively. Assuming complete carbon utilization, this corresponds to yields of 10.8 ± 0.1 on glucose and 17.3 ± 0.1 % (Cmol Cmol−1) on glycerol. In the course of the cultivation, no accumulation of trans -cinnamate was observed, confirming the efficient operation of the Phd pathway. To the best of our knowledge, this is the first approach demonstrating de novo benzoate biosynthesis applying a synthetic pathway. Moreover, high titers and yields were achieved for a microbial benzoate production process with a minimal medium, thereby producing benzoate solely from glucose or glycerol.