Microbial Cell Factories (Sep 2022)

Glucose transport engineering allows mimicking fed-batch performance in batch mode and selection of superior producer strains

  • Daniela Velazquez,
  • Juan-Carlos Sigala,
  • Luz María Martínez,
  • Paul Gaytán,
  • Guillermo Gosset,
  • Alvaro R. Lara

DOI
https://doi.org/10.1186/s12934-022-01906-1
Journal volume & issue
Vol. 21, no. 1
pp. 1 – 14

Abstract

Read online

Abstract Background Fed-batch mode is the standard culture technology for industrial bioprocesses. Nevertheless, most of the early-stage cell and process development is carried out in batch cultures, which can bias the initial selection of expression systems. Cell engineering can provide an alternative to fed-batch cultures for high-throughput screening and host selection. We have previously reported a library of Escherichia coli strains with single and multiple deletions of genes involved in glucose transport. Compared to their wild type (W3110), the mutant strains displayed lower glucose uptake, growth and aerobic acetate production rates. Therefore, when cultured in batch mode, such mutants may perform similar to W3110 cultured in fed-batch mode. To test that hypothesis, we evaluated the constitutive expression of the green fluorescence protein (GFP) in batch cultures in microbioreactors using a semi defined medium supplemented with 10 or 20 g/L glucose + 0.4 g yeast extract/g glucose. Results The mutant strains cultured in batch mode displayed a fast-growth phase (growth rate between 0.40 and 0.60 h−1) followed by a slow-growth phase (growth rate between 0.05 and 0.15 h−1), similar to typical fed-batch cultures. The phase of slow growth is most probably caused by depletion of key amino acids. Three mutants attained the highest GFP fluorescence. Particularly, a mutant named WHIC (ΔptsHIcrr, ΔmglABC), reached a GFP fluorescence up to 14-fold greater than that of W3110. Strain WHIC was cultured in 2 L bioreactors in batch mode with 100 g/L glucose + 50 g/L yeast extract. These cultures were compared with exponentially fed-batch cultures of W3110 maintaining the same slow-growth of WHIC (0.05 h−1) and using the same total amount of glucose and yeast extract than in WHIC cultures. The WHIC strain produced approx. 450 mg/L GFP, while W3110 only 220 mg/L. Conclusion The combination of cell engineering and high throughput screening allowed the selection of a particular mutant that mimics fed-batch behavior in batch cultures. Moreover, the amount of GFP produced by the strain WHIC was substantially higher than that of W3110 under both, batch and fed-batch schemes. Therefore, our results represent a valuable technology for accelerated bioprocess development.

Keywords