Phosphite synthetic auxotrophy as an effective biocontainment strategy for the industrial chassis Pseudomonas putida

Abstract

The inclusion of biosafety strategies into strain engineering pipelines is crucial for safe-by-design biobased processes. This in turn might enable a more rapid regulatory acceptance of bioengineered organisms in both industrial and environmental applications. For this reason, we equipped the industrially relevant microbial chassis Pseudomonas putida KT2440 with an effective biocontainment strategy based on a synthetic dependency on phosphite, which is generally not readily available in the environment. The produced PSAG-9 strain was first engineered to assimilate phosphite through the genome-integration of a phosphite dehydrogenase and a phosphite-specific transport complex. Subsequently, to deter the strain from growing on naturally assimilated phosphate, all native genes related to its transport were identified and deleted generating a strain unable to grow on media containing any phosphorous source other than phosphite. PSAG-9 exhibited fitness levels with phosphite similar to those of the wild type with phosphate, and low levels of escape frequency. Beyond biosafety, this strategy endowed P. putida with the capacity to be cultured under non-sterile conditions using phosphite as the sole phosphorous source with a reduced risk of contamination by other microbes, while displaying enhanced NADH regenerative capacity. These industrially beneficial features complement the metabolic advantages for which this species is known for, thereby strengthening it as a synthetic biology chassis with potential uses in industry, with suitability towards environmental release.

Keywords: Biocontainment; NADH; Non-sterile conditions; Phosphite; Pseudomonas putida; Synthetic auxotrophy.

Fig. 1

Graphical representation of the different stages of the generation of a phosphite auxotrophic P. putida strain. A In blue, P. putida KT2440 with its native Pi-transport-related genes. Pi, Pn and G3Pi stand for phosphite, phosphonate and glycerol-3-phosphate. B In green, P. putida PSAG with its corresponding native genes, plus the hypophosphite transporter HtxBCDE from P. stutzeri WM88 and the phosphite dehydrogenase PtxD from Ralstonia sp. 4506. Pt stands for phosphite. C In yellow, P. putida PSAG-9 equipped only with the phosphite assimilation genes htxBCDE and ptxD and deprived of its native Pi-transport related genes

Fig. 2

Growth of P. putida in different P sources before and after addition of the PSAG assimilation genes and the removal of its Pi native transporters. A Growth of P. putida wild type (blue) and PSAG (yellow) in MOPS with different P sources. B Growth of P. putida wild type (blue) and PSAG-2 (yellow) in MOPS with different P sources. C Growth of P. putida wild type (blue) and PSAG-4 (yellow) in MOPS with different P sources. Growth was monitored by measuring OD₆₀₀ in an ELx808 Absorbance Microplate Reader (BioTek Instruments, Inc., VT, U.S.). Error bars represent the standard deviation among biological duplicates and technical triplicates for each condition

Fig. 3

PSAG-9 strain only grows on Pt-supplemented media. A Growth of P. putida wild type (blue) and PSAG-9 (yellow) in MOPS with different P sources. B Growth of P. putida wild type (blue) and PSAG-9 (yellow) in LB with and without Pt. Growth was monitored by measuring OD₆₀₀ in 500 mL flasks. Error bars represent the standard deviation between biological triplicates in each condition

Fig. 4

Phenotypic characterization of PSAG-9. Growth in MOPS supplemented with 50 mM glucose (A), 50 mM succinate (B), and 50 mM citrate (C). D Growth in MOPS supplemented with 50 mM glucose to test different Pt sources. As P sources, Pi (triangle) was used to grow P. putida wild type, and Pt (square) was used to grow P. putida PSAG-9, employing for this purpose either phosphorous acid (yellow) or sodium phosphite (orange) as Pt source. E Comparison between the transformation efficiency in P. putida KT2440 and P. putida PSAG-9. Transformation efficiency was calculated using the expression vector pSEVAb23 pAND105 sfGFP. CFU count was performed in plates containing MOPS-glucose agar containing either 1 mM Pi (wild type) or 1 mM Pt (PSAG-9) 48 h after electrotransformation of the strains. F Growth of strains harboring the expression vector pSEVAb23 pAND105 sfGFP, in MOPS supplemented with 50 mM glucose and corresponding P sources. G Relative fluorescence of strains harboring the expression vector pSEVAb23 pAND105 sfGFP. Data represent normalized fluorescence levels, expressed as a ratio with the OD₆₀₀ over 24 h. Growth and fluorescence were monitored using a Synergy Mx Multi-Mode Microplate Reader or an Epoch 2 Microplate Spectrophotometer (BioTek Instruments, Inc., VT, U.S.). Error bars represent the standard deviation among biological duplicates and technical triplicates for each condition

Fig. 5

PSAG-9 could not use other various P sources except Pt. A spot assay was performed on seven different types of agar-containing media: LB, Columbia, TB, Soil extract, MOPS-0 (P-free MOPS-glucose), MOPS-glucose-Pi, and MOPS-glucose-Pt. Three biological replicates were plated in different dilutions for the wild type (1, 2 and 3) and PSAG-9 (4, 5 and 6) respectively. For the spots, dilutions from an overnight culture ranging from 10^–1 to 10^–6 were used. Pictures were taken after 21 days of incubation at 30 °C

Fig. 6

Pt as sole P source effectively inhibited the growth of other biological contaminants. PSAG-9 clones were cultured under non-sterile conditions in 250 mL flasks. Blank cultures of MOPS-Pi and MOPS-Pt were included as controls and each group was incubated in triplicates. Pictures were taken after 5 days of incubation at 30 °C and 200 rpm. A The blanks with only MOPS-Pi medium; B the blanks with only MOPS-Pt medium; C the growth of PSAG-9 at 5 days post-inoculation in MOPS-Pt; and, D comparison of the three groups

Fig. 7

NADH production rate in wild type (blue) and PtxD-containing (green) cell extracts. Wells were seeded with 0.3–0.45 mg/mL of total protein from clarified cell extracts. Values were normalised for total protein concentration in each extract. OD₃₄₀ was measured for 90 min in 30-s intervals and the depicted bars represent the increasing NADH concentration rate throughout that period. Error bars represent the standard deviation between biological triplicates and technical duplicates in each condition