Table 1.
Strains list.
Fig 1.
Schematic of an intact pks island coding for colibactin.
This cluster consists of 19 genes starting with clbA and ending with clbS. The region of the pks island deleted is shown by dashed lines.
Fig 2.
Δpks EcN does not produce colibactin.
A) The maturation of colibactin by ClbP results in two molecules of N-myristoyl-D-asparagine, which can subsequently be used as biomarker for colibactin production in EcN strains. B) N-myristoyl-D-asparagine is not detected in Δpks EcN, while it is detected in WT EcN strain. ND, none detected.
Fig 3.
In vitro evaluation of EcN and Δpks EcN fitness.
A) Growth curves of WT EcN (dark blue), Δpks EcN (pink) and both strains in mixed culture/competition (inoculated at a 1:1 ratio, light blue). Data are the mean and SD of triplicate cultures grown in LB media. B) Competitive index for Δpks EcN and WT EcN over a 24h period of growth in LB media. Competitive indexes were calculated at each time point by dividing the number of recovered Δpks EcN CFU (kanamycin resistant) by the number of recovered WT EcN CFU (streptomycin resistant). Each dot represents the CI determined from a single culture. A bar indicates the mean of the data. A competitive index of 1.0 indicates no difference in CFU recovery between the two strains.
Fig 4.
In vitro activity of engineered EcN strains in the absence of pks.
A) N-myristoyl-D-asparagine (colibactin biomarker) levels in supernatants of parental strains SYNB1618 (dark blue) and SYNB1934 (pink), while no biomarker is detected in Δpks SYNB1618 and Δpks SYNB1934. Data are the mean ± SD of 3 samples after growth in LB for 18h. ND = not detected, (LLOQ) was 0.8μg/mL. B) SYNB1618 and SYNB1934 Phe metabolizing activity in vitro, before and after deletion of the pks island. Bar graph shows the mean ± SD (n = 3) rate of TCA production in SYNB1618 (dark blue), Δpks SYNB1618 (light blue), SYNB1934 (pink), and Δpks SYNB1934 (orange) using an in vitro assay of TCA production over 1h. C) SYNB8802 oxalate metabolism with and without the pks island. Oxalate metabolism by SYNB8802 (intact pks island; dark blue) and Δpks SYNB8802 (deleted pks island; pink) were measured over 1h. Mean ± SD of 3 samples at each time point are shown.
Fig 5.
In vivo kinetics of WT EcN and Δpks EcN strains in mice.
(A) Fecal excretion of EcN strains in C57BL/6J mice (n = 5 per group) following oral administration of a single dose (1x1010 CFU) of WT EcN (pink) or Δpks EcN (dark blue) over 48h. (B) Fecal excretion of EcN strains in streptomycin-treated C57BL/6J mice (n = 10 per group) following oral administration of a combined dose (5x109 CFU of each bacterial strain) of WT EcN (pink) and Δpks EcN (dark blue) over 72h. In both cases, bacterial abundance was determined by CFU enumeration at each time point. Data are presented as the mean ± SEM. Statistical analysis was performed using 2-way repeated ANOVA followed by Sidak’s multiple comparison test, (p >0.05 for all time points).
Fig 6.
Kinetics and activity of WT EcN and Δpks EcN in nonhuman primates (NHPs).
(A) Fecal excretion of EcN strains in NHPs (n = 6 per group) following oral administration of a combined dose (1x1012 CFU of each bacterial strain) of WT EcN (pink) or Δpks EcN (dark blue) over 120h, (LLOQ = 1.5x102 CFU/g feces). Data are presented as mean ± SEM. Statistical analysis was performed using 2-way repeated ANOVA followed by Sidak’s multiple comparison test (p>0.05). (B) Appearance of plasma D5-TCA (B) and urinary recovery of D5-HA (C) in NHPs (n = 12 per group) following oral administration of a single dose (1x1012 CFU) of SYNB1934 (pink) or Δpks SYNB1934 (dark blue). Data are presented as mean ± SEM. Statistical analysis was performed using 2-way repeated ANOVA followed by Sidak’s multiple comparison test (B) and unpaired t-test with Welch’s correction (C).