Total biosynthesis of opiates by stepwise fermentation using engineered Escherichia coli

Abstract

Opiates such as morphine and codeine are mainly obtained by extraction from opium poppies. Fermentative opiate production in microbes has also been investigated, and complete biosynthesis of opiates from a simple carbon source has recently been accomplished in yeast. Here we demonstrate that Escherichia coli serves as an efficient, robust and flexible platform for total opiate synthesis. Thebaine, the most important raw material in opioid preparations, is produced by stepwise culture of four engineered strains at yields of 2.1 mg l(-1) from glycerol, corresponding to a 300-fold increase from recently developed yeast systems. This improvement is presumably due to strong activity of enzymes related to thebaine synthesis from (R)-reticuline in E. coli. Furthermore, by adding two genes to the thebaine production system, we demonstrate the biosynthesis of hydrocodone, a clinically important opioid. Improvements in opiate production in this E. coli system represent a major step towards the development of alternative opiate production systems.

Figure 1. Total biosynthesis of thebaine using four-step culture.

The brackets indicate the reactions in the individual strain of each culture step. ATR2, NADPH-cytochrome P450 reductase 2; CNMT, coclaurine N-methyltransferase; DODC, dopa decarboxylase; MAO, monoamine oxidase; SalSNcut, N-terminally-truncated salutaridine synthase; SalR, salutaridine reductase; SalAT, salutaridinol acetyltransferase; TYR, tyrosinase; 3,4-DHPAA, 3,4-dihydroxyphenylacetaldehyde; 4′OMT, 3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase.

Figure 2. Total biosynthesis of (R,S)-reticuline.

(a) (R,S)-Reticuline synthetic pathway from (R,S)-THP. (b) The chirality analysis of pure (R,S)-reticuline. (c) The products from the culture of AN1752. (d) The products from the culture of AN1600. Experiments in b,c were conducted three times, and same tendency was observed. (e) (R,S)-Reticuline production following the addition of various concentrations of (R,S)-THP to the AN1600 culture. The error bar indicates the standard deviation of three independent experiments.

Figure 3. Thebaine production from pure (R,S)-reticuline or (R,S)-reticuline synthesized by total biosynthesis.

(a) LC-MS analysis of thebaine (m/z=312) from the culture of the parental strain (salutaridine producer, AN1096; upper panel), the thebaine producer (AN1829; mid panel) and the thebaine standard (lower panel). (b) MS/MS fragment pattern of the products of AN1829 (upper panel) and the thebaine standard (lower panel). Experiments in a and b were conducted at least three times, and same tendency was observed. (c) Time-course analysis of the total biosynthesis of thebaine from fermentatively produced (R,S)-reticuline in the four-step culture. (R,S)-Reticuline synthesized by total biosynthesis was added to the fourth step of the culture at time zero. The error bar indicates the standard deviation of three independent experiments.

Figure 4. Total biosynthesis of hydrocodone.

(a) Schematic representation of the total biosynthesis of hydrocodone. (b) LC-MS analysis of hydrocodone (m/z=300) from the culture of the parental strain (thebaine producer, AN1829; upper panel), hydrocodone producer (AN1942; mid panel) and the peak standard of hydrocodone (lower panel). The peak indicated by the arrow was analysed for its MS/MS fragment pattern. (c) MS/MS fragment pattern of the products of AN1942 (upper panel) and the peak standard of hydrocodone (lower panel). Experiments in b,c were conducted three times, and same tendency was observed.