Browse Prior Art Database

Highly Selective Enzymatic Amoxicillin Synthesis Disclosure Number: IPCOM000229412D
Publication Date: 2013-Jul-28
Document File: 4 page(s) / 154K

Publishing Venue

The Prior Art Database

Related People

Peter W. Sutton: AUTHOR [+14]

Related Documents

Eur. Pat. Appl. 1991, EP 0453048-A1: PATAPP [+10]


The Achromobacter sp. CCM 4824 penicillin G acylase βF24A mutant, immobilised on commercially available epoxy resin supports, is a suitably robust biocatalyst for the preparation of multiple batches of amoxicillin with only a 1.05 equiv excess of D-4-hydroxyphenylglycine methyl ester per batch.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 26% of the total text.

Page 01 of 4

Highly Selective Enzymatic Amoxicillin Synthesis

   Murray Brown, Ted Chapman, Laiq Chaudry, Isabelle Davis, Colin Edge, Andrew Fosberry, Michelle Gee, Simon Hayes, Emma Jones, Lucy Joyce, Bill Leavens, Mark Shipton, Dominic Smith, Peter W. Sutton,* Charles Wade

GlaxoSmithKline, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY


Penicillin-G-acylases (PGAs) are well established for the production of 6-amino-penicillanic acid (6- APA) through hydrolysis of the penicillin G side-chain. They are now enjoying increased application in the production of semi-synthetic beta-lactam antibiotics, such as amoxicillin, through the more challenging reverse hydrolysis reaction as they can deliver improved product quality and reduced environmental impact over the well established chemical process (Scheme 1).

Scheme 1: Penicillin G acylase catalysed preparation of amoxicillin (where X = OMe or NH2) Unfortunately, wild-type PGAs suffer from moderate chemoselectivity towards synthesis of the desired product versus hydrolysis of the ester or primary amide starting material or desired product (so called "synthesis/hydrolysis" or "S/H" ratio). Therefore, a large excess of the electrophilic component is generally required which can significantly increase cost and waste and has led to the relatively slow adoption of this process for the production of these high volume, low cost products.

The low product cost makes the single use of the free enzyme for antibiotic production uneconomical. Instead, more expensive immobilised PGA biocatalysts are used, where their increased cost is more than off-set by their recyclability (typically the biocatalyst is reused hundreds of times). Many different immobilisation techniques have been reported.1 Unfortunately, commonly used enzyme supports are said to lead to further reductions in S/H ratio over the free enzyme. To this end, a large body of research has been devoted towards the discovery of more efficient biocatalysts and processes for enzymatic antibiotic production.

Environmental screening for better enzymes has led to the identification of a handful acylases and amino ester hydrolases with moderately improved chemoselectivity over the archetypal PGA from E. coli (EC PGA). In contrast, enzyme engineering has proved more fruitful,2 with positions on EC PGA that are reported to influence chemoselectivity residing primarily in the phenyl binding pocket (positions α146, α 145, β24).3 Some changes at the β24 position appear to have a particularly strong positive impact on S/H ratio.

First identified as an important position for selectivity and activity by Gist-Brocades,4 Alkema et al later demonstrated that the βF24A mutant of EC PGA is capable of synthesising ampicillin from 6-

*Corresponding author: GlaxoSmithKline, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY; Tel.: +44 1438768674; e-mail:

Page 02 of 4

APA and phenylglycine methyl ester with a 3-fold increase in S/H...