From Concept to Process:
Metabolic Engineering for Production of Glucosamine
Glucosamine (GlcN) and its derivative N-acetylglucosamine (GlcNAc or NAG) are synthesized in all organisms, including bacteria, yeast, filamentous fungi, plants and animals. In humans, GlcN and GlcNAc are precursors of the disaccharide units in glycosaminoglycans (such as hyaluronic acid, chondroitin sulfate and keratan sulfate), which are necessary to repair and maintain healthy cartilage and joint function. Clinical trails with GlcN for treatment of arthritis started in the early 1980's. Since then, GlcN has been increasingly used as a dietary supplement. It is estimated that 33 million people suffer from osteoarthritis in the United States. The National Institutes of Health (NIH) is conducting a Phase III clinical trial, a multiple-site, double-blind randomized study on the effectiveness of GlcN for treatment of knee osteoarthritis (NIH GlcN/Chondroitin Arthritis Intervention Trial (GAIT); http://clinicaltrials.gov/show/NCT00032890). Results of the study are expected to be published in March, 2005.
Currently, GlcN is produced by acid hydrolysis of chitin (a linear polymer of GlcNAc) extracted from crab and shrimp shells. Concentrated hydrochloric acid breaks down the polymer and deacetylates GlcNAc to form GlcN. GlcNAc is produced by chemical acetylation of GlcN using acetic anhydride. GlcN production could become limited by variable raw material supply as the demand continues to increase. Moreover, GlcN from shellfish may not be suitable for people with shellfish allergies. Some filamentous fungi contain chitosan (a linear co-polymer of GlcN and GlcNAc, with GlcN accounting for over 65%) in their cell walls. The amount of chitosan in fungal biomass is low, generally less than 25% of dry cell weight. Current economics limit GlcN production to the use of fungal biomass from fermentation processes dedicated to citric acid or other primary products.
Bio-Technical Resources developed a fermentation-based process for production of high quality and low cost GlcN and GlcNAc. Metabolic engineering was used to develop recombinant E. coli strains as production hosts. Pathways for the synthesis and metabolism of GlcN are well-characterized in the bacterium Escherichia coli and other organisms. As shown in Figure 1, GlcN synthase (encoded by glmS) converts fructose-6-phosphate (fructose-6-P) to GlcN-6-P using glutamine as an amino donor. GlmS homologues from eukaryotic organisms are normally called GFA1 (glutamine-fructose-6-P amidotansferase). For the synthesis of peptidoglycan and lipopolysaccharide, essential components of the cell wall of gram-negative bacteria, GlcN-6-P is converted to GlcN-1-P by phosphoglucosamine mutase (encoded by glmM). The product is further converted by a bifunctional enzyme, GlcN-1-P N-acetyltransferase/GlcNAc-1-P uridyltransferase (encoded by glmU), to GlcNAc-1-P and then to UDP-GlcNAc.
GlcN and GlcNAc also serve as alternative carbon and nitrogen sources. GlcN is transported and phosphorylated by a mannose transporter (IIMan, encoded by the operon manXYZ) and the glucose transporter (IIGlc, encoded by ptsG). GlcNAc is transported and phosphorylated by the mannose transporter and a GlcNAc-specific transporter (IINAG, encoded by nagE). GlcNAc-6-P deacetylase (encoded by nagA) converts GlcNAc-6-P to GlcN-6-P, which is further converted to fructose-6-P by GlcN-6-P deaminase (encoded by nagB). In the E. coli genome, glmS and glmU constitute the operon glmUS while nagA, -B, -C, -D and -E form a regulon. The nagC gene encodes a regulatory protein that functions as a repressor of the nag regulon as well as both an activator and repressor of the glmUS operon. The function of nagD is not known, but is likely related to amino sugar metabolism, as it resides in the nag regulon.
Due to complex regulation, the levels of glucosamine detected in the growth medium of E. coli cultures are very low, typically several mg per liter. Metabolic engineering was used to maximize the capacity of amino sugar production. The initial strategy was to inactivate genes involved in glucosamine transport and catabolism, and to over-express the biosynthesis gene (glucosamine synthase). This led to a 15-fold increase in glucosamine production, but the titer remained at the milligram level. Feedback inhibition of glucosamine synthase was identified as a critical factor in regulating the pathway. Directed evolution and screening of the enzyme and over-expression of a product-resistant variant proved to be very effective to increase glucosamine production, and the titer reached multi-gram levels. However, rapid glucosamine degradation and inhibitory effects of glucosamine and its degradation products on the host cell prevented further titer increase.
Extending the pathway to N-acetylglucosamine, an amino sugar that is stable and inert to the E. coli host, by over-expressing a yeast glucosamine N-acetyltransferase gene, represented a significant conceptual breakthrough enabling high titer production. Finally, development and optimization of a two-phase fed batch process (growth phase and production phase) minimized inhibitory effects of N-acetylglucosamine pathway intermediates on cell growth and resulted in the development of a competitive process for the production of N-acetylglucosamine and glucosamine.