|
February 26th, 2010 12:00pm EST, 9:00am PST
|
A new strategy for the directed evolution of glycosyltransferases for the synthesis of therapeutic glycoconjugates
Kerry Routenberg Love On the origin of phylogenetically related enzymes: Lessons from a directed evolution story Patrice Soumillion |
- Cosponsored by: GLOBAL
- Organizers: Zhilei Chen, David W. Wood, and Xinle Wu
- Registration Fee: $190*
* Academic attendees can register for free: email admin@biotwebinar.com for more information.
|
Abstracts |
|
A new strategy for the directed evolution of glycosyltransferases for the synthesis of therapeutic glycoconjugates
Kerry Routenberg Love, and J. Christopher Love. Department of Chemical Engineering, Massachusetts Institute of Technology, E25-537, 77 Massachusetts Ave., Cambridge, MA 02139 |
Engineered glycosyltransferases (GTases) have enormous potential for the synthesis of biologically relevant glycoconjugates, either by improving on the catalytic efficiency of native glycosylation or by incorporating non-natural sugar residues. Few attempts, however, have been made to engineer GTases by directed evolution, largely due to the lack of methods for screening and selecting mutants on the basis of GTase activity. Evolving enzymes from a library of variants requires a simple strategy to link desirable enzyme characteristics to the encoding library member. Spatially separating library members into individual containers would allow the identification of variants with unique catalytic properties without the requirement of substrate uptake or surface attachment. To that end, we have developed a technique using microfabricated chambers to physically separate a library of yeast cells, each capable of secreting an enzyme of interest (see figure). The library of segregated yeast cells can be interrogated with substrates that will yield a fluorescence signal upon successful enzymatic turnover. Clones from fluorescent wells are retrieved by micromanipulation and evolved for further rounds of selection. Since the intensity of signal correlates directly with product formation, real-time fluorescence monitoring allows selection of library members on the basis of both kinetic efficiency and substrate specificity.
|
On the origin of phylogenetically related enzymes: Lessons from a directed evolution story
Patrice Soumillion, Life Science Institute, Université catholique de Louvain, Place Croix du Sud, 4-5, bte 3, Louvain-la-Neuve, 1348, Belgium |
D-Alanyl-D-Alanine-peptidases and beta-lactamases are phylogenetically linked but feature large differences of reactivity towards penicillins: while beta-lactamases hydrolyse them very fast, leading to their inactivation, the same penicillins covalently inhibit DD-peptidases. Hence, DD-peptidases are also called penicillin binding proteins (PBPs). Starting from a PBP (called PBP-A) closely related to beta-lactamases, a 90 fold increase in the penicillinase rate was obtained by a single mutation (L158E). However, the mutant is still 5 orders of magnitude slower then a typical beta-lactamase. The active site elements of PBP-A-L158E are almost perfectly superimposed with those of a class A beta-lactamase and structure guided mutations were unsuccessful in further improving the activity. In vivo selection from libraries of mutants also failed but some results indicated a toxicity problem that could mean that DD-peptidase and beta-lactamase activities are mutually exclusive. This was further supported by an in vitro neutral drift experiment in which selection of PBPs from phage-displayed mutants led to a protein with severely impaired activity towards a DD-peptidase substrate. Our results emphasize the fact that, if toxic, an evolution path implying a generalist intermediate is forbidden. As a consequence, evolution paths may often be more twisted than thought at first sight.
