Covalent attachment of proteins to solid supports and surfaces via Sortase-mediated ligation

Abstract

Background: There is growing interest in the attachment of proteins to solid supports for the development of supported catalysts, affinity matrices, and micro devices as well as for the development of planar and bead based protein arrays for multiplexed assays of protein concentration, interactions, and activity. A critical requirement for these applications is the generation of a stable linkage between the solid support and the immobilized, but still functional, protein.

Methodology: Solid supports including crosslinked polymer beads, beaded agarose, and planar glass surfaces, were modified to present an oligoglycine motif to solution. A range of proteins were ligated to the various surfaces using the Sortase A enzyme of S. aureus. Reactions were carried out in aqueous buffer conditions at room temperature for times between one and twelve hours.

Conclusions: The Sortase A transpeptidase of S. aureus provides a general, robust, and gentle approach to the selective covalent immobilization of proteins on three very different solid supports. The proteins remain functional and accessible to solution. Sortase mediated ligation is therefore a straightforward methodology for the preparation of solid supported enzymes and bead based assays, as well as the modification of planar surfaces for microanalytical devices and protein arrays.

Figure 1. Ligation of fluorescent proteins to polymer beads.

(a) GMA beads modified with one, two, or four glycine residues were incubated with EGFP-LPETGG-His₆ and Sortase. Samples were taken at specific time points and analyzed on a BD FACSAria. Controls contained beads with no glycine or diglycine beads without Sortase. Error bars showing the standard error in the mean fluorescence are omitted as they are generally smaller than the data symbols. Errors are given in Supplementary Data S2.

Figure 2. Fluorescence micrographs of labeled solid supports.

(a) Diglycine GMA beads and (b) oligoglycine modified Affigel resin were separately labeled with EGFP and DsRed and then mixed. Fluorescence images were recorded as separate gray scale images (see Supplementary Figure S2) with FITC and Cy3 filter sets and then combined and false coloured.

Figure 3. Tus protein ligated to GMA beads is accessible to its cognate DNA ligand (Ter).

The sequence specific DNA-binding protein Tus was ligated to diglycine GMA beads. The Tus-labeled beads were incubated with varying proportions of fluorescein labeled Ter DNA and Cy5 labeled non-specific DNA and the bead fluorescence analysed by FACS. The curve is a model fit for a single binding process with a K _D of 29±8 nM. Error bars showing the standard error in the mean fluorescence are omitted as they are generally smaller than the data symbols. Errors are given in Supplementary Data S2.

Figure 4. (a,b) Ligation of fluorescent proteins to Affi-Gel resin.

Left to right, negative control, BFP-, EGFP-, and DsRed-LPETGG-His₆ were ligated to Affi-Gel 102 Resin modified with oligoglycine. The negative control reaction contained EGFP-LPETGG-His₆ and no Sortase. After washing with buffer pictures were taken with a white light (a) and UV transilluminator (b, 312 nm). (c–f) Ligation of EGFP to a glass surface. Microscope coverslips were modified with triethoxy(aminopropyl) silane and oligoglycine before incubation with EGFP-LPETGG-His₆. The slides were washed with Sortase buffer containing 1% SDS and photographed using the FITC filter set on an Axiovert 200 microscope. (c) Glycine modified surface with EGFP-LPETGG-His₆ but no Sortase, (d) amino modified surface (i.e. without glycine) with EGFP-LPETGG-His₆ and Sortase, (e) Glycine modified surface with EGFP-LPETGG-His₆ and Sortase with same exposure settings as negative controls, (f) same as (e) with five-fold reduced exposure time.