ISISBio:Protocols/Sortase mediated ligation/Solid support ligation
We (and others) have used Sortase to attach a range of proteins to a small selection of solid supports, mostly resins or polymer beads, but also glass surfaces. Steric hindrance appears to be a significant problem for these ligation reactions so consideration of the spacer between oligoglycine and the solid support is important.
In the case of ligation to solid supports it is generally not possible to raise the concentration of the ligation partner in large excess over that of the target protein. Nonetheless we have generally found that the standard conditions used for small molecule labelling work well. We have not been able to accurately quantify the yield of protein ligated to solid supports, however experiments aimed at determining the depletion of target protein from solution suggest the yields are low. This is not generally a problem as generally relatively low coverages are desired.
Design of solid support
We have taken two approaches to the preparation of solid supports. Glycidylmethacrylate beads with one, two, or four glycines at the end of an eight carbon spacer were prepared by conventional fmoc solid phase synthesis and the effect of this on ligation yield and rate were followed. There is a clear effect of the number of glycines which we believe to be primarily a steric effect. In general two glycines at the end of a spacer seems the optimal balance of synthetic work and yields.
Where repeated cycles of deprotection and coupling are not possible or not convenient we have used the random EDC coupling of diglycine onto surface amines. The conditions used are somewhat unorthodox but are likely to lead to relatively short, sparsely packed chains. This followed unsuccessful attempts at ligation to H-GnC-OH (n=1,2,4) treated gold surfaces. Attempts to couple ligate protein to glass surfaces modified with an aminosilane and fmoc-GG-OH, followed by deprotection, were also unsuccessful. By contrast the random coupling of diglycine onto glass modified with an aminosilane was generally successful.
Overall our experience is that where background binding is not a major concern random coupling of oligoglycine is a convenient route to a glycine modified surface. Where background binding needs to be reduced more sophisticated approaches will be required. The major limitations on the ligation reaction appear to be steric with the direct coupling of ligation parterns onto surfaces generally not providing effective coupling. In these cases a larger spacer is recommended, ideally one that will stand away from the surface of the solid support.
Preparation of spotted arrays
We have attempted to spot ligation solution onto glass surfaces modified as above to prepare arrays. Although in some experiments this appeared to work the amine surface leads to a large background binding and we were unable to consistently show ligation above non-specific binding background to our satisfaction. This would probably be overcome by more sophisticated preparation of the surface to reduce the number of free (non-glycine) amines on the surface. Amine surfaces generally give a poor background binding response. The random co-coupling of a hydroxyacid and amino acid onto the amine surface followed by oligoglcyine coupling and alkaline hydrolysis of esters is one approach that could be explored. An alternative would be an orthogonal protecting group strategy.
- Sortase A
- LPETGG-tagged protein target
- Solid support
- Sortase buffer
- Target protein: 5 – 200 µM
- Ligation partner: For GMA beads we have routinely coupled 1 mg of beads with a surface glycine loading of 2-5 µmol.g-1 in 20 – 50 µL reactions. For glass surfaces we have not characterised surface loading but have prepared them as described [##ref###].
- Sortase A: 50 nM
- Standard Sortase buffer
Reactions are incubated for around one hour or overnight with shaking (beaded resins) or overnight with rocking (glass surfaces) at room temperature. For the surfaces described in ## the glass slides were submerged in the ligation reaction and rocked overnight. In those reactions where we attempted to spot the ligation reaction onto a surface we spotted 1 – 10 µL onto the surface and incubating overnight in a H2O saturated atmosphere. This appeared to work but was confounded by background binding.
The small molecule ligation partner can be readily removed by gel filtration and residual Sortase can be removed if required by gel filtration (if the sizes are sufficiently different) or nickel affinity chromatography. Remember the desired product is in the flow through.
The yield of desired product generally follows the ratio of protein to ligation partner. Reducing the concentration of ligation partner will therefore reduce the yield. It also tends to slightly increase the amount of hydrolysis product observed. Increasing the Sortase concentration will speed up the reaction but will also increase the amount of hydrolysis product. Reactions are probably complete in 4-6 hours depending on target protein concentration but we have found overnight incubation convenient.
Relevant papers and books
- Goldbeter A and Koshland DE Jr. . pmid:6947258.
- JACOB F and MONOD J. . pmid:13718526.
- Mark Ptashne. A genetic switch. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, 2004. isbn:0879697164.
- Cameron Neylon 11:50, 28 May 2008 (EDT):Who has experience with this protocol?