On-column Trypsinisation

by Victoria Doronina Labtimes 02/2017

Studying cellular kinomes with affinity chromatography-based multiplex inhibitor assays is a rather cumbersome and extensive procedure. A new improved protocol from Norway takes some sweat out of the method.

Low abundance of their proteins of interest is a curse for people, who work with kinases. How do you study something that has ten molecules per cell and, if overexpressed, kills your cell culture? Something that even mass-spectrometry (MS) struggles to detect?

The probably obvious answer is to fish out the kinase molecules by affinity chromatography, selectively – sepharose beads packed in a column – and then send the result to mass spectrometry (MS). The kinase inhibitors crosslinked to the column serve as a bait and the assay bears the grand name of Multiplex Inhibitor Beads Assay (MIB).

But there are always drawbacks. The published protocols require as many as seven kinase inhibitors with many prepared in-house because of difficulties with finding a suitable linker. But even if you found and crosslinked your inhibitors, the main snag of the affinity chromatography method lies in the elution step.

The elution is usually done by adding ATP or hot SDS (Cell 149: 307-21). But ATP binds to myriads of ATPases Associated with various cellular Activities (AAA+) proteins and is, therefore, not specific enough. Hot SDS destroys the beads and requires a thorough removal prior to MS. A lot of work for a single-use experiment.

Marit Otterlei’s group simplified the affinity chromatography of kinases. Photo: NTNU discovery

Marit Otterlei and her co-workers from the Norwegian University of Science and Technology in Trondheim decided to skip a few steps of the classical MIB (Anal Biochem, 523:10-16).

They attached two commercial kinase inhibitors – PurvB and BisX – and one home-made to the sepharose 4B beads and applied the resulting column as a kinase bait. PurvB and home-made SB6-060-05 (patent number US200090238808) contain a carboxyl group and BisX, an amino group, which allow immobilisation of the inhibitor molecules on the sepharose beads.

The extract was prepared from three human cell lines treated with cytotoxic drugs. The authors incubated the sepharose-inhibitors column for ten minutes and then slowly let the extract run through. Subsequently, they added trypsin directly to the columns, making a nice shortcut. The beads and trypsin were mixed and incubated overnight as a standard for sepharose-based pull-downs. After eluting, the resulting peptides were dried and residual solvents removed using 0.5 ml spin columns.

LS-MS of the eluted peptides detected around 6,000 proteins, with many proteins found in at least two out of three assays. The number of proteins varied from 5,078 from U2OS cells to 4,351 in A549 cells. The authors explain this discrepancy by the difference in protein expression in various cell lines.

Amongst them were 40 percent (216) of annotated human kinases and other signalling proteins as well as metabolic and structural proteins. Comparison of extracts, prepared from untreated cells and cells treated with inhibitors, revealed significant changes in expression levels of more than 2,000 proteins.

Hot SDS destroys beads

Hot SDS elution was performed in parallel but the result was poor, as fewer kinases were detected. To make things even worse, beads were destroyed by hot SDS after three elutions, while in the modified assay the same columns were used for 1.5 years (>10 times), without significant deterioration of the results.

The amount of detected proteins estimated by Western Blot varied widely between the experiments. On the bright side, the qualitative data from Western Blot correlated well with data from the modified MIB assay. Both methods showed, for example, that MAPK3 was upregulated in H460 cells after treatment with cisplatin or docetaxel, while MAPK1 was downregulated in U2OS cell lines.

The authors explain the pull down of a large number of proteins, which usually do not bind to kinase inhibitors (that tie to kinases near the ATP binding site) through interaction of proteins: they bind non-specifically to kinase inhibitors, stick to the immobilised target proteins via protein-protein interactions or bind non-specifically to sepharose beads.

It’s hard to remove proteins involved in high-affinity interaction with target kinases or their partners. The structural proteins will probably remain in detectable quantities due to a sheer number of molecules.

But the detection of a high number of non-specific proteins suggests that the modified MIB method probably could be further improved. First of all, the number of non-specific binding can be reduced by the classical extract pre-incubation with sepharose beads. Also, the group applied one low salt and one high salt wash before trypsinisation, to remove impurities. More stringent washes with increasing salt concentration would reduce unspecific binding.

Otterlei and colleagues suggest that their technique does not depend on the particular inhibitors used and can be ­extended to any custom-made inhibitor with a group suitable for crosslinking to the beads.

In the end, the modified MIB ­protocol is a reasonable alternative to the standard MIB assay and should be worth a try.

Last Changed: 26.04.2017