Post-column derivatization

By bringing together post-column derivatization and two-color laser-induced fluorescence (LIF) detection, two Japanese chemists have developed a novel capillary electrophoresis (CE) method for identifying proteins that have undergone post-translational modifications (PTMs). Initially, the chemists used their method to distinguish glycosylated proteins from unglycosylated proteins, but they say that a similar approach could work with any type of PTM.

Post-column derivatization involves attaching some kind of detectable label, usually a fluorescent molecule, to an analyte, often a protein, allowing it to be detected with a sensitive technique such as LIF after separation by CE. Although it tends to be easier to attach the labels to the proteins offline before the separation step, the advantage of attaching them afterwards is that it ensures the label doesn’t interfere with the separation process.

In conjunction with a few colleagues, Takashi Kaneta from Okayama University came up with a technique for conducting post-column derivatization a few years ago. It employs naphthalene-2,3-dicarbaldehyde, which readily binds to proteins, as the fluorescent label, and a T-shaped connector to combine the labels with proteins exiting the separation capillary, which is inserted through the arms forming the top of the T. The labels are introduced into the trunk of the connector, and then the labels and proteins react together in a reaction capillary attached to the far end of the connector, where they are also detected by LIF.

Two-color LIF

After showing that this technique can detect the proteins in biological samples, Kaneta wondered whether it could be combined with two-color LIF, which had never been attempted before. As its name suggests, two-color LIF involves stimulating fluorescence at two different wavelengths of light, offering a way to discriminate proteins that can’t be clearly separated by CE.

The idea is to label one protein with a fluorescent molecule excited at a specific wavelength and label another protein with a fluorescent molecule excited at a different wavelength. Two lasers alternate between emitting light at each wavelength, allowing the two proteins to be distinguished based on when they fluoresce, even if they can’t be separated by CE.

Kaneta realized that combining his post-column derivatization technique with two-color LIF might produce an effective way to distinguish proteins that had undergone a PTM from those that hadn’t, as the two can often be difficult to separate by CE. But labelling two proteins with different fluorescent molecules after separation is obviously much more of a challenge than simply labelling all the proteins with the same molecule.

Just add sugar

Together with his colleague Ayumi Tabara, Kaneta has now found a way to meet this challenge, at least for glycosylated proteins, which are proteins modified with an attached sugar group. The idea is to ensure that the second label should only be able to bind with sugar groups. Thus, any glycosylated proteins in a sample will bind with both labels, with the first label binding to the body of the protein and the second label binding to the attached sugar group, producing a dual fluorescent signal. Unglycosylated proteins, on the other hand, will only bind with the first label, producing a single fluorescent signal.

Kaneta and Tabara used naphthalene-2,3-dicarbaldehyde, which fluoresces at 450nm, as the first fluorescent label. For the second label, they used a complex of the lectin concanavalin A (con A), which binds with certain sugar groups, and the fluorescent molecule tetramethylrhodamine, which fluoresces at 532nm. When they tried out this method on a mixture of the glycosylated protein thyroglobulin and the unglycosylated protein albumin, they found that thyroglobulin produced a dual signal and albumin produced a single signal. This allowed the two proteins to be distinguished, even though they migrated from the separation capillary at similar times.

Because con A can only bind with certain sugar groups, this method is rather restricted at the moment, but Kaneta and Tabara are hoping to expand it to other glycosylated proteins by using other lectins. They are also helping to develop versions that are able to detect other PTMs.

‘I think the proposed method would be applicable to the detection of phosphorylated proteins using an appropriate probe such as a fluorescently-labeled antibody,’ Kaneta told separationsNOW. ‘I would [also] like to try selective detection of DNA methylation by the proposed method, using a methyl-CpG binding protein and an intercalator.’