How to label a protein for FIDA measurements?

To measure a molecule with Flow Induced Dispersion Analysis (FIDA), it needs to be fluorescent: either through its own intrinsic fluorescence or with a fluorescent label. Intrinsic fluorescence is measured with the 280 nm (UV) detector, and labels with the 480, 530 or 640 nm detectors, depending on the fluorophore. Which label is best depends on your molecule and your experiment. FIDA also detects free fluorophore in your labelled sample and flags it to you, with its percentage, so you don't always need to purify your sample again after labelling.

How to choose a label?

Some helpful questions to help choose your label are:

• Does your molecule of interest have tryptophans or tyrosines? Intrinsic fluorescence (a label might not be required).
• Are you performing your assay in lysates? Fusion proteins, covalent labels.
• Does your recombinantly expressed protein have a tag for covalent labelling or a His-tag? SNAP tags, Tris-NTA.
• Are you looking for additional binding confirmation in terms of changes in fluorescence intensity (BRIC, Lambda Dynamics)? FITC, Cy5-related dyes.

After determining the optimal label for your sample, choose the detector that matches the excitation and emission ranges of the fluorophore.

Which detector to use?

• 280 nm (UV): uses the intrinsic fluorescence of tryptophan or tyrosine in your molecule of interest. Generally higher sample concentrations compared to other detectors. Excitation 270 to 290 nm, emission 320 to 430 nm.
• 480 nm: the most commonly used detector, suitable for most biological fluorescent tags and dyes. Compatible fluorophores: FITC, Alexa Fluor 488, Atto 488, GFP, Rhodamine B. Excitation 460 to 500 nm, emission 512 to 568 nm.
• 530 nm: minimal autofluorescence. Compatible fluorophores: mCherry, Alexa Fluor 568, Atto 565. Excitation 500 to 580 nm, emission 587 to 667 nm.
• 640 nm: high sensitivity, lowest possible background. Compatible fluorophores: Cy5, ALC 647, Atto 633, Atto 647, Alexa Fluor 633, Alexa Fluor 647. Excitation 590 to 650 nm, emission 662 to 735 nm.

The 280 nm, 480 nm and 640 nm detectors contain two photomultiplier tubes, enabling Lambda Dynamics. Switching between detectors is a simple five-minute procedure. See How does the Fida instrument detect a signal?

Tutorial: Labelling strategies for Fida experiments

This tutorial provides an overview of various labelling strategies for Fida experiments, focusing on fluorescent detection-based assays. It covers selecting labels, detectors, and labelling approaches suitable for different experimental setups. Watch the video.

Labelling strategies

Fusion constructs with fluorescent proteins:

1. Suitable for assays in lysates.
2. Proteins are expressed fused to fluorescent tags (commonly green or yellow) at N or C terminals.
3. No purification needed; can be used directly in cell lysate.
4. Main limitation: size increase due to the 28 kDa fluorescent protein tags.
5. The tag can sometimes be cleaved, leading to free fluorophore in the sample.

SNAP tags:

1. A smaller alternative (19 kDa) to fluorescent protein tags.
2. Allows precise label positioning and access to a broad range of dyes.
3. Requires a chemical reaction for dye conjugation.

Covalent labelling kits using NHS coupling:

1. Fidabio offers three kits (ALC 480, Fl 480 and ALC 640) for labelling primary amines.
2. Suitable for all purified proteins containing lysine residues; the buffer should be free of primary amines.
3. Requires a chemical reaction for dye conjugation, and it is common to have free dye after labelling.
4. ALC 480: insensitive to pH and environmental changes.
5. Fl 480: sensitive to pH and environmental changes, suitable for small molecule binding studies.

Thiol labelling:

1. Direct labelling of cysteines, which offers control over label placement.
2. Requires at least one free cysteine and avoidance of reducing agents.
3. Requires a chemical reaction for dye conjugation, and it is common to have free dye after labelling.

Click chemistry:

1. Involves incorporating unnatural amino acids with an alkyne group.
2. Requires specialised expression systems and DNA sequences.
3. Offers complete control over label placement.

Labelled peptides or nucleotides:

1. Due to the challenges involved with labelling small peptides or nucleic acid molecules, it is recommended to conjugate them with a fluorophore during synthesis.
2. Suitable for assays with minimal consumption and high sensitivity.
3. Occasionally leads to unconjugated fluorophore.

Noncovalent labelling strategies:

1. Commonly use Oregon Green and Tris NTA nickel complex for histidine-tagged proteins.
2. Proteins remain native and are not chemically modified, but affinity constants must be considered, as the interaction between the probe and the His-tag is affinity based.
3. A percentage of the probe can remain unconjugated, resulting in free fluorophore in the sample.
4. High specificity options include labelled nanobodies or antibodies.

Biomolecule specific dyes:

1. Examples: Thioflavin T for amyloid fibrils, SYBR gold for nucleotides.
2. Fida assays benefit from no size bias in these dyes.
3. Generally require higher concentrations due to lower affinity.

Is labelling your molecule mandatory to perform a FIDA assay?

No. The 280 nm (UV) detector uses the intrinsic fluorescence of tryptophan or tyrosine in your molecule of interest, enabling label-free assays.

Can Fida instruments work with noncovalent labelling?

Yes, there are several non-covalent labels that will work. Examples that have been applied in the past are the HIS Lite™ OG488-Tris NTA-Ni complex against the His-tag, single domain antibodies against the ALFA-tag, and nanobodies specific for your protein of interest. For example, the poster Optimizing Solution-Phase Epitope Mapping with ALFA-Tagged Peptides and FIDA (2026) used ALFA-tagged peptides with an Alexa-labelled anti-ALFA VHH as the indicator.

Could over-labelling your molecule have a negative effect on FIDA measurement?

On some occasions it can. Poor or excessive labelling can alter the molecule's conformation, increasing its hydrodynamic radius. Thanks to the labelling quality readout, FIDA detects such changes in hydrodynamic radius and flags them to you. In addition, for all technologies that rely on labelling to detect interactions, over-labelling risks putting a label in or near the binding site, inhibiting the binding due to steric hindrance.

How to check sample labelling quality?

An unconjugated label is much smaller than a label conjugated to a biomolecule. In FIDA, one can identify multiple species and their relative abundance. This is the feature used to give labelling quality, as described below.

How to detect free dye?

FIDA detects free dye in a labelled sample in the same measurement, without any separation of the species, and the Fida software reports the percentage of free dye.

Why can there be free dye in a labelled sample?

After chemical labelling of a biomolecule, the unreacted free dye is removed using a desalting column. However, one-step desalting columns often leave some free dye in the sample.

How does FIDA detect free dye?

FIDA can be used to measure the percentage of free dye present in a sample. Although all detectable molecules in a sample travel at the same time and their signals are superimposed, the Fida software can distinguish the free dye from the labelled molecule, and indicates the percentage of each in the sample (Figure 1). To learn more about how free dye is detected, see the Labelling Quality (free fluorophore) course, chapter Data interpretation, in the Sample Characterization & QC section of the Fidabio Learning Hub.

For example, Figure 2 A shows a Protein X sample which has been fitted both to single and two species. Both fits give similar sizes; however, the two species fit detects 1% free dye present. Figure 2 B shows a mutant of Protein X fitted to both single and double species. In this case, the two species fit detects 10% free dye in this sample. Notice that this sample exhibits a sharper peak, relative to the wild type in A, indicating the presence of a smaller detectable molecule. R_h with the two species fit is in better agreement with the R_h of the wild type in A, which is more reasonable.
With practice, it is relatively easy to spot the presence of free dye if the percentage of free dye is over five, such as in this example. However, a lower percentage of free dye is harder to spot with the naked eye.

How much free dye is acceptable?

Although it is preferable to work with samples that have zero or less than 2 percent free dye, it is possible to work with samples with higher free dye. In circumstances where further desalting results in an unacceptable protein loss, such samples could still be used by applying the two species fit. In these cases, it is recommended to use the R_h values from the two species fit for affinity and conformational change determination. Up to five percent of free dye is acceptable, but in general we recommend keeping free dye below 30 percent.

How to remove free dye?

The best remedy for a sample with excess free dye is to perform another desalting step. There are different ways of minimising free fluorophore:

• Fidabio dye removal resin: comes with a cut-off of 3 kDa. Molecules smaller than the cut-off get trapped on the resin beads, and larger molecules are collected in the flow-through. Sometimes more than one round of purification is required.
• Affinity columns: commercially available resin with affinity for free fluorophore. All the free dye binds to the resin beads, and the labelled protein is eluted.
• Dialysis: small volumes of labelled sample can be dialysed using a dialysis cassette. This method is slow but efficient for dye removal and prevents sample loss.
• Size exclusion chromatography (SEC) columns: due to the big difference between the size of free dye and dye conjugated to the protein, separation can be performed using SEC columns.

Can free dye be an artefact?

Yes. Free dye can be an artefact of sample stickiness: a labelled protein sticking to an uncoated capillary can appear as 100% free fluorophore, and after coating the capillary the labelled protein is detected. See How do you stop protein sticking to surfaces? and How to coat a capillary to prevent sample sticking?

Related resources

Want to see how FIDA reports labelling quality? The labelling quality readout shows the free label measurement, and the blog How to label samples efficiently? Towards Informed Labelling & Purification Practices covers labelling and purification workflows. The publication Electromembrane Extraction of Unconjugated Fluorescein Isothiocyanate from Solutions of Labeled Proteins Prior to Flow Induced Dispersion Analysis (Analytical Chemistry, 2019) describes removing unconjugated dye from labelled protein samples in a 96-well format.

The Labelling Quality (free fluorophore) course in the Sample Characterization & QC section of the Fidabio Learning Hub covers free fluorophore in more depth, and the Choosing the right detector wavelength course in the Hardware section covers labels and detectors.