2D-DIGE vs 2D Gel Electrophoresis: What DIGE Is and When to Use It
Two-dimensional difference gel electrophoresis (2D-DIGE) is standard 2D gel electrophoresis with one change: the protein samples are labeled with spectrally distinct fluorescent dyes before the run, so that up to three samples can be separated on the same gel and imaged separately. Because the samples share a gel, gel-to-gel variation is removed from the comparison, and a pooled internal standard run on every gel allows accurate quantitation across a whole experiment. This article explains how DIGE works, how it differs from conventional 2D gels, and when the extra cost is worth paying.
What is 2D-DIGE?
DIGE stands for difference gel electrophoresis. It was introduced by Ünlü, Morgan and Minden in 1997 as a way to detect differences between two protein samples on a single 2D gel, by tagging each sample with a different fluorescent dye, running both on one gel, imaging each dye separately and superimposing the images. The dyes were designed so that a protein has the same mobility whichever dye it carries, which means a spot that appears in one image and not the other is a real difference between the samples rather than an artifact of running two gels.
Today the method uses three cyanine dyes, Cy2, Cy3 and Cy5, and is almost always run with a pooled internal standard in the Cy2 channel. Everything else about the separation is identical to conventional 2D gel electrophoresis: isoelectric focusing in the first dimension, SDS-PAGE in the second. If you are new to the technique, start with our guide to 2D electrophoresis.
How 2D-DIGE works
1. Label
Each sample is labeled with one CyDye. In minimal labeling, the most common form, the dye’s NHS ester reacts with the ε-amino group of lysine, and the dye-to-protein ratio is set so that roughly one dye molecule is attached per protein. A typical reaction is 400 pmol of dye to 50 µg of protein, incubated on ice in the dark for 30 minutes, then quenched with 10 mM lysine for a further 10 minutes.
2. Mix.
The labeled samples, usually two experimental samples plus the Cy2-labeled internal standard, are combined into one tube.
3. Run.
The mixture is loaded onto a single IPG strip and separated by isoelectric focusing and then SDS-PAGE, exactly as for a conventional 2D gel.
4. Scan.
The gel is imaged with a fluorescence scanner at the excitation wavelength of each dye in turn, producing one image per sample from the same gel.
5. Compare.
Because the images come from the same gel, they are already in register. Software overlays them, quantifies each spot in each channel, and normalizes every measurement against the internal standard.
The Cy2 pooled internal standard
The internal standard is a pool made by mixing equal amounts of every sample in the experiment. It is labeled with Cy2 and included on every gel alongside the two experimental samples.
Because the same standard is present on every gel, every spot on every gel can be expressed as a ratio to the same reference. That ratio is unaffected by how much of the gel ran into the buffer, how evenly the strip rehydrated, or how the scanner was set that day, because those effects apply equally to the standard and the samples on that gel. The standard therefore links the gels together and turns an experiment of many gels into one dataset with a common scale.
It also gives every spot a value on every gel, because a spot present in any sample is present in the pool. This matters for the statistics: it removes one of the main sources of missing values in 2D gel data.
The Cy2 pool appears on every gel, so every spot on every gel is measured against the same reference.
Minimal vs saturation labeling
Minimal labeling
Minimal labeling uses NHS-ester dyes that bind lysine. The dye is limiting, so only a small fraction of the molecules of any protein carry a dye, most of them a single one. This keeps the labeled protein’s mobility close to that of the unlabeled majority, and because the bulk of the protein is unlabeled, the gel can be post-stained and spots picked for mass spectrometry in the usual way. Minimal labeling detects proteins at around 1 ng and is the standard choice for most DIGE experiments.
Saturation labeling
Saturation labeling uses maleimide dyes that bind cysteine, with dye in excess so that every available cysteine on every protein is labeled. It is considerably more sensitive: 0.1 ng of albumin has been detected with a Cy5 saturation dye against 1 ng with the minimal dye, and a spot map of more than 5,000 proteins can be produced from as little as 5 µg of total protein. The dynamic range is around three to four orders of magnitude, an order of magnitude better than SYPRO Ruby or minimal labeling. Saturation labeling is the choice when sample is scarce, for example laser-microdissected tissue or biopsies. Its costs are that proteins with no cysteine are invisible, that reaction stoichiometry has to be optimized carefully for each sample type, and that the labeled protein’s mobility shifts, so spots must be matched to a post-stained preparative gel for picking.
2D-DIGE vs standard 2D gel electrophoresis
| Property | Standard 2D gel electrophoresis | 2D-DIGE |
|---|---|---|
| Detection | Post-run stain: Coomassie, silver or a fluorescent stain | Pre-run fluorescent labeling with Cy2, Cy3 and Cy5 |
| Samples per gel | One | Up to three, typically two samples plus the internal standard |
| Gel-to-gel variation | Present in every comparison; corrected by software alignment and replicates | Removed within a gel; corrected across gels by the internal standard |
| Internal standard | None | Cy2-labeled pool of all samples on every gel |
| Sensitivity | Coomassie lowest; silver and fluorescent stains reach the low nanogram range | About 1 ng with minimal labeling; about 0.1 ng with saturation labeling |
| Gels needed | One per sample per replicate | Roughly half: for example, 8 samples in triplicate on 12 gels instead of 24 |
| Missing values | Common, where a spot is detected on some gels and not others | Reduced, because every spot is measured in the internal standard on every gel |
| Cost per gel | Lower: standard reagents and stains | Higher: CyDyes and a fluorescence scanner |
| Compatibility with spot picking | Direct | Direct for minimal labeling; via a post-stained preparative gel for saturation labeling |
| Best for | Routine comparisons, teaching, labs without a fluorescence scanner | Quantitative comparisons across many samples, low-abundance changes, scarce samples |
Sensitivity and gel count figures from references 2
and 3 below.
When DIGE is worth it
The case for DIGE is strongest when the question is quantitative and the experiment is large. If you are comparing eight conditions in triplicate, that is 24 conventional gels, each with its own variation, against 12 DIGE gels tied together by an internal standard. The saving in gels roughly offsets the cost of the dyes, and the gain in statistical power is real because the noise you are trying to see through has been removed at source rather than modeled afterward.
DIGE is also the right choice when the differences you expect are small. A twofold change is visible on almost any gel; a 20% change across a cohort is not, unless the measurement is precise enough to separate it from run-to-run noise. The internal standard is what makes that precision possible.
Conventional 2D gels remain the sensible choice for a first look at an unfamiliar sample, for method development, for teaching, for a lab without a fluorescence scanner, and for any experiment where the comparison is qualitative: is this spot present or absent.
Limitations of 2D-DIGE
DIGE inherits every limitation of 2D gel electrophoresis: hydrophobic, very large, very small and extreme-pI proteins are under-represented, and abundant proteins mask rare ones unless the sample is prefractionated. On top of that, the dyes and the scanner are a significant cost, minimal labeling depends on lysine content so proteins with few lysines label weakly, saturation labeling misses proteins with no cysteine, and the labeling reaction has to be run to a consistent protocol or the internal standard cannot do its job. Finally, DIGE changes how the images must be analyzed. Three channels per gel, normalization to a standard, and spot matching across a dozen or more gels are not tasks for general-purpose image software.
Analyzing DIGE gels
A DIGE experiment produces three images per gel and a dozen or more gels. The analysis software has to align all of those images into one coordinate space, detect a single spot pattern across the whole experiment, quantify every spot in every channel, express each measurement as a ratio to the Cy2 standard, and then run the statistics.
SameSpots was built for exactly this. It aligns every image first and detects one spot pattern across the whole experiment, so every spot is measured on every gel and in every channel and the statistics run on a complete dataset. It supports DIGE experiments with a pooled internal standard as standard, with no separate DIGE module or license, and it runs one-way, two-way and repeated measures ANOVA on the normalized ratios. Mike Dunn, Professor of Biomedical Proteomics at the UCD Conway Institute, moved to SameSpots because of its handling of DIGE experiments; his words are on the product page.
SameSpots 2D-DIGE analysis software
For host cell protein antibody coverage, where the assay is a 2D-DIBE (difference in-blot electrophoresis) comparison of a total-protein channel against an antibody-bound channel, TotalLab provides SpotMap 2D, which is built for that specific calculation and is 21 CFR Part 11 ready with AuditSafe.
Frequently asked questions
Q: What is a DIGE?
A: DIGE is difference gel electrophoresis: 2D gel electrophoresis in which the samples are labeled with different fluorescent dyes before the run so that up to three samples can be separated on one gel and imaged separately.
Q: What does DIGE stand for?
A: Difference gel electrophoresis. “2D-DIGE” is two-dimensional difference gel electrophoresis, the usual form.
Q: What is the difference between 2D-DIGE and 2D gel electrophoresis?
A: The separation is the same. In 2D-DIGE the samples are pre-labeled with Cy2, Cy3 and Cy5 dyes and run together on one gel, which removes gel-to-gel variation from the comparison. In standard 2D gel electrophoresis each sample runs on its own gel and is stained afterward.
Q: How many samples can you run on one 2D-DIGE gel?
A: Up to three, one per dye. In most experiments two are experimental samples and the third, labeled with Cy2, is the pooled internal standard.
Q: What is the Cy2 internal standard?
A: A pool of equal amounts of every sample in the experiment, labeled with Cy2 and run on every gel. Every spot on every gel is measured as a ratio to it, which puts all the gels in an experiment on the same scale.
Q: What is the difference between minimal and saturation labeling?
A: Minimal labeling attaches roughly one dye per protein molecule, at lysine, and leaves most protein unlabeled, so gels can be post-stained and picked. Saturation labeling attaches a dye to every cysteine, is about ten times more sensitive and works from micrograms of protein, but misses cysteine-free proteins and requires a separate preparative gel for picking.
Q: How sensitive is 2D-DIGE?
A: Around 1 ng of protein per spot with minimal labeling and around 0.1 ng with saturation labeling, based on published detection of albumin.
Q: Is 2D-DIGE compatible with mass spectrometry?
A: Yes. With minimal labeling the majority of each protein is unlabeled, so spots can be picked from the DIGE gel directly after post-staining. With saturation labeling, spots are usually picked from a separately run, post-stained preparative gel matched to the DIGE image.
Q: Which software analyzes 2D-DIGE gels?
A: Dedicated software that handles multiple channels per gel and normalizes to the internal standard. SameSpots from TotalLab includes DIGE analysis in its standard license. Melanie sells DIGE as a separate module. DeCyder, the original DIGE analysis package from GE Healthcare, is no longer developed.
References
1. Ünlü M, Morgan ME, Minden JS. Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis. 1997;18(11):2071-7. https://doi.org/10.1002/elps.1150181133
(Source for: the origin of DIGE, the single-gel two-dye design, and dyes matched so that a protein has the same mobility whichever dye it carries.)
2. Diez R, Herbstreith M, Osorio C, Alzate O. 2-D Fluorescence Difference Gel Electrophoresis (DIGE) in Neuroproteomics. In: Alzate O, editor. Neuroproteomics. Boca Raton: CRC Press; 2010. Chapter 4. https://www.ncbi.nlm.nih.gov/books/NBK56019/
(Source for: NHS-ester dyes binding lysine with about one dye per protein; 0.1 ng albumin detected with a Cy5 saturation dye against 1 ng with the minimal dye; dynamic range of three to four orders of magnitude for saturation dyes; a spot map of more than 5,000 proteins from 5 µg of protein; halving the number of gels.)
3. Cytiva (GE Healthcare). 2-D experimental design using Ettan DIGE system. Application note. https://cdn.cytivalifesciences.com/api/public/content/digi-13606-pdf
(Source for: 400 pmol dye to 50 µg protein; 30 minutes on ice in the dark; quenching with 10 mM lysine for 10 minutes; each gel containing the pooled standard and two samples; eight samples in triplicate on 12 gels.)
4. Görg A, Weiss W, Dunn MJ. Current two-dimensional electrophoresis technology for proteomics. Proteomics. 2004;4(12):3665-85. https://doi.org/10.1002/pmic.200401031
(Source for: the limitations that DIGE inherits from 2D gel electrophoresis.)
Analyze your DIGE experiment in SameSpots
Three channels per gel, a pooled standard on every gel, and a complete dataset with no missing values. SameSpots includes 2D-DIGE analysis in its standard license and runs the statistics in the same window. Request a trial and analyze your own DIGE images.