The Light That Builds Itself – The Green Fluorescent Protein and Making Life Visible

The Hook

In the summer of 1961, two researchers in Friday Harbor on the west coast of North America were fishing jellyfish out of the water. Osamu Shimomura and his colleague Johnson were interested in a single question: where does the glow of the jellyfish Aequorea victoria come from? According to the Nobel Committee's popular account, they squeezed the glowing rims of about 10,000 jellyfish and obtained an extract from them – soberly called "squeezate" in the lab.

What they found first was aequorin, a protein that glows blue in the presence of calcium. But the same 1962 paper contained a side finding: a second protein that fluoresced green under UV light. This by-product of basic curiosity became, three decades later, one of the most important tools in cell biology. In 2008, Shimomura, Martin Chalfie and Roger Y. Tsien jointly received the Nobel Prize in Chemistry for it – "for the discovery and development of the green fluorescent protein, GFP".

This article explains why this particular protein is so special. The answer is biochemically surprising: GFP builds its own dye. It needs no enzyme and no special cofactor to do so – but, as we will see, it very much needs oxygen.

The Core Concept: A Protein That Makes Its Own Dye

Most coloured proteins in nature are coloured "on loan". Hemoglobin is red because it carries a heme group that the body produces in a separate metabolic pathway. Chlorophyll proteins are green because they bind chlorophyll. The protein itself is merely the scaffold; the dye comes from outside.

GFP is different. Its chromophore – the part of the molecule that absorbs light and emits it again – forms from three amino acids of its own chain, namely at positions 65 to 67: serine, tyrosine and glycine. Roger Tsien described this process in three steps in his influential 1998 review (Annual Review of Biochemistry 67, 509–544):

1. Folding. First, the chain of 238 amino acids has to fold into its native shape. Only this shape brings the three amino acids into the correct position relative to each other.
2. Cyclization and dehydration. The amide group of glycine 67 attacks the carbonyl group of residue 65 nucleophilically; a ring forms, and water is released in the process.
3. Oxidation. Molecular oxygen dehydrogenates a bond at tyrosine 66. Only in this step does the chromophore acquire its visible absorbance and fluorescence. Tsien writes that hydrogen peroxide is presumably released in a 1:1 ratio to mature GFP.

Two things follow from this that should be kept apart. First: no jellyfish-specific enzymes are required – the gene alone suffices to obtain glowing protein in a foreign organism. Second: the process is not entirely "unconditional". Tsien stresses that atmospheric oxygen is required for fluorescence to develop. GFP formed in the absence of oxygen only begins to glow once air is readmitted.

Why does the jellyfish actually glow green rather than blue? Aequorin emits blue light around 470 nm, close to an excitation peak of GFP. In the jellyfish, this energy is transferred to GFP, which releases it as green light. Tsien cites the early suggestion by Morin and Hastings that this is a radiationless energy transfer.

The Architecture: A Lantern in a Can

How does a protein manage to let such a delicate chemical event take place inside it and protect the resulting dye? The answer came in 1996, when two groups solved the three-dimensional structure at almost the same time: Ormö, Cubitt, Kallio, Gross, Tsien and Remington in Science (273, 1392–1395) and Yang, Moss and Phillips in Nature Biotechnology (14, 1246–1251). Both structures reached a resolution of 1.9 Å.

The picture is elegant: eleven β-strands form the wall of a cylinder, through whose axis an α-helix runs. The chromophore forms from this central helix and thus sits inside the barrel. Yang, Moss and Phillips named this new fold the "β-can"; Tsien adopted the term in his review. The cylinder shields the chromophore from the surrounding water.

I am of the opinion that this architecture is the key to understanding GFP: the protein is at the same time raw material, reaction vessel and protective shell of its own dye. This also explains why a protein that misfolds simply stays dark.

Example 1: The Moment GFP Became a Tool

After Shimomura's discovery, little happened for a long time. Only in 1988 did Martin Chalfie hear about GFP at a seminar at Columbia University. Douglas Prasher of the Woods Hole Oceanographic Institution was already looking for the corresponding gene; he cloned it and later sent it to Chalfie. According to the Nobel Committee, Chalfie's graduate student Euskirchen got E. coli bacteria to glow green about a month later.

The publication followed in February 1994 in Science (263, 802–805) under the title "Green fluorescent protein as a marker for gene expression". The core statement of the abstract: a cDNA for the GFP of Aequorea victoria yields a fluorescent product in both prokaryotes (E. coli) and eukaryotes (the roundworm Caenorhabditis elegans), and no exogenous substrates or cofactors are required for this. In the worm, Chalfie placed the GFP gene behind a promoter that is active in six touch receptor neurons – these cells glowed in the living animal.

Chalfie et al. conclude in the abstract that GFP can be used to monitor gene expression and protein localization in living organisms. I am of the opinion that this is exactly where the real paradigm shift lay: away from snapshots of fixed, stained specimens, towards observing processes in the living animal.

Example 2: One Amino Acid, a Better Signal – S65T and EGFP

Wild-type GFP has properties that are inconvenient for everyday lab work. According to Tsien, it has a major excitation peak at 395 nm, about three times higher than a minor peak at 475 nm. The two peaks belong to two forms of the chromophore, a neutral and an anionic one. Excitation at 395 nm yields emission peaking at 508 nm.

Tsien's group replaced serine 65 with threonine (mutation S65T). This suppresses the 395 nm peak and enhances the anionic peak, which shifts to 489–490 nm. In addition, according to Tsien, oxidation proceeds about four times faster than in the wild type. The variant "EGFP" contains, among other changes, S65T and F64L; the F64L mutation was introduced to improve folding at 37 °C.

The background: wild-type GFP folds fairly efficiently at or below room temperature, but its folding efficiency declines steeply at higher temperatures. The temperature sensitivity concerns only the folding process – GFP that has matured correctly remains stable and fluorescent, according to Tsien, up to at least 65 °C. For work with mammalian cells at 37 °C, improved folding was therefore a practical advance.

Example 3: A Rainbow from a Jellyfish

If a single mutation can change the spectrum so strongly, the next question is obvious: can other colours be produced? Tsien's 1998 review lists several classes; the table shows a selection:

Values after Tsien 1998, Table 1; for the wild type, the excitation peaks are given from the text.

The blue variant BFP arises when tyrosine 66 is replaced by histidine. Tsien notes, however, that BFP suffers from a relatively low quantum yield and bleaches relatively easily. In the yellow variants (YFP), according to Tsien, an aromatic residue at position 203 stacks with the chromophore.

For red, the jellyfish was not enough. Here proteins from corals came into play: according to the Nobel Committee, Mikhail Matz and Sergei Lukyanov found further fluorescent coral proteins, including the red DsRED. This one, however, consists of four amino acid chains; Tsien's group redesigned it so that it fluoresces as a single chain.

The Nobel Committee itself cites an impressive application: researchers at Harvard University modified mice so that nerve cells in the brain produce varying amounts of yellow, cyan and red fluorescent proteins. Individual neurons thus appear in different hues – the method was named "Brainbow" (Livet et al., Nature 450, 2007).

Framework: What GFP Can – and Cannot – Do

When planning an experiment, a sober balance helps. The following comparison is based on Tsien 1998 and the structural data:

I am of the opinion that the right-hand column is often lost in popular accounts. GFP is not a "neutral" marker but an additional protein with its own folding kinetics, its own oxygen requirement and its own photophysical quirks. Good experiments take this into account with controls.

Example 4: What the Story Teaches About Research

Between Shimomura's first description in 1962 and the Nobel Prize in 2008, 46 years passed. Shimomura was not looking for a tool; he wanted to understand how a jellyfish glows. Chalfie, in turn, was not a bioluminescence specialist but a neurobiologist who wanted to make nerve cells visible in a worm. Tsien, finally, was a chemist who wanted to redesign molecules in a targeted way.

Prasher, whose cloning of the gene formed the bridge between discovery and application, was not among the laureates; the Nobel Committee, however, explicitly mentions his role in its popular account.

I am of the opinion that GFP is a textbook example of how curiosity-driven basic research and goal-oriented tool development work together – and of the fact that scientific breakthroughs often rest on more shoulders than a prize can reflect.

The Central Takeaway

1. GFP builds its chromophore itself – from Ser65, Tyr66 and Gly67 – but needs oxygen to do so. The precise statement is: no additional enzymes, substrates or cofactors, but O₂ for the oxidation step.
2. Structure is function. The eleven-stranded β-barrel is both reaction space and protective shell; no correct folding, no light.
3. Small mutations, big effects. Single amino acid substitutions such as S65T or Y66H noticeably change brightness, maturation and colour.
4. A marker is never invisible. Maturation time, temperature sensitivity, dimerization and photophysics belong in the interpretation of every fluorescence experiment.

As a practical nudge: the next time you see an impressive fluorescence image, ask about the controls – how was it ruled out that the fusion protein localizes or behaves differently because of the attached can?

Cross-References in the Vault

• The Machine Made of RNA: The Ribosome, the Ribozyme, and the Translation of the Genetic Code – how the GFP gene becomes an amino acid chain.
• Four That Act in Concert: Allostery, Hemoglobin, and the Logic of Cooperative Binding – hemoglobin as a counterexample: colour through a "borrowed" cofactor.
• The Molecular Kiss of Death: The Ubiquitin-Proteasome System and the Art of Controlled Destruction – a system whose dynamics can be studied with fluorescent fusions.
• The Machine That Splits Water: Photosystem II and the Origin of Oxygen – another encounter between protein and light.

Food for Thought

GFP was not searched for but found – as a side finding of a question nobody considered useful. Which "useless" curiosity in your own field would you fund with time and resources today, without knowing what it might one day be good for?

Fact Check

Automated AI check against primary sources; no human expert review.

About 98 claims were checked (approx. 48 hard facts, 2 quotes, approx. 20 source attributions, approx. 28 synthesis claims including tables and "The Central Takeaway"). 8 passages were corrected, weakened or removed. The most important corrections:

• The name "β-can" does not originate with Tsien but with Yang, Moss and Phillips (1996); Tsien adopted it.
• The colour table wrongly gave a quantum yield of 0.4 for ECFP; in Tsien's table this value belongs to a different Y66W variant.
• Details not supported by the retrieved sources (first names of some authors, laser/filter suitability of S65T, the US state) were removed; the statement about the paradigm shift is marked as a personal assessment.

Sources and Further Reading

1. Nobelprize.org (2008): Press release – The Nobel Prize in Chemistry 2008. https://www.nobelprize.org/prizes/chemistry/2008/press-release/
2. Nobelprize.org (2008): Popular information – The Nobel Prize in Chemistry 2008: How the jellyfish's green light revolutionised bioscience. https://www.nobelprize.org/prizes/chemistry/2008/popular-information/
3. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., Prasher, D. C. (1994): Green fluorescent protein as a marker for gene expression. Science 263 (5148), 802–805. https://pubmed.ncbi.nlm.nih.gov/8303295/
4. Tsien, R. Y. (1998): The green fluorescent protein. Annual Review of Biochemistry 67, 509–544. https://doi.org/10.1146/annurev.biochem.67.1.509 – PDF: https://web.math.princeton.edu/~sswang/GECI/GFP_Tsien_1998_annual_reviews.pdf
5. Ormö, M., Cubitt, A. B., Kallio, K., Gross, L. A., Tsien, R. Y., Remington, S. J. (1996): Crystal structure of the Aequorea victoria green fluorescent protein. Science 273 (5280), 1392–1395. https://doi.org/10.1126/science.273.5280.1392 – structure data: https://www.rcsb.org/structure/1EMA
6. Yang, F., Moss, L. G., Phillips, G. N. Jr. (1996): The molecular structure of green fluorescent protein. Nature Biotechnology 14 (10), 1246–1251. https://doi.org/10.1038/nbt1096-1246

← All articles