A new harvest of fluorescent proteins

A suite of improved fluorescent proteins offers better tools for cell biology.

In this issue, Shaner et al.1 describe a veritable cornucopia of new fluorescent proteins, including mCherry, mBanana, mOrange, mStrawberry, mTangerine and mHoneydew. Fluorescent proteins make possible the relatively straightforward study of favorite proteins within living cells, and cell biologists have become increasingly dependent on them as research tools. Why should this latest crop interest the average cell biologist? The limitations of existing fluorescent proteins have direct consequences for experimental design and interpretation. Shaner et al. address some of these drawbacks and present new variants that should extend our capabilities.

The most prominent fluorescent proteins have been derived from the Aequorea victoria green fluorescent protein (avGFP)2,3, and these molecules have been thoroughly altered and scrutinized since the expression of avGFP in heterologous organisms about 10 years ago. The limitations of most fluorescent molecules are generally associated with molecular brightness and/or stability. However, avGFPs and its initial descendents have additional complications involving protein folding, chromophore maturation and self-association. Most limitations associated with these properties have been alleviated and the proteins vastly improved (Fig. 1) with a variety of mutations to produce reliable fluorescent markers (see refs. 4 and 5 for reviews).

Figure 1: An overview of the development of selected monomeric fluorescent proteins.

Bob Crimi

The numbers in parentheses indicate key steps (listed below) in the development of these proteins. Each protein is positioned at the year of publication on the y-axis and at the approximate emission wavelength on the x-axis. (1) Cloning of the GFP gene from Aequorea victoria2. (2) Expression of Aequorea victoria GFP (avGFP) in heterologous organisms3. (3) Mutagenesis of avGFP produces blue, cyan and green fluorescent proteins4,5. (4) Crystal structure shows that GFP is an 11-strand β-barrel with the chromophore protected from the external environment. Rational design based on the structure produces the yellow fluorescent proteins4,5. (5) YFP mutants with decreased pH and halide sensitivity are produced4,5. (6) One of three mutations, A206K, L221K or F223R, is found sufficient to disrupt avGFP dimerization4. (7) A green fluorescent protein, “Azami-Green” (AG), is isolated from Galaxeidae coral and converted into a monomeric protein9. (8) Directed mutagenesis of CFP produces Cerulean, which is 2.5× brighter and has a single fluorescence lifetime10. (9) DsRed (drFP583) is cloned from Discosoma and expressed in heterologous systems6. (10) DsRed is codon-optimized for expression in mammalian tissues; the molecule is prone to mature slowly into the red fluorescent form, to form obligate tetramers and to aggregate in some cases. (11) DsRed mutants are produced to decrease aggregation and decrease the time required for maturation into the red form. (12) DsRed1.T1, which rapidly matures (<1 hour) into the red form, yet still forms obligate tetramer4,5. (13) The DsRed tetramer is disrupted with 33 mutations to produce a monomeric RFP17. (14) An orange fluorescent protein, 'Kusabira-Orange' (KO), is isolated from the stony coral, Fungia concinna, and converted into a monomeric protein11.(15) The pitfalls associated with mRFP1, such as insufficient brightness and photostability, are addressed. New monomeric spectral mutants are produced from the latest round of mRFP1 mutagenesis.

The first of the red fluorescent proteins, Discosoma DsRed6, also has several pronounced undesirable properties, but mutant proteins that overcome many of these problems were quickly reported (see refs. 4 and 5 for reviews). One of these undesirable properties is that DsRed is an obligate tetramer. Although oligomerization may seem innocuous, protein-protein associations driven by the read-out marker can disrupt the normal localization, trafficking and protein-protein interactions of the protein of interest. This is particularly problematic if one considers single molecules participating in more than one oligomeric group at a time and essentially linking several oligomeric units together into a much larger complex. The production of monomeric RFP1 from DsRed by means of 33 mutations was a major breakthrough for red fluorescent proteins7. Yet, as good as it is, mRFP1 has several drawbacks relative to DsRed, such as decreased brightness and reduced photostability, which limit its usefulness as an imaging tool.

Shaner et al. avoided the effort required to monomerize the numerous other red fluorescent molecules available8 by using mRFP1 as the starting molecule. By doing so, they evolved monomeric fluorescent proteins having altered spectra or improved brightness with each round of mutagenesis (Fig. 1). Taking advantage of the similarity between the avGFP and mRFP1 structures (see ref. 4 for review), they directed their improvements with knowledge gained from many years of work performed with avGFPs. For instance, the fluorescence of mRFP1 is significantly decreased when fused to the C terminus of a protein of interest, whereas avGFP is generally unaffected when fused to either end of the protein of interest. Therefore, by simply replacing the N terminus of mRFP1 with that of avGFP and adding the avGFP C terminus to mRFP1, the authors produced a protein that develops fully the red fluorescence signal. Secondly, Shaner et al. targeted the residues Q66 and Y67 in mRFP1 (equivalent residues in avGFP are S65 and Y66, respectively), which are known to have key roles in determining the spectral characteristics of the avGFPs. The final result, after mutagenizing these positions along with additional random mutagenesis, is a series of monomeric fluorescent proteins with emission peaks ranging from 537 nm to 610 nm.

What is the significance of this diversity of fluorescent proteins? The new proteins essentially fill the 'gap' between the most red-shifted avGFP variant, YFP, and the red fluorescent proteins. One of the paper's supplementary figures shows FACS analysis of a mixture of bacteria expressing one of six different fluorescent proteins. By separating six fluorophores in this single experiment, the authors nicely demonstrate the importance of having diverse fluorescent proteins.

Although some of the new fluorescent proteins probably lack the brightness and/or stability needed for many imaging experiments, their existence is encouraging as it suggests that we will eventually have bright, stable, monomeric fluorescent proteins across the entire visible spectrum. Coupled with the technical advances in spectral imaging now offered by many of the commercially available microscopes, they will also allow single experiments on single cells to encompass a wider range of molecules of interest.

The diversity of the new fluorescent proteins brings up a dilemma. Because it is not often that a researcher needs to observe several different proteins at once, which of the fluorescent proteins are best for experiments requiring only one or two markers? As the authors indicate in their discussion, none of the new monomeric fluorescent proteins are clearly the best for all purposes. Putting aside the spectral differences, all of the reported variants, with the exception of mHoneydew, mBanana and mTangerine, are brighter than mRFP1. This alone indicates that any of the remaining fruits should be better than mRFP1. However, consideration of other attributes, such as pH sensitivity, maturation time or photostability, will help narrow the choice.

The fluorescence of mOrange is the brightest of the monomers but displays a moderate sensitivity to pH. Depending on where the protein is targeted in the cell (Is the organelle acidic?), this may be of little consequence. All of the new fluorescent proteins mature within a reasonable time period, so this property does little to narrow the decision. Finally, one should keep in mind the issue of photostability. Of the monomeric forms, mCherry is the most photostable (tenfold better than mRFP1 and six- to tenfold better than the other variants) and will be the best choice for experiments in which photobleaching has been the major limitation.

What can we expect in the future for these and other fluorescent proteins? The proteins described by Shaner et al. will ease restrictions on imaging experiments requiring a 'red' protein, either by providing one that's brighter, such as mOrange, or one that's more stable, such as mCherry. Thus, many of these molecules will be seamlessly incorporated into the bank of reliable fluorescent proteins and open new avenues for studying our favorite molecules. Considering the history of fluorescent protein development, each of these new proteins will undoubtedly serve as the starting point for further improvements and alterations. And furthermore, the patterns that emerge from the structural biology and fluorescence spectroscopy of the molecules produced in this and other studies may aid in future rational design of fluorescent proteins.


  1. 1

    Shaner et al. Nat. Biotech. 22, 1567–1572 (2004).

    CAS  Article  Google Scholar 

  2. 2

    Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G. & Cormier, M.J. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233 (1992).

    CAS  Article  Google Scholar 

  3. 3

    Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W. & Prasher, D.C. Green fluorescent protein as a marker for gene expression. Science 263, 802–805 (1994).

    CAS  Article  Google Scholar 

  4. 4

    Zhang, J., Campbell, R.E., Ting, A.Y. & Tsien, R.Y. Creating new fluorescent probes for cell biology. Nat. Rev. Mol. Cell Biol. 3, 906–918 (2002).

    CAS  Article  Google Scholar 

  5. 5

    Lippincott-Schwartz, J. & Patterson, G.H. Development and use of fluorescent protein markers in living cells. Science 300, 87–91 (2003).

    CAS  Article  Google Scholar 

  6. 6

    Matz, M.V. et al. Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17, 969–973 (1999).

    CAS  Article  Google Scholar 

  7. 7

    Campbell, R.E. et al. A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99, 7877–7882 (2002).

    CAS  Article  Google Scholar 

  8. 8

    Shagin, D. et al. GFP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 21, 841–850 (2004).

    CAS  Article  Google Scholar 

  9. 9

    Karasawa, S., Araki, T., Yamamoto-Hino, M. & Miyawaki, A. A green-emitting fluorescent protein from Galaxeidae coral and its monomeric version for use in fluorescent labeling. J. Biol. Chem. 278, 34167–34171 (2003).

    CAS  Article  Google Scholar 

  10. 10

    Rizzo, M.A., Springer, G.H., Granada, B. & Piston, D.W. An improved cyan fluorescent protein variant useful for FRET. Nat. Biotechnol. 22, 445–449 (2004).

    CAS  Article  Google Scholar 

  11. 11

    Karasawa, S., Araki, T., Nagai, T., Mizuno, H. & Miyawaki, A. Cyan-emitting and orange-emitting fluorescent proteins as a donor/acceptor pair for fluorescence energy transfer. Biochem. J. 381, 307–312 (2004).

    CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Patterson, G. A new harvest of fluorescent proteins. Nat Biotechnol 22, 1524–1525 (2004). https://doi.org/10.1038/nbt1204-1524

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing