Exploiting the cyanobacterial light-harvesting machinery for developing fluorescent probes

Researchers develop a new class of near-infrared fluorescent proteins from a light-harvesting phycobiliprotein.

Over millions of years under severe environmental pressure in murky waters, the cyanobacterial light-harvesting phycobiliprotein allophycocyanin (APC) evolved for excitation and emission maxima in the near-infrared (NIR) optical window (650–900 nm). Now experimental biologists, such as Rodriguez et al.1, who describe their work in an Article in this issue of Nature Methods, are harnessing the power of this natural selection to develop a new generation of bright and robust fluorescent proteins (FPs) for a range of applications, including deep imaging of mammalian tissue.

FPs with excitation/emission maxima within the optical window for efficient light penetration are required for deep imaging, but the majority of currently known FPs—jellyfish/coral-derived GFP and GFP-like proteins—are spectrally limited to excitation maxima of <610 nm. Furthermore, the autocatalytic formation of their chromophores requires O2, leading these FPs to mature poorly in hypoxic tissue regions and to produce a stoichiometric amount of H2O2 (ref. 2), possibly making them harmful to these vulnerable regions.

Many attempts have been made to develop proteins that become fluorescent by incorporating natural π-conjugated compounds (pigments) as chromophores. Among such pigments is the bilin molecule, consisting of an open chain of four pyrrole rings. Bilins are heme metabolites, as represented by biliverdin IXα (BV), the product of heme breakdown by heme oxygenase-1 (HO-1). BV is a π-conjugated compound that absorbs red light; it is colored blue-green and can be seen in skin bruises. In mammalian cells, BV is reduced to bilirubin, which is later excreted into bile. In plants and microorganisms, BV is reduced to phytochromobilin (PΦB) by the enzyme HY2 or to phycocyanobilin (PCB) by PcyA (Fig. 1). PΦB, PCB, and BV are linked by covalent bonds to biliproteins found across many non-animal kingdoms, forming phytochromes and phycobiliproteins.

Figure 1: Schematic depiction of bilin-based fluorescent proteins.

Various bilins (gray rectangles and chemical structures) are shown, as well as the enzymes that modify them. Metabolic pathways are shown by dashed arrows. Solid black arrows indicate covalent (unidirectional) or noncovalent (bidirectional) bilin incorporation. Also named are the phytochromes (left) and phycobiliproteins (right) that bind the respective bilins. FPs are indicated by shading in the color of their fluorescence, with engineered FPs in bold. Phy, plant phytochrome.

Natural phytochromes show very weak fluorescence. They photo-transform reversibly between Pr (inactive) and Pfr (active) states that absorb red and far-red (730 nm) light, respectively. Plant phytochromes use PΦB as their chromophore and regulate numerous photomorphogenic processes, including seed germination, chloroplast and leaf development, shade avoidance, circadian rhythm entrainment, and flowering time3. In contrast, bacterial phytochromes (BphPs) incorporate BV on a positionally conserved Cys. In 2007, the Tsien lab reported mammalian expression of NIR FPs (excitation/emission maxima at 684/708 nm) engineered from a BphP from Deinococcus radiodurans (IFPs) that were bright enough for practical use4 (Fig. 1). In recent years, a variety of NIR FPs have been developed from different BphPs, such as Rhodopseudomonas palustris RpBphP2 and RpBphP6 (ref. 5). However, the BphP-derived NIR FPs are characterized by weak fluorescence and low stability1. In fact, it seems that BphPs may be better engineered as optogenetic tools rather than FPs, considering their photo-transforming activities in various signaling cascades.

Phycobiliproteins are brilliantly colored, highly fluorescent protein components of the photosynthetic light-harvesting antenna complexes of cyanobacteria, red algae, and cryptomonads. The three common phycobiliproteins are phycoerythrin (PE), with phycoerythrobilin chromophores, and phycocyanin (PC) and APC, both with PCB chromophores. The attachment of these chromophores requires lyase activity. Fluorescent phycobiliproteins are assembled into organized cellular structures called phycobilisomes. Because phycobiliproteins absorb intense incident light directly, they are expected to be photostable. They also participate in an efficient energy-transfer chain within the phycobilisome in the following sequence: PE → PC → APC → chlorophyll. Accordingly, phycobiliproteins have high molar extinction coefficients and fluorescence quantum yields. However, reconstitution of fluorescent APC with combined use of HO-1, PcyA, and lyase has been limited to Escherichia coli6,7.

Rodriguez et al.1 describe the successful development of a bright, stable, far-red fluorescent reporter using an APC α-subunit from Trichodesmium erythraem (TeAPCα). After performing mutagenesis relentlessly, they created apo-APCα mutants that can autocatalytically incorporate BV as well as PCB to fluoresce, bypassing the need for the lyase enzyme. The final evolved FP, named small ultra-red FP (smURFP), is a dimer with excitation/emission maxima of 642/670 nm (with BV attachment), which are very similar to those of the popular far-red dye Cy5 (649/670 nm). smURFP also shows photostability similar to that of standard EGFP in various mammalian cell types. A far-red/NIR Fucci probe8 was successfully created via fusion of smURFP and IFP2.0 to the ubiquitination domains of CDT1 and geminin, respectively. The effectiveness of smURFP for in vivo deep imaging was demonstrated with BV supplementation achieved by coexpression of HO-1 or use of a membrane-permeant BV analog (BVMe2).

Unfortunately, this study does not clarify the mechanism of BV attachment in smURFP. In fact, there is some biological diversity and uncertainty regarding BV attachment to BphP proteins as well5. Elucidation of the mechanism in detail will facilitate the rational design of biliprotein variants with more efficient reconstitution, as this represents a weakness of the current smURFP tool. Also, future work should examine how cells are affected by the sequestration and exogenous addition of BV, so that more physiological imaging experiments with biliproteins can be conducted. As an alternative strategy, mammalian expression of native APC may be achieved via the coexpression of PcyA, which can supply PCB.

In addition to the development of a new far-red FP, the work by Rodriguez et al.1 may also represent a step toward the creation of an artificial photosynthetic system in mammalian cells. In this regard, the complex interdependent quaternary structure of the phycobilisome containing APC hexamers can be seen as consisting of delicate building blocks, and their successful engineering inspires optimism about the eventual creation of such an artificial photosynthetic system, which could open new vistas in bioimaging and bioengineering.

Other novel types of biliproteins, such as UnaG9, which is derived from eel muscle and reversibly binds bilirubin as a fluorogenic chromophore, will also contribute to our understanding of the bilin–heme system and its applications. These technical and methodological innovations with bilins and their metabolites are now under way.


  1. 1

    Rodrigue, E.A. et al. Nat. Methods 13, 763–769 (2016).

    Article  Google Scholar 

  2. 2

    Tsien, R.Y. Annu. Rev. Biochem. 67, 509–544 (1998).

    CAS  Article  Google Scholar 

  3. 3

    Vierstra, R.D. & Zhang, J. Trends Plant Sci. 16, 417–426 (2011).

    CAS  Article  Google Scholar 

  4. 4

    Shu, X. et al. Science 324, 804–807 (2009).

    Article  Google Scholar 

  5. 5

    Shcherbakova, D.M., Baloban, M. & Verkhusha, V.V. Curr. Opin. Chem. Biol. 27, 52–63 (2015).

    CAS  Article  Google Scholar 

  6. 6

    Tooley, A.J., Cai, Y.A. & Glazer, A.N. Proc. Natl. Acad. Sci. USA 98, 10560–10565 (2001).

    CAS  Article  Google Scholar 

  7. 7

    Zhang, J. et al. Angew. Chem. Int. Edn. Engl. 49, 5456–5458 (2010).

    CAS  Article  Google Scholar 

  8. 8

    Sakaue-Sawano, A. et al. Cell 132, 487–498 (2008).

    CAS  Article  Google Scholar 

  9. 9

    Kumagai, A. et al. Cell 153, 1602–1611 (2013).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Atsushi Miyawaki.

Ethics declarations

Competing interests

The author declares no competing financial interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Miyawaki, A. Exploiting the cyanobacterial light-harvesting machinery for developing fluorescent probes. Nat Methods 13, 729–730 (2016). https://doi.org/10.1038/nmeth.3983

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