Bright and photostable yellow fluorescent proteins for extended imaging

Published on 2025/04/14                                                                                                             Research powered by Mightex’s Polygon1000     


This research was recently presented by Jihwan Lee at the ASCB (2024) in San Diego,CA,USA and was supported by a Mightex Travel Award to promote scientific communication and sharing of research using the Mightex Polygon.                                                                                                                                                                                                                                                                                                                                                                                                          Congratulations to Jihwan Lee! We can’t wait to see more from this exciting project!                                                                                                              

Jihwan Lee
Baylor College of Medicine,
Rice University, USA

Prof. François St-Pierre
Baylor College of Medicine,
Rice University, USA

 

Introduction

Metagenomic screening and protein engineering have produced a broad color palette of fluorescent  proteins (FPs) to visualize biological events and structures. However, many FPs rapidly  photobleach under repeated or prolonged illumination, limiting the duration and temporal  resolution in live cell microscopy. Rapid photobleaching is especially pronounced in high-power 

and long-exposure techniques such as super-resolution1, single-molecule2,3, and voltage imaging  4. StayGold, a highly photostable green FP reported recently, facilitates prolonged imaging  experiments 5–8. However, the photolability of FPs in other spectral ranges persists, limiting  extended multi-spectral imaging. Notably, yellow FPs (YFPs) typically photobleach faster than  counterparts from other spectral classes. Given the widespread use of YFPs in fluorescent  biosensors 9–11 and multi-spectral imaging 12–14, enhancing their photostability is imperative. 

Results 

Single-cell screening platform identifies two bright YFPs with improved photostability 

We sought to address the relative photolability of current YFPs via high-throughput protein  engineering. Screening was performed using SPOTlight, our pooled single-cell screening platform  15 (Fig. 1). We randomly mutagenized mGold 15, the most photostable monomeric YFP we  previously derived from mVenus 15 (Fig. 1a, step 1). We screened the resulting libraries in yeast  cells, evaluating the brightness and photostability of 1,125,438 cells representing 204,987 variants  across 7 rounds of screening (Fig. 1a, step 2). In each screening round, we optically labeled 200- 

400 single variants (Fig. 1a, step 3→4) for retrieval via Fluorescence Activated Cell Sorting  (FACS). Using the Mightex Polygon 1000 digital micromirror device was critical because it  enabled precise single-cell optical labeling. Compared to DMD from another company,  Mightex’s DMD provided greater illumination strength on the sample plane, enabling faster  optical tagging. Optically labeled cells were then grown into colonies on agar plates, picked,  expanded, and plated onto multi-well plates (Fig. 1a, step 4). We performed population-level  validation by quantifying the mean brightness and photostability of thousands of cells per variant,  and the top 5-10 variants were sequenced. 

Characterization and application of mGold2s and mGold2t 

The two best variants based on brightness and photostability were named mGold2s and mGold2t. When we imaged live mammalian cells expressing mGold2s and mGold2t using continuous widefield illumination of 520/5 nm (central wavelength/spectral width) light at 2.4 mW/mm2 for 1  hour, mGold2s and mGold2t retained 50% and 43% of their original fluorescence, corresponding  to photobleaching half-lives of 60 and 42 minutes, respectively (Fig. 1b). In contrast, mVenus and  mGold rapidly photobleached with half-lives of 3 and 12 minutes, respectively. mGold2s and  mGold2t showed different photobleaching patterns: mGold2t initially photobleached more rapidly  than mGold2s, but this trend reversed at later time points (Fig. 1c).

To demonstrate the utility of the photostable mGold2 variants, we assessed the performance of mGold2s and mGold2t in super-resolution imaging using structure-illuminated microscopy (SIM).  We imaged actin filaments in COS-7 cells expressing YFP-LifeAct fusions at a rate of 9 Hz for  ~1.5 minutes, using a 488-nm laser at 95 mW/mm2. Each time point comprised nine raw frames 

reconstructed into a single SIM image. The resolution of SIM images was further enhanced  through sparse deconvolution 16 (Fig. 1cd). mGold2s and mGold2t maintained over 80% of their  initial fluorescence, significantly outperforming mVenus, which retained only about 25% (Fig. 2e). mGold2s enabled visualization of actin dynamics similar to the “actin blip” dynamics described  previously 17 (Fig. 2f). This dynamic event involves actin punctum formation at the network node  followed by disassembly of the punctum and can detected as a spike in fluorescence intensity.  These dynamics could not be observed using photolabile mVenus because of rapid photobleaching.  

Conclusion 

We anticipate that these new variants will set a new gold standard in the realm of monomeric YFPs,  offering extended imaging durations and improved temporal resolution for biological research and  biotechnological applications. 

 

 

Figure 1. Development, characterization, and application of mGold2 YFPs. (a) SPOTlight screening  workflow used to develop bright and photostable YFP variants. (b) mGold2s and mGold2t expressed in  HEK293A cells retained around half of their fluorescence after continuous illumination of 520/5-n light at 2.4  mW/mm2 for 1 hr. (c) YFP-LifeAct fusions were expressed in COS-7 cells to visualize actin filaments. (d) Imaging workflow. (e) mGold2s and mGold2t shoed greater photostability than mVenus under TIRF-SIM with  488 nm light at 95 mW/mm2. Bottom graph: Traces denoted the mean, and the thin shaded area are the 95% CI. n = 40-50 regions of interest across 10 cells. (f) mGold2s enabled extended visualization of actin dynamics. The  arrow denotes an actin blip.

References 

1. Lambert, T. J. & Waters, J. C. Navigating challenges in the application of superresolution  microscopy. J. Cell Biol. 216, 53–63 (2016). 

2. Schirripa Spagnolo, C. & Luin, S. Choosing the Probe for Single-Molecule Fluorescence  Microscopy. Int. J. Mol. Sci. 23, 14949 (2022). 

3. Jain, A. et al. Probing cellular protein complexes using single-molecule pull-down. Nature 473, 484–488 (2011). 

4. Yang, H. H. & St-Pierre, F. Genetically Encoded Voltage Indicators: Opportunities and  Challenges. J. Neurosci. 36, 9977–9989 (2016). 

5. Hirano, M. et al. A highly photostable and bright green fluorescent protein. Nat. Biotechnol. 40, 1132–1142 (2022). 

6. Ando, R. et al. StayGold variants for molecular fusion and membrane-targeting applications.  Nat. Methods (2023) doi:10.1038/s41592-023-02085-6. 

7. Ivorra-Molla, E. et al. A monomeric StayGold fluorescent protein. Nat. Biotechnol. (2023)  doi:10.1038/s41587-023-02018-w. 

8. Zhang, H. et al. Bright and stable monomeric green fluorescent protein derived from  StayGold. Nat. Methods (2024) doi:10.1038/s41592-024-02203-y. 

9. Tutol, J. N. et al. Engineering the ChlorON Series: Turn-On Fluorescent Protein Sensors for  Imaging Labile Chloride in Living Cells. ACS Cent. Sci. 10, 77–86 (2024). 10. Ni, Q. et al. Signaling diversity of PKA achieved via a Ca2+-cAMP-PKA oscillatory circuit.  Nat. Chem. Biol. 7, 34–40 (2011). 

10. Ponsioen, B. et al. Detecting cAMP-induced Epac activation by fluorescence resonance  energy transfer: Epac as a novel cAMP indicator. EMBO Rep. 5, 1176–1180 (2004). 12. Cohen, S., Valm, A. M. & Lippincott-Schwartz, J. Multispectral live-cell imaging. Curr.  Protoc. Cell Biol. 79, e46 (2018). 

11. Livet, J. et al. Transgenic strategies for combinatorial expression of fluorescent proteins in  the nervous system. Nature 450, 56–62 (2007). 

12. Cai, D., Cohen, K. B., Luo, T., Lichtman, J. W. & Sanes, J. R. Improved tools for the  Brainbow toolbox. Nat. Methods 10, 540–547 (2013). 

13. Lee, J. et al. Versatile phenotype-activated cell sorting. Sci. Adv. 6, eabb7438 (2020). 16. Zhao, W. et al. Sparse deconvolution improves the resolution of live-cell super-resolution  fluorescence microscopy. Nat. Biotechnol. 40, 606–617 (2022). 

14. Mo, Y. et al. Quantitative structured illumination microscopy via a physical model-based  background filtering algorithm reveals actin dynamics. Nat. Commun. 14, 3089 (2023).

 

Author: Jihwan Lee

Bio: Temporary post-doctoral associate at St-Pierre Lab, Department of Neuroscience, Baylor College of Medicine, Rice University.                              Advisor: Dr. François St-Pierre.