TL;DR
Researchers have developed mNG-GECO1, a new genetically encoded calcium indicator (GECI) based on the bright mNeonGreen fluorescent protein. This indicator offers significantly improved brightness compared to existing GCaMP sensors while maintaining comparable performance in various biological systems, paving the way for next-generation calcium imaging tools.
What Problem Does This Paper Address?
Imaging calcium ion (Ca²⁺) dynamics is crucial for understanding cell signaling and neural activity. Genetically encoded calcium indicators (GECIs) based on fluorescent proteins (FPs) are powerful tools for this purpose, allowing for targeted expression and long-term imaging. However, the widely used GCaMP series, derived from Aequorea victoria GFP (avGFP), may be approaching its performance limits due to the inherent properties of avGFP. This necessitates the development of new GECIs with improved brightness and performance to sustain progress in the field. Additionally, existing GECIs can suffer from Ca²⁺ buffering, leading to potential artifacts and altered cellular function, which can be mitigated by reducing expression or improving indicator properties.
Methods
The development of mNG-GECO1 involved a multi-pronged approach combining rational design, linker optimization, and iterative directed evolution.
- Rational Design & Initial Construction: The team started with a topological variant of a previously developed GECI (REX-GECO1) and inserted a calmodulin (CaM) and RS20 peptide calcium-sensing domain into the mNeonGreen (mNG) scaffold. This resulted in an initial prototype, mNG-GECO0.1.
- Optimization of Insertion Site: Deletion of a specific residue (Ala146) near the insertion site significantly improved the Ca²⁺ response, leading to mNG-GECO0.2.
- Iterative Directed Evolution: This core methodology involved rounds of error-prone PCR or site-saturation mutagenesis to generate libraries of variants. These libraries were then screened using a fluorescent colony screening system to identify variants with increased brightness and Ca²⁺ response. Promising candidates were re-tested and used as templates for subsequent rounds.
- Linker Optimization: Specific linker regions connecting mNG to the CaM-RS20 domain were further optimized using site-saturation mutagenesis to improve response kinetics and amplitude.
- Performance Optimization: Further directed evolution rounds focused on enhancing overall performance, leading to the final mNG-GECO1 variant.
- Characterization: The performance of mNG-GECO1 was rigorously evaluated in vitro (purified protein), in cultured cells (HeLa), dissociated neurons, and in vivo (zebrafish larvae). Techniques included measuring excitation/emission spectra, Ca²⁺ response (ΔF/F₀), binding affinity (K<0xE1><0xB5><0x83>), kinetic rates (k<0xE1><0xB5><0x92><0xE2><0x82><0x9F>, k<0xE1><0xB5><0x92><0xE2><0x82><0x9F>), extinction coefficients, quantum yields, two-photon cross-sections, molecular brightness, and signal-to-noise ratios (SNR).
Datasets
The study did not introduce new benchmark datasets. Instead, it utilized established experimental systems for evaluation:
- Cultured Cells: HeLa cells were used for imaging Ca²⁺ oscillations.
- Dissociated Neurons: Primary rat hippocampal neurons were used for field stimulation assays.
- Zebrafish Larvae: Transgenic zebrafish lines were generated for in vivo imaging of neural activity.
- Human iPSC-derived Cardiomyocytes: These cells were used to assess mNG-GECO1's performance in a different cell type and compare it to traditional chemical dyes.
Key Insights and Ideas
- mNeonGreen as a Superior Scaffold: The bright and monomeric mNeonGreen FP proved to be an excellent starting point for developing a new generation of GECIs, surpassing the limitations of avGFP.
- Directed Evolution is Key: Iterative directed evolution was instrumental in achieving significant improvements in both brightness and Ca²⁺ response, demonstrating its power in optimizing fluorescent biosensors.
- Linker Optimization Matters: Fine-tuning the linker regions between the FP and the calcium-sensing domain significantly impacts the indicator's performance, particularly its kinetics.
- Brightness Trade-offs: While mNG-GECO1 is brighter, its Ca²⁺ affinity (K<0xE1><0xB5><0x83>) is higher than GCaMP6s, suggesting a potential trade-off that might affect single-spike detection in certain scenarios.
- Broad Applicability: mNG-GECO1 demonstrates robust performance across diverse biological contexts, from cultured cells and neurons to in vivo zebrafish and cardiomyocytes, highlighting its versatility.
Key Results
- Brightness: mNG-GECO1 is approximately 60% brighter in vitro than GCaMP6s. Its in vitro molecular brightness is 21.3, compared to 16.1 for GCaMP6s.
- Ca²⁺ Response (ΔF/F₀):
- In vitro: mNG-GECO1 showed a ΔF/F₀ of 35, comparable to GCaMP6s (39).
- HeLa cells: Average ΔF/F₀ was 4.50 ± 2.96 for mNG-GECO1 vs. 3.48 ± 2.40 for GCaMP6s. Maximum ΔF/F₀ was 16.8 ± 10.5 vs. 12.8 ± 6.11, respectively.
- Dissociated neurons: Single action potential (AP) ΔF/F₀ was 0.19 ± 0.04 for mNG-GECO1 vs. 0.27 ± 0.09 for GCaMP6s. For 160 APs, it was 6.5 ± 0.8 vs. 9.0 ± 0.47.
- Zebrafish larvae: Maximum ΔF/F₀ was 3.09 ± 0.08 for mNG-GECO1 vs. 4.56 ± 0.11 for GCaMP6s.
- Kinetics:
- In vitro koff: mNG-GECO1 (1.57 s⁻¹) is faster than GCaMP6s (1.06 s⁻¹).
- Dissociated neurons: Half rise time for mNG-GECO1 was 49 ms vs. 65 ms for GCaMP6s (3 APs). Half decay time was 582 ms vs. 1000 ms.
- Zebrafish larvae: Half decay time was 1.98 s⁻¹ for mNG-GECO1 vs. 3.00 s⁻¹ for GCaMP6s.
- Ca²⁺ Affinity (K<0xE1><0xB5><0x83>): mNG-GECO1 has a higher K<0xE1><0xB5><0x83> (810 nM) compared to GCaMP6s (147 nM), indicating lower affinity for Ca²⁺.
- Signal-to-Noise Ratio (SNR): In zebrafish larvae, mNG-GECO1 had a higher SNR (6.63 ± 0.07) than GCaMP6s (5.25 ± 0.04).
- Toxicity: mNG-GECO1 showed no obvious morphological anomalies in zebrafish larvae and performed comparably or better than chemical dyes in iPSC-CMs, suggesting low toxicity.
Limitations and Future Work
- Ca²⁺ Affinity: The higher K<0xE1><0xB5><0x83> of mNG-GECO1 compared to GCaMP6s might limit its ability to accurately detect single action potentials in certain contexts, especially at lower firing rates. The authors suggest that increasing Ca²⁺ affinity would be a key area for future development to enable more accurate single spike detection.
- Further Optimization: While mNG-GECO1 represents a significant advancement, the authors note that further directed evolution efforts will be required to produce variants with substantial advantages over the jGCaMP7 series. They propose that mNG-GECO1, like the first-generation GCaMP, will serve as a parent for a new lineage of improved GECIs.
- Comparison to Latest GCaMP Variants: While compared to GCaMP6s, the abstract mentions GCaMP7 as the most recent generation, and the conclusion suggests further development is needed to surpass jGCaMP7. This implies that while mNG-GECO1 is a strong contender, it may not yet outperform the absolute cutting edge of GCaMP technology in all aspects.
Summary updated