introduction
Intracellular calcium (Ca2+) signaling is ubiquitous in biology and is a key element of cell signaling. While existing fluorescent sensors and reporters can detect activated cells with elevated Ca2+ concentrations, these methods require the delivery of light signals to deep tissues via implants and cannot be used non-invasively in free-roaming animals. On August 5, Nature Methods reported "Rapid, biochemical tagging of cellular activity history in vivo", introducing a novel enzyme-catalyzed method to rapidly biochemically label cells with elevated Ca2+ concentrations in vivo. The method utilizes a Ca2+-activated split-TurboID (CaST) enzyme to label activated cells within 10 minutes with the assistance of exogenous biotin molecules. The enzymatic signal is enhanced with increasing Ca2+ concentration and biotin labeling time, suggesting that CaST can act as a time-gated integrator of total Ca2+ activity. In addition, unlike transcriptional reporters, which take hours to generate a signal, readouts of CaST can occur immediately after active labeling. The dynamic change in the concentration of ions within the cell allows the cell to respond and adapt to its local environment, which ultimately contributes to the normal physiological functions of the body. For example, neurons, as the basic functional units of the brain, can be activated by various external stimuli or pharmacological compounds, resulting in rapid fluctuations in intracellular Ca2+ concentrations. Thus, the activity of complex neural networks can be directly measured by changes in intracellular Ca2+ levels. Genetically encoded Ca2+ indicators have dramatically changed our ability to record neural activity in awake and active animals. However, a major limitation of fluorescence sensors is that their readout signals are transient and often require invasive methods to obtain optical signals of deep brain structures. This makes it challenging to combine the activity history of a given neuron with its numerous other cellular properties, such as precise spatial localization, RNA expression, or protein expression. To overcome this issue, previous studies have designed orthogonal transcriptional reporter genes (e.g., FLARE, FLiCRE, Cal-Light) or fluorescent proteins (e.g., CaMPARI) to stably label cells activated at high intracellular Ca2+ levels. However, these methods all rely on light-sensitive proteins, requiring blue or ultraviolet light to limit the time window for labeling cellular activity. This requirement limits its scalability in deep brain regions or body regions where fiber implantation is not possible. Another method of stable labeling includes transcriptional reporters based on immediate early genes (IEGs) such as TRAP2 and tetTag, which utilize drug injection rather than light exposure to control the active labeling window. However, while IEG activity has been shown to correlate with neural activity in a variety of cell types, it is far inferior to Ca2+ as a universal readout signal. In addition, the slow onset of IEG expression limits the ability to immediately label and recognize activated neurons within a specific time window. In the study, the researchers engineered a Ca2+-dependent enzyme that reported elevated Ca2+ levels within living cells by attaching exogenous biotin molecules to activated cells. A proximity marker enzyme, split-TurboID, has been redesigned and repurposed to report elevated intracellular Ca2+ levels by tagging proteins with exogenous biotin molecules within living cells. This method is able to label cells in minutes without the need for light, providing a rapid, non-invasive new strategy for neural activity recording.
Intracellular calcium ion (Ca2+), as an important signaling molecule, is involved in almost all cell signaling processes. In neurons, changes in Ca2+ concentration are closely related to neural activity and are important indicators for studying neural network activity. Currently, activated cells with elevated Ca2+ concentrations can be detected using fluorescent sensors and reporter genes, but these methods often require invasive manipulations, such as implants, to deliver light signals to deep tissues, which makes non-invasive detection difficult in free-roaming animals. To overcome these limitations, the researchers devised an enzyme-catalyzed, biochemical labeling technique capable of rapidly labeling cells with elevated Ca2+ concentrations. This technology, called Ca2+-activated split-TurboID (CaST), is labeled in less than 10 minutes with the assistance of an exogenous biotin molecule. This method is not only fast, but also enables cell labeling without the use of light, greatly increasing its potential for application in deep brain regions and other in vivo tissues.
The research team first designed a Ca2+-dependent enzyme that recombinates and activates when Ca2+ concentrations rise, thereby labeling target cells. Specifically, CaST consists of two parts: calmodulin (CaM), which binds to Ca2+, and synthetic peptide M13. These two parts are attached to the two inactive fragments of split-TurboID (sTb(N) and sTb(C)), respectively. When intracellular Ca2+ concentrations are elevated, the CaM moiety is recruited to M13, allowing split-TurboID to reassemble and activate. In the presence of exogenous biotin, the recombinant split-TurboID labels itself and surrounding proteins.
The researchers expressed different versions of the CaST tool in HEK293T cells, treated with biotin and Ca2+, and stained with Alexa Fluor 647 (SA-647)-conjugated streptavidin to detect biotinylated proteins. The results showed that CaST was able to effectively label the target protein in the presence of exogenous biotin when the concentration of Ca2+ increased. To verify the effectiveness of CaST, the research team conducted a variety of experiments. They ligated the two fragments of CaST to different proteins and expressed them in HEK293T cells in different combinations. Through co-treatment of biotin and Ca2+, the researchers found that one of the combinations (membrane-bound CD4-sTb(C)-M13-GFP and CaM-V5-sTb(N) in the cytosol) exhibited the highest signal-to-background ratio (SBR). They further optimized the transfection ratio and found that a 5:2 transfection ratio (CD4-sTb(C)-M13-GFP(N)) yielded the best labeling results.
Using fluorescence microscopy and western blot, the researchers verified the labeling effect of CaST at different Ca2+ concentrations and stimulation times. The results showed that the labeling signal of CaST increased with the increase of Ca2+ concentration and the prolongation of stimulation time, indicating that it had a good time and concentration dependence. In specific experiments, the researchers expressed CaST in HEK293T cells and co-treated the cells with biotin and Ca2+. Fluorescence microscopy showed that the intracellular biotin labeling signal was significantly enhanced in the co-presence of biotin and Ca2+, while the labeling signal was weak or absent in the presence of only biotin or Ca2+ alone.
CaST的功能和有效性验证（Credit: Nature Methods）
Protein structure prediction: The protein structure of two halves of CaST was predicted using AlphaFold2. The diagram illustrates the structure of the two parts in the separate state (when no Ca2+ is expected) and in the composite state (when high Ca2+ is expected). The two-part protein revisibly reassembles under Ca2+ dependence. The predicted biotin binding sites are shown in blue. CaST design: Schematic diagram of the expression design of CaST in HEK cells. The moiety containing sTb(C)-M13-GFP is anchored to the membrane by the transmembrane domain of the CD4 cell membrane protein, while the CaM-V5-tag-sTb(N) moiety is expressed in the cytosol. CaST labels proteins only when cells are treated with biotin and exhibit elevated intracellular Ca2+. Fluorescence image: Confocal image of HEK cells transfected with both fractions of CaST simultaneously and treated with biotin for 30 min under conditions ± Ca2+. Cells were washed, fixed, and stained with anti-V5 and SA-647. The anti-V5 signal labels the CaM-sTb(N) moiety, while the GFP fluorescence shows the CD4-sTb(C)-M13 moiety. Biotinylation of proteins was detected by SA-647 staining. Western blot analysis: HEK cells were transfected with CaST and treated ± 50 μM biotin and ± Ca2+ (5 mM CaCl2 and 1 μM ionomycin) for 30 minutes. Cells were then washed with Dulbecco's phosphate buffered saline (DPBS), whole cell lysates were collected, and Western blot analysis was performed using streptavidin-horseradish peroxidase (SA-HRP) or anti-V5/HRP. 'N' indicates the expected CaM-V5-sTb(N) fragment size, while 'C' indicates the expected CD4-sTb(C)-M13-GFP fragment size. Quantification of biotinylated proteins: Biotinylated proteins in Western blot experiments were quantified. Two independent biological replicates were quantified, and a 75-kDa endogenous biotinylated band below the entire band was included in the quantification (sum of total raw intensity pixel values).
To test the stability of CaST, the researchers treated HEK cells with Ca2+ for 30 minutes, washed them for 10 minutes, and then added biotin for labeling for 30 minutes. The results showed that there was no significant labeling signal in the washed cells, indicating that the recombination and activation of CaST was reversible. In addition, the titration experiment was performed by increasing the concentration of Ca2+, and the results showed that the labeled signal of CaST was linearly correlated with the concentration of Ca2+, which further verified its high sensitivity and specificity.
Studies have shown that CaST is able to efficiently label cells with elevated Ca2+ concentrations in vitro and in vivo. In HEK cell experiments, CaST exhibits a significant labeling signal when biotin and Ca2+ are co-present, while the labeling signal is weak or absent when only biotin or Ca2+ is present. This suggests that CaST can be used as a Ca2+-dependent labeling tool to effectively distinguish cellular activity under different conditions. In further experiments, the researchers verified the labeling effect of CaST at different Ca2+ concentrations and stimulation times by fluorescence microscopy and Western blotting. The results showed that the labeling signal of CaST increased with the increase of Ca2+ concentration and the prolongation of stimulation time, indicating that it had a good time and concentration dependence.
CaST的性能（Credit: Nature Methods）
Fluorescence microscopy image of HEK cells: The image shows CaST-transfected HEK cells treated at 50 μM biotin and ± Ca2+ (5 mM CaCl2 and 1 μM ionomycin) for 30 minutes. The top shows SA-647-stained biotinylated protein, and the bottom shows CD4-sTb(C)-M13-GFP. Single-cell analysis: Single-cell analysis results show scatter plots of SA-647 labeling and GFP fluorescence intensity of cells at different GFP expression levels. The results showed that SA-647 staining increased significantly under the addition of biotin and Ca2+, but not when biotin was added. Distribution of SA-647/GFP Fluorescence Ratio: The violin plot illustrates the distribution of SA-647/GFP fluorescence ratio in single cells under different treatment conditions. The SA-647/GFP ratio was significantly higher than that of other conditions under the condition of biotin and Ca2+. Schematic diagram of CaST-IRES construction design: The design of a bicistronic CaST-IRES construction is shown for simultaneous expression of two parts of CaST in HEK cells. Fluorescence image of CaST-IRES: HEK cells transfected with CaST-IRES were treated at 50 μM biotin and ± Ca2+ for 30 minutes. Similar to before, the top shows SA-647-stained biotinylated protein, and the bottom shows CD4-sTb(C)-M13-GFP. Single-cell analysis (CaST-IRES): Scatter plot showing the average SA-647 vs. the average GFP fluorescence intensity of GFP-positive cells under different treatment conditions. The results showed that SA-647 staining increased significantly under the addition of biotin and Ca2+, but not when biotin was added. Distribution of SA-647/GFP fluorescence ratios (CaST-IRES): Violin plot showing the distribution of SA-647/GFP fluorescence ratios in single cells under different treatment conditions. The SA-647/GFP ratio was significantly higher than that of other conditions under the condition of biotin and Ca2+. CaST Non-IRES vs. IRES Versions: The chart shows the distribution of the SA-647/GFP fluorescence ratio between CaST non-IRES and IRES versions under different conditions. The labeling effect of the IRES version of CaST with biotin and Ca2+ was significantly better than that of the non-IRES version. ROC Curve Analysis: ROC curve analysis shows the ability of CaST non-IRES and IRES versions to distinguish between Ca2+-treated and non-treated cell populations. The AUC of the non-IRES version was 0.87, while the AUC of the IRES version was 0.93, indicating that the IRES version has a higher precision in differentiating between activated and unactivated cells.
Neuronal experimentsResearchers applied CaST to cultured neurons to verify its effectiveness in nerve cells. They used adeno-associated virus (AAV) to express CaST in rat hippocampal neurons and stimulated the neurons with potassium ions (KCl) to observe changes in Ca2+ concentration and CaST labeling. The results showed that CaST was able to effectively label neurons with elevated Ca2+ concentrations within 10 minutes after KCl stimulation. In addition, the researchers also used CaST to detect the effects of different pharmacological agents on the concentration of Ca2+ in neurons. For example, the modulatory effect of dopamine (DA) and 2,5-dimethoxy-4-iodoamphetamine (DOI) on neuronal Ca2+ concentrations. The results showed that DOI could significantly increase the Ca2+ concentration in neurons and cause CaST labeling, while DA could not significantly affect the Ca2+ concentration.
CaST在培养的神经元中的性能（Credit: Nature Methods）
Expression and labeling of CaST in neurons: Part A shows fluorescence microscopy images of CaST expression in cultured rat hippocampal neurons. These neurons are infected with adeno-associated virus (AAV), express CD4-sTb(C)-M13-GFP and CaM-sTb(N) moieties, and are treated for 30 min under ±biotin and ±KCl conditions. The image shows SA-647 staining of the biotinylated protein with GFP fluorescently labeled CD4-sTb(C)-M13. SA-647/GFP ratios under different treatment conditions: Part b quantifies the SA-647/GFP fluorescence ratio under different treatment conditions in panel a. The results showed that the ratio of +biotin +KCl was significantly higher than that of other conditions. CaST labeling after 10 min KCl stimulation: Part c shows fluorescence microscopy images of neurons after treatment under 10 min ± biotin and ± KCl conditions. The results showed that SA-647 labeling was significantly enhanced under the co-treatment of biotin and KCl. 10 min labeled SA-647/GFP ratio: Part d quantifies the SA-647/GFP fluorescence ratio under different treatment conditions in panel c. The results showed that the SA-647/GFP ratio was significantly increased under 10 minutes of KCl and biotin co-treatment. SA-647 labeling ratio in GFP+ neurons: Part e quantifies the proportion of SA-647+ cells in GFP+ neurons during 10 min and 30 min labeling experiments. The results showed that SA-647+ cells labeled at 10 minutes accounted for about 35% of GFP+ neurons and about 65% labeled at 30 minutes under the co-treatment of biotin and KCl. However, the proportion of SA-647+ cells was lower (about 10%) under biotin-only treatment. CaST labeling after different drug treatments: Part f shows fluorescence microscopy images of rat hippocampal neurons after 30 min treatment at 50 μM biotin and 10 μM DA, 10 μM DOI, or 30 mM KCl. The results showed that DOI and KCl treatments significantly increased SA-647 markers, while DA treatments did not significantly affect SA-647 markers. SA-647/GFP ratios under different drug treatment conditions: Part g quantifies the SA-647/GFP fluorescence ratio under different drug treatment conditions in panel f. The results showed that KCl and DOI treatments significantly increased the SA-647/GFP ratio, while DA treatment did not significantly affect the ratio.
Animal researchers further validated the application of CaST in mouse models. They express CaST in the mouse prefrontal cortex (PFC) and induce neuronal activity by psychedelic drugs such as psilocybin. By measuring the head-twitch response (HTR) in mice, the researchers found that psilocybin was able to significantly increase the Ca2+ concentration of PFC neurons and cause CaST labeling. This result demonstrates that CaST is able to non-invasively label and detect neural activity in free-moving mice.
CaST在小鼠体内非侵入性标记和检测psilocybin激活的神经元的能力（Credit: Nature Methods）
Schematic diagram of the experiment: Part a shows the experimental workflow for neurons activated with the CaST tag psilocybin. After injection of CaST virus, mice are treated with psilocybin in a freely active state and activated neurons are labeled by biotin. SA-647 staining and head swing response (HTR) measurements were subsequently performed. SA-647 and GFP fluorescence images in the mPFC region: Part b shows fluorescence microscopy images of the prefrontal cortex (mPFC) region compared to mice injected with biotin + saline and biotin + psilocybin. The results showed a significant increase in SA-647 labeling in the mPFC region of psilocybin-treated mice. SA-647 vs. GFP fluorescence intensities of individual neurons: Part c quantifies the SA-647 vs. GFP fluorescence intensities of each GFP+ neuron in Part b. The horizontal dashed line represents the 90th percentile threshold for all SA-647 neurons in the biotin + saline group. SA-647/GFP ratio of FOV: Part d quantifies the SA-647/GFP fluorescence ratio for different fields of view (FOV) in part c. The results showed that the proportion of mice treated with psilocybin was significantly higher than that of mice treated with normal saline. Proportion of SA-647+ neurons: Part e shows the proportion of SA-647+ neurons among all GFP+ neurons. The results showed that approximately 70% of GFP+ neurons in the psilocybin-treated group exhibited strong SA-647 labeling, while this proportion was lower in the saline-treated group. Correlation of SA-647+ neurons with HTR: Part f shows the cell mask of SA-647+ mPFC neurons during HTR measurements. Fields of view with the same amount of HTR from the same mice, but independent CaST injections from the contralateral hemisphere. Relationship between the number of HTRs and the number of SA-647+ neurons: Part g quantifies the number of SA-647+ neurons per square millimeter in the data in Part F as a function of the number of HTRs. The results showed that the number of SA-647+ neurons was positively correlated with the number of HTRs. Relationship of HTR number to SA-647/GFP ratio: Part h quantifies the relationship between the average SA-647/GFP ratio and HTR number for all neurons in part f data. The results showed a significant increase in the SA-647/GFP ratio in mice with HTR in the psilocybin-treated group. Comparison with c-Fos: Part I shows CaST GFP, SA-647, and c-Fos stained images of the mPFC region after psilocybin treatment. The results showed that psilocybin treatment significantly increased SA-647 labeling, but had little effect on c-Fos labeling. Number of c-Fos+ neurons: Part J quantifies the number of c-Fos+ neurons per square millimeter. The results showed that psilocybin treatment did not significantly increase the c-Fos labeling in the mPFC region. Number of SA-647+ neurons: The number of SA-647+ neurons per square millimeter is quantified in the k part. The results showed that psilocybin treatment significantly increased SA-647 labeling in the mPFC region. SA-647+/GFP+ neuron ratio: Part l shows the proportion of SA-647+ neurons to GFP+ neurons per square millimeter FOV. The results showed that psilocybin treatment significantly increased this ratio.
The advent of CaST technology provides a new tool for non-invasive labeling of cellular activity. Compared with traditional fluorescence sensors and transcriptional reporters, CaST is characterized by fast labeling, non-invasiveness, and high efficiency. This makes it promising for a wide range of applications in deep brain regions and other tissues in the body that are difficult to access optically. In the future, CaST is expected to be used to study the relationship between neural network activity and behavior. For example, when studying the effects of psychedelic drugs on neural activity, CaST can be used to label and analyze neuronal activity under the influence of drugs. In addition, CaST can be combined with other spatial molecular imaging techniques to reveal gene expression and protein distribution of activated neurons within a specific time window, providing a more comprehensive view of neuroscience research.
In conclusion, CaST, as a fast, non-invasive, and efficient tool for labeling cellular activity, has broad application prospects and important research value. By further optimizing and expanding its range of applications, CaST is expected to have a profound impact in neuroscience and other biomedical fields.
bibliography
Zhang R, Anguiano M, Aarrestad IK, Lin S, Chandra J, Vadde SS, Olson DE, Kim CK. Rapid, biochemical tagging of cellular activity history in vivo. Nat Methods. 2024 Aug 5. doi: 10.1038/s41592-024-02375-7. Epub ahead of print. PMID: 39103446.https://www.nature.com/articles/s41592-024-02375-7
Editor-in-charge|Explore Jun
Typography |
Please indicate the source of reprinting【Biological Exploration】
End