Summary

4D Light-sheet Imaging of Zebrafish Cardiac Contraction

Published: January 05, 2024
doi:

Summary

This protocol utilizes light-sheet imaging to investigate cardiac contractile function in zebrafish larvae and gain insights into cardiac mechanics through cell tracking and interactive analysis.

Abstract

Zebrafish is an intriguing model organism known for its remarkable cardiac regeneration capacity. Studying the contracting heart in vivo is essential for gaining insights into structural and functional changes in response to injuries. However, obtaining high-resolution and high-speed 4-dimensional (4D, 3D spatial + 1D temporal) images of the zebrafish heart to assess cardiac architecture and contractility remains challenging. In this context, an in-house light-sheet microscope (LSM) and customized computational analysis are used to overcome these technical limitations. This strategy, involving LSM system construction, retrospective synchronization, single cell tracking, and user-directed analysis, enables one to investigate the micro-structure and contractile function across the entire heart at the single-cell resolution in the transgenic Tg(myl7:nucGFP) zebrafish larvae. Additionally, we are able to further incorporate microinjection of small molecule compounds to induce cardiac injury in a precise and controlled manner. Overall, this framework allows one to track physiological and pathophysiological changes, as well as the regional mechanics at the single-cell level during cardiac morphogenesis and regeneration.

Introduction

The zebrafish (Danio rerio) is a widely used model organism for studying cardiac development, physiology, and repair due to its optical transparency, genetic tractability, and regenerative capacity1,2,3,4. Upon myocardial infarction, while structural and functional changes impact the cardiac ejection and hemodynamics, technical limitations continue to hinder the ability to investigate the dynamic process during cardiac regeneration with the high spatiotemporal resolution. For example, conventional imaging methods, such as confocal microscopy, have limitations in terms of imaging depth, temporal resolution, or phototoxicity for capturing the dynamic changes and assessing cardiac contractile function during multiple cardiac cycles5.

Light-sheet microscopy represents a state-of-the-art imaging method that successfully addresses these issues by quickly sweeping the laser across the heart's ventricle and atrium, achieving detailed images with enhanced spatiotemporal resolution and negligible photo-bleaching and photo-toxic effects6,7,8,9,10,11.

This protocol introduces a comprehensive imaging strategy that includes LSM system construction, 4D image reconstruction, 3D cell tracking, and interactive analysis to capture and analyze the dynamics of cardiomyocytes across the entire heart during multiple cardiac cycles12. The customized imaging system and computational methodology allow one to track the myocardial microstructure and contractile function at the single-cell level in transgenic Tg(myl7:nucGFP) zebrafish larvae. Furthermore, small molecule compounds were delivered into the embryos using microinjection to assess drug-induced cardiac injury and subsequent regeneration. This holistic strategy provides an entry point to in vivo investigate structural, functional, and mechanical properties of myocardium at the single-cell level during cardiac development and regeneration.

Protocol

Approval for this study was granted by the Institutional Animal Care and Use Committee (IACUC) of the University of Texas at Dallas, under protocol number #20-07. Tg(myl7:nucGFP) transgenic zebrafish larvae12 were used for the present study. All data acquisition and image post-processing were carried out using open-source software or platforms with research or educational licenses. The resources are available from the authors upon reasonable request.

1. Zebrafish breeding and embryo microinjection

Timing: 2 days

  1. Maintain and breed adult zebrafish through standard care and breed procedures. For detailed methods, refer to previous report13.
  2. Perform microinjection following established protocol14. For the present study, microinjection was performed at the 1-2 cell (zygote) stage.
    ​NOTE: It is crucial to accurately identify this stage for effective microinjection and to ensure developmental consistency. During the one-cell stage, a small dome forms atop the yolk, and this stage typically lasts approximately 12 min. Upon progressing to the two-cell stage, two adjacent domes become visible, and this stage endures for about 45 min post-fertilization15. More detailed information of system construction can be found in other reports or protocols12,16,17

2. Zebrafish embryos/larvae preparation and mounting

Timing: 7 days

  1. Transfer embryos at 1 day post fertilization (dpf) to E3 water with 0.2 mM 1-Phenyl-2-thiourea (PTU) (see Table of Materials) to prevent pigment formation.
  2. Replace the medium with fresh E3 water with PTU every day until LSM imaging.
  3. Image embryos or larvae with the stereo microscope to record their development at the desired timepoints between 0 and 7 dpf.
  4. Prepare segmented fluorinated ethylene propylene (FEP) tubes with 2 mm inner diameter for embedding zebrafish larvae under the LSM system12.
    NOTE: The FEP tube is utilized to secure the sample with refractive index matching materials that closely resemble the surrounding medium, such as water.
  5. Starting from 3 dpf, transfer fish to a 150 mg/L tricaine (MS-222) solution (see Table of Materials) to immobilize fish before imaging.
  6. Prepare 0.8 % low-melting agarose with 150 mg/L tricaine and use a transfer pipette to move anesthetized fish to the agarose once it has cooled to room temperature18.
  7. Use another transfer pipette to mount the fish with agarose in FEP tubes (Figure 1).
    NOTE: To ascertain the minimal stress in the developing embryos, pre- and post-fixation examinations of cardiac activity, such as heart rate, were conducted under microscopic observation. These assessments aimed to identify any significant differences before and after tricaine anesthesia. An additional examination of the skin and musculature for irregularities in cardiac structure, such as edema, could also be conducted post-fixation. The concentration of tricaine also plays a crucial role in cardiac imaging19, and therefore the 150 mg/L concentration tricaine solution was utilized for the imaging of contraction in zebrafish larvae in this project. While the present retrospective synchronization allows us to address the irregular heartbeats20, a comprehensive solution for the 4D imaging of cardiac arrhythmias and long-term imaging in zebrafish is still under development.

3. Light-sheet imaging system setup and configuration

Timing: 3-14 days

  1. Construct an in-house LSM system based on a cylindrical lens using continuous-wave diode-pumped solid-state (DPSS) laser systems at specific wavelengths such as 473 nm and 532 nm as illumination sources (Figure 2A).
  2. Customize LabVIEW (see Table of Materials) control codes for the synchronization of translational sample stage, light-sheet illumination, and exposure of sCMOS camera to capture image sequences.
    ​NOTE: More detailed information of system construction can be found in other reports or protocols12,16,17. We encourage research groups to seek collaborative opportunities with laboratories that possess established expertise in optical imaging during system construction.

4. Zebrafish imaging preparation and data collection

Timing: 1 day

  1. Calibrate LSM system spatial resolution and minimize opacity variation at varying depths by measuring fluorescence beads.
    1. First, dilute fluorescent beads (see Table of Materials) having the diameter of 0.53 µm to a concentration of 1: 1.5 x 105 using 0.8 % low-melting agarose and transfer the solution to the segmented FEP tube.
    2. Second, use the LSM system to image the beads and measure the point spread function (PSF) across the entire sample, validating the spatial resolution and system alignment12.
      NOTE: In this system, the analysis of representative results for the full width at half maximum (FWHM) indicates lateral and axial resolutions of 1.26 ± 0.15 and 2.48 ± 0.15 µm (n = 60 beads), respectively. This suggests a consistent spatial resolution throughout the imaging depth.
  2. Attach the FEP tube with mounted zebrafish larva securely to the 6-axis (x, y, z, pitch, yaw, and roll) motorized sample stage, and submerge the tubes into the sample chamber filled with E3 water. To avoid the light scattering by the yolk sac and better locate the position of the heart, rotate the zebrafish larva to take image from the ventral side (Figure 2B). In this case, most captured images would include both the ventricle and atrium (Figure 2C).
    NOTE: Given that the current imaging procedures typically lasted less than one hour per fish and were conducted in a controlled room temperature environment, the chamber temperature control and oxygen level regulation are deemed optional for this project. However, for longer-term time-lapse imaging studies, particularly those focused on developmental processes, continuous temperature monitoring and regulation at 27 °C within the sample chamber is suggested to ensure the physiological environment of the zebrafish and to prevent any potential developmental abnormalities.
  3. Connect the motorized sample stage to the workstation and configure all necessary parameters, including the desired moving mode.
  4. Establish a connection between the sCMOS camera and the workstation. Adjust all essential settings, such as an exposure time of 5 ms and the desired region of interest.
  5. Configure the number of image sequences and frames within the customized LabVIEW control program.
  6. Power on the laser and initiate LSM imaging of the sample. Maintain the process until all image sequences have been successfully recorded.
  7. Record 300 frames as a 2D image sequence to cover 3-5 cardiac cycles of the larva with a framerate of 200 frames/second. The recording captures cardiac contractility at the selective plane illuminated by the light-sheet.
  8. Move the larva at the step size of 1 µm across the light-sheet illumination to virtually slice the sample.
    NOTE: The step size of the sample stage should be set based on the axial resolution of the light-sheet system, following the Nyquist-Shannon sampling theorem12.
  9. Repeat step 7 to record another image sequence of the new sample slice until the whole heart is covered. Generally, 100-200 image sequences with the step size of 1 µm ensure the coverage of the entire heart of a zebrafish larva (Figure 2D).
    ​NOTE: Step 4.7-4.9 are controlled by the LabVIEW program and automatically performed by the LSM system (Figure 3). Given the extensive volume of image data generated, a standardized naming convention is recommended for image sequences. For example, designate folders with names such as "z-1", "z-2", etc., to categorize data across various image sequences. Within each sequence, the images should be named sequentially (e.g., "1", "2", …) to denote distinct time points. The total time for mounting fish, adjusting the imaging angle, and collecting all the data for one zebrafish larva is around 1 h.

5. 4D image reconstruction with parallel computation

Timing: 1 day
NOTE: The 4D reconstruction algorithm developed by our group and sample data are publicly accessible21. This method allows one to reconstruct the 4D heart image from the image sequences collected in previous steps (Table 1).

  1. Open the file test_Parallel.m in MATLAB (see Table of Materials). Specify the folder location where the raw image sequences are stored in the variable "baseDir".
  2. Assign the variable "numOfSlice" with the total number of image sequences (typically 100-150) and the variable "numOfImage" with the number of images (typically 300-500) in each sequence.
  3. Inspect the image sequence that shows the middle plane of the zebrafish heart (for example, the 51st sequence out of 101 total sequences). Identify the frame numbers of the first and fourth systoles in this sequence and assign them to the variables systolicPoint_1st and systolicPoint_4th.
  4. Click on Run to start the process.
    NOTE: The length of the cardiac cycle in each image sequence will first be identified by the MATLAB program through a comparison of the similarity (sum of squared differences) between images at different time points. Subsequently, the cardiac cycle will be estimated using the average of the cycle lengths from all image sequences. Next, the starting points of image sequences at different axial locations will be aligned by the program through an iterative comparison of the image similarity from the first image sequence to the last image sequence.
  5. Check the results in the output folder. The program will save the images as ".tif" files with time stamps. Each ".tif" file is a 3D heart image at a specific time point. The time interval between two 3D images is equal to the exposure time that is set.
    ​NOTE: Reconstruct the acquired images from the previous steps to depict the 4D (3D spatial + 1D temporal) cardiac contractions. To enhance proficiency in high-throughput investigations during the retrospective synchronization, this approach enables simultaneous computational operations on a GPU by leveraging multiple CPU cores through the MATLAB parallel computing toolbox and transforming images into the gpuArray format.

6. 3D cell segmentation and cell tracking

Timing: 1 day

  1. Download 3DeeCellTracker package (see Table of Materials) and set up the Python environment22.
  2. Download ITK-SNAP annotation software23 (see Table of Materials) and use it to manually label the 3D heart image in two selected timepoints, one at the ventricular diastole and the other at the ventricular systole, to create the training and validation datasets.
    NOTE: Other labeling software may also be applicable.
  3. In Python, run the 3DeeCellTracker training program and initialize the noise_level(100 in this case), folder_path and model parameters in TrainingUNet3D function to set the predefined 3D U-Net model.
  4. In MATLAB, use imageDimConverter.m program to convert and rename the training and validation dataset to the proper format for loading.
  5. In Python, load the training and validation datasets by using trainer.load_dataset() and trainer.draw_dataset() functions.
  6. In Python, run the first part of 3DeeCellTracker tracking program and initialize the parameters.
    NOTE: This includes defining image parameters to characterize image dimensions, segmentation parameters to select the machine learning model applied for cell segmentation, and tracking parameters to select the pre-trained deep neural network models employed for cell registration. For the first time use, employing the U-Net model for cell segmentation and FFN+ PR-GLS for cell registration is recommended. It is needed to store data, model, and results in folders that will be automatically created in this step.
  7. In MATLAB, use imageDimConverter.m program to convert and rename all the 3D heart images to the proper format and transfer them to the data folder created in the last step.
  8. In Python, run the second part of 3DeeCellTracker program to start segmentation.
  9. Once the first 3D image is segmented, compare the segmentation result with the raw image, and perform manual correction if any incorrect segmentation is found. Move the corrected segmentation to the created "/manual_vol1" folder.
  10. In Python, run the third part of 3DeeCellTracker program to segment all the images.
  11. Use Amira (see Table of Materials) to perform a visual assessment of tracking outcomes by comparing the positions of tracked cells with their corresponding raw images. If test results are not satisfying, recommence the procedure from step 6.6, employ an alternative machine learning model for segmentation or cell registration, and then re-initiate the tracking process.
    ​NOTE: After the segmentation, the data of cell labels is stored as an 8-bit image (0 to 255) by default, which means only 256 scales are provided as labels for cell classification. There are two solutions for addressing this issue when additional classes are needed. One solution is to set the format of data of cell labels as a 16-bit image in the tracking program, which will cause an exponential increase in the processing time, more than two days in the present case. The other solution is to use the "separateTrackingResults.py" program to post-process the cell labels, which will separate cells with the same labels based on their physical location and assign each cell a different label. This process takes around 10 min in this case.

7. Cardiac contractility analysis in the virtual reality mode

Timing: 1 day

  1. Acquire the cell tracking outcome from previous steps.
  2. Manually validate the data and select cells with consistent image intensity across all volumes (about 500 cells out of approximately 600).
  3. Use cellLabelsToObj.ipynb script21 to generate a surface mesh for each individual cell through 3D Slicer software24 (see Table of Materials) and assign a unique color code to each cell.
  4. Export each 3D model, consisting of all cells in a single time point, as a single .obj file consisting of multiple sub-objects accompanied by a .mtl file to describe the cell label.
  5. Import the models into Unity using the educational license, a development engine used for extended reality.
  6. Apply the customized scripts consisting of functions written in C# to the models and user-interface elements to allow for 4D visualization and interactive analysis. Perform VR interaction following step 7.7.
  7. Interact with the models in virtual reality (VR) using the following functions (Figure 4):
    1. Cell selection: Select up to two cells at a time and view their trajectories, velocities, volumes, surface areas, and relative distances.
    2. Time point selection: Choose a specific time point in the data to analyze the outputs of the selected cells, such as at t = 0 ms during diastole, or at t = 200 ms during systole.
    3. Time pause: Freeze the dynamic model to focus on the selected cells and their outputs.
    4. Contrast: Modulate the contrast between the selected cells and surrounding cells in the model.

Representative Results

The current protocol consists of three main steps: zebrafish preparation and microinjection, light-sheet imaging and 4D image reconstruction, and cell tracking and VR interaction. Adult zebrafish were allowed to mate, the fertilized eggs were collected, and performed microinjection as needed for the proposed experiments (Figure 1). This step provides an entry point to explore zebrafish applications in the investigation of cardiac development and regeneration, and it also plays a crucial role in the subsequent imaging and analysis. The contracting heart was imaged at different stages (from 3 dpf to 7 dpf) using the custom-built LSM system and aligned the image sequences along the z-axis to reconstruct the 4D zebrafish heart model (Figure 2 and Figure 3). These models provide a basis for deep phenotypic characterizations and are pivotal in revealing the dynamics of the cardiac morphology and function across developmental timelines. The individual cells in the zebrafish heart were tracked and their motion and interaction were quantified using the customized VR platform. The velocity and relative distance changes of selected cells were also compared during one cardiac cycle in the atrium and ventricle to assess regional contractility and investigate local strain (Figure 4). The regional strain was determined based on the variation in displacement between two adjacent myocardial cells in the region of interest. The fractional shortening (FS) and ejection fraction (EF) were estimated using methodologies described in the published literature25. The equations for FS and EF are as follows:

Equation 1

Equation 2

where Dd and Ds are the ventricle diameters (short axis, measured by distances between different ventricle cells) at the end-diastolic and end-systolic stages, respectively, and Vd and Vs are the ventricular volumes (calculated by long and short axis diameters of ventricle) at the end-diastolic and end-systolic stages, respectively.

Collectively, these results showcase a new strategy that integrates both physiology and engineering to facilitate volumetric imaging and data interpretation in zebrafish models, holding great promise for advancing the exploration of cardiac morphogenesis and mechanics.

Figure 1
Figure 1: Zebrafish egg collection and microinjection process. The process involves mating adult zebrafish, collecting fertilized eggs, and performing the optional microinjection. During microinjection, the pre-pulled glass pipette is loaded with the desired materials. Fertilized eggs are aligned in straight rows within agarose mold slits and positioned perpendicular to the needle. The microinjection is conducted under constant microscopic observation of both the eggs and the needle tip. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Light-sheet imaging process and 4D image reconstruction. (A) Schematic of the in-house light-sheet imaging system structure. CL: cylindrical lens. EO: excitation objective. DO: detection objective. TL: tube lens. FL: filter. CAM: sCMOS camera. (B) Schematic of a zebrafish mounted in the FEP tube and embedded by agarose. (C) Raw 2D image sequence captured from a 4 dpf zebrafish larva. The clock on the upper left of each frame indicates the cardiac phase starting from end-systole. (D) Illustration of retrospective synchronization for 4D zebrafish image registration. Image sequence indicates a continuous recording at a certain depth along the z-axis. Each frame in the image sequence is represented by a red dot, and the starting and ending phases from end-diastole to end-systole are highlighted in the yellow box. (E) 4D reconstructed heart models from the 2D image sequences. Dpf: days post fertilization. This figure is adapted from Zhang et al.12. Please click here to view a larger version of this figure.

Figure 3
Figure 3: LabVIEW control panel. (A) The settings used in the control panel encompass the configuration of the laser, camera, image path, and motor pattern. (B) An example of a raw zebrafish heart image captured by the LabVIEW program. Scale bar: 30 µm. Please click here to view a larger version of this figure.

Figure 4
Figure 4: Cell tracking and VR interaction assist in revealing zebrafish cardiac contractility. (A) Tracked cells of Tg(myl7:nucGFP) zebrafish heart at 3 dpf. (B) The VR platform provides users with an immersive viewing and interactive experience of a 4D zebrafish heart model, allowing one to visualize heart function over time through user-defined analysis. (C) After collecting measurements from the VR platform, velocity and relative distance changes were compared during one cardiac cycle between selected cells at 3 dpf and 7 dpf. The first two figures present the average velocity changes in five ventricular cells and five atrial cells, and the others illustrate the relative distance changes between three groups of cells, i.e., two ventricular cells, a ventricular cell and an atrial cell, and two atrial cells. (D) Cardiac function assessments of 3 dpf and 7 dpf zebrafish hearts. A total of 370 cells were tracked in zebrafish hearts at 3 dpf, and 580 cells were tracked at 7 dpf. This figure is adapted from Zhang et al.12. Please click here to view a larger version of this figure.

NAME USAGE IDENTIFIER
Customized programs and algorithms
PSF.m To calculate PSF from FWHM measurements results of fluorescence beads. MATLAB program
4D Zebrafish Imaging.vi To control and synchronize the hardware in the LSM system. LabVIEW program
test_Parallel.m, etc. To reconstruct 4D images from 2D image sequences. MATLAB program
imageDimConverter.m To covert image data to a specified format for cell tracking. MATLAB program
separateTrackingResults.py To post process cell tracking results for separating cells with same labels. Python program
cellTracking_All.py To select cells with consistent image intensity across all volumes. Python program
cellLabelsToObj.ipynb To generate a surface mesh for each individual cell through 3D slicer. Python program
DynamicHeartModel.cs, etc. To create a VR environment and interact with objects.  C# program
Customized hardware design
3D-printed sample chamber To use with water-immersion objectives. SolidWorks design
3D-printed sample holder To adapt with the sample stage and hold the sample. SolidWorks design

Table 1: Customized resources table. The sample codes and data are uploaded to Zenodo21.

Discussion

The integration of the zebrafish model with engineering methods holds immense potential for the in vivo exploration of myocardial infarction, arrhythmias, and congenital heart defects. Leveraging its optical transparency, regenerative capacity, and genetic and physiological similarities to humans, zebrafish embryos and larvae have become extensively utilized in research1,2,4. The superior spatiotemporal resolution, minimal photodamage, and optical sectioning capabilities of light-sheet imaging set it apart for the 4D investigation of cardiac morphology and contractile function in zebrafish larvae12,26,27,28. This protocol introduces a comprehensive approach that integrates light-sheet imaging, retrospective synchronization, 3D cell tracking, and an interactive virtual reality (VR) platform for quantitative assessments. It provides an entry point to capture and reconstruct cardiac contraction, track and quantify the cellular dynamics, and assess the structural, functional, and mechanical changes during cardiac development and regeneration. This method could also be empowered by multi-channel fluorescence imaging to track various cell types and lineages for investigating cellular heterogeneity and intercellular interaction. While the zebrafish model's dichotomy from mammalian systems and its limitation as an acute model warrant careful consideration, advancements in genetic engineering may allow for the development of transgenic zebrafish lines that more closely resemble human disease states, thereby enhancing the translational value of findings derived from zebrafish studies29.

This retrospective synchronization based on the assumption of periodical cardiac cycle is used to reconstruct the light-sheet images of the contracting zebrafish heart. The implementation of this method in light-sheet imaging enables one to visualize and analyze the contracting heart at over 200 volumes per second. To address the challenge posed by the large volume of data in the current analysis, parallel computation with multi-core CPU and GPU was used, resulting in more than ten-fold improvement in image reconstruction efficiency. During the retrospective synchronization, it was presupposed that the cardiomyocytes return to their baseline positions with each heartbeat. While this assumption is relatively safe given the low variability in heartbeats seen in zebrafish26 as demonstrated in Figures 4C, it remains a simplification that may not account for all biological nuances. Innovations in imaging, such as compressive fluorescence and light-field microscopy, could mitigate effects stemming from this assumption by decreasing slice redundancy.

To enable the in-depth analysis in the intricate 4D heart model, a computational framework was developed including both 3DeeCellTracker-based cell segmentation and VR platform for user-directed investigations. The inherent capabilities of this framework, including the high efficiency and interactive manipulation, allows one to quantify the cellular velocity change, investigate the cell-cell interactions, and assess the regional and global myocardial mechanics such as fractional shortening, ejection fraction, and regional strain30 (Figure 4). The absence of prior assessments of zebrafish heart function at 7 dpf within the existing literature is noted. However, when compared to data at 3 dpf provided by other researchers, the present results are consistent with previous indications of a decline in both ejection fraction (EF) and fractional shortening (FS) during the interval from 3 dpf to 7 dpf25,31. These analytical methods possess significant potential for elucidating cardiac dynamics in zebrafish cardiac injury models12,25.

As VR technology and software contain unique features such as stereoscopic vision and automatic registration of cell positions, this platform allows for an immersive and streamlined method to select specific cells and analyze their trajectory throughout the cardiac cycle. The utilization of VR provides an intuitive way to perform complex tasks such as cell segmentation and annotation, with greater efficiency and accuracy32,33,34. It also supports multiple users to utilize unique tools when working together on large, complicated datasets34. Although prolonged using VR may induce side effects like dizziness, this can be mitigated by improved ergonomic designs and untethered headsets. Furthermore, to address the considerable computational power required for rendering complex biological structures in real-time, optimizing software for instantaneous processing and leveraging cloud-based computing solutions could enhance the system's ability to manage intricate visualizations.

With hardware devices and computation power continue advancing, this strategy can be extended to investigate cardiac arrhythmias in the myocardium at both cellular and tissue levels, ultimately holding the potential to unravel the underlying mechanism of cardiac morphogenesis and advancing the development of therapeutic interventions.

Declarações

The authors have nothing to disclose.

Acknowledgements

We express our gratitude to Dr. Caroline Burns at Boston Children's Hospital for generously sharing the transgenic zebrafish. We thank Ms. Elizabeth Ibanez for her help in husbanding zebrafish at UT Dallas. We also appreciate all the constructive comments provided by D-incubator members at UT Dallas. This work was supported by NIH R00HL148493 (Y.D.), R01HL162635 (Y.D.), and UT Dallas STARS program (Y.D.).

Materials

RESOURCE SOURCE/Reference IDENTIFIER
Animal models
Tg(myl7:nucGFP) transgenic zebrafish Burns Lab in Boston Children's Hospital ZDB-TGCONSTRCT-070117-49
Software and algorithms
MATLAB The MathWorks Inc. R2023a
LabVIEW National Instruments Corporation 2017 SP1
HCImage Live Hamamatsu Photonics 4.6.1.2
Python The Python Software Foundation 3.9.0
Fiji-ImageJ Schneider et al.18 1.54f
3DeeCellTracker Chentao Wen et al.15 v0.5.2
Unity Unity Software Inc. 2020.3.2f1
Amira Thermo Fisher Scientific 2021.2
3D Slicer Andriy Fedorov et al.17 5.2.1
ITK SNAP Paul A Yushkevich et al.16 4
Light-sheet system
Cylindrical lens Thorlabs ACY254-050-A
4X Illumination objective Nikon MRH00045
20X Detection objective Olympus 1-U2M585
sCMOS camera Hamamatsu C13440-20CU
Motorized XYZ stage Thorlabs PT3/M-Z8
Two-axis tilt stage Thorlabs GN2/M
Rotation stepper motor Pololu 1474
Fluorescent beads Spherotech FP-0556-2
473nm DPSS Laser Laserglow R471003GX
532nm DPSS laser Laserglow R531003FX
Microinjector and vacuum pump
Microinjector WPI PV850
Vacuum pump Welch 2522B-01
Pre-Pulled Glass Pipettes WPI TIP10LT
Capillary tip for gel loading Bio-Rad 2239912
Virtual reality hardware
VR headset Meta Quest 2
30mg/L PTU solution
PTU Sigma-Aldrich P7629
1X E3 working solution
1% Agarose
Low-melt agarose Thermo Fisher 16520050
Deionized water
10g/L Tricaine stock solution
Tricaine Syndel SYNC-M-GR-US02
Deionized water
Sodium bicarbonate Sigma-Aldrich S6014
150mg/L Tricaine working solution
10g/L Tricaine stock solution
Deionized water
60X E3 stock solution
Sodium Chloride Lab Animal Resource Center (LARC), The University of Texas at Dallas NaCl
Potassium Chloride KCL
Calcium Chloride Dihydrate CaCL2 x 2H2O
Magnesium Sulfate Heptahydrate MgSO4 x 7H2O
RO Water
1X E3 working solution
60X E3 stock solution Lab Animal Resource Center (LARC), The University of Texas at Dallas
RO Water
1% Methylene Blue (optional)  C16H18ClN3S

Referências

  1. Power, R. M., Huisken, J. A guide to light-sheet fluorescence microscopy for multiscale imaging. Nat. Methods. 14 (4), 360-373 (2017).
  2. Liu, J., Stainier, D. Y. R. Zebrafish in the study of early cardiac development. Circ. Res. 110 (6), 870 (2012).
  3. Ding, Y., Bu, H., Xu, X. Modeling inherited cardiomyopathies in adult zebrafish for precision medicine. Front. Physiol. 11, 599244 (2020).
  4. Giardoglou, P., Beis, D. On zebrafish disease models and matters of the heart. Biomedicines. 7 (1), 15 (2019).
  5. Zhang, X., Alexander, R. V., Yuan, J., Ding, Y. Computational analysis of cardiac contractile function. Curr. Cardiol. Rep. 24 (12), 1983-1994 (2022).
  6. Weber, M., Huisken, J. In vivo imaging of cardiac development and function in zebrafish using light sheet microscopy. Swiss Med. Wkly. 145 (51), w14227 (2015).
  7. Vedula, V., et al. A method to quantify mechanobiologic forces during zebrafish cardiac development using 4-D light sheet imaging and computational modeling. PLOS Comput. Biol. 13 (10), e1005828 (2017).
  8. Yalcin, H. C., Amindari, A., Butcher, J. T., Althani, A., Yacoub, M. Heart function and hemodynamic analysis for zebrafish embryos. Dev. Dyn. 246 (11), 868-880 (2017).
  9. Baek, K. I., et al. Advanced microscopy to elucidate cardiovascular injury and regeneration: 4D light-sheet imaging. Prog. Biophys. Mol. Biol. 138, 105-115 (2018).
  10. Salman, H. E., Yalcin, H. C. Advanced blood flow assessment in Zebrafish via experimental digital particle image velocimetry and computational fluid dynamics modeling. Micron. 130, 102801 (2020).
  11. Sodimu, O., et al. Light sheet imaging and interactive analysis of the cardiac structure in neonatal mice. J. Biophotonics. 16 (5), e202200278 (2023).
  12. Zhang, X., et al. 4D light-sheet imaging and interactive analysis of cardiac contractility in zebrafish larvae. APL Bioeng. 7 (2), 026112 (2023).
  13. Westerfield, M. The Zebrafish Book. Eugene. , (2000).
  14. Rosen, J. N., Sweeney, M. F., Mably, J. D. Microinjection of zebrafish embryos to analyze gene function. J Vis Exp. 25, e1115 (2009).
  15. Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B., Schilling, T. F. Stages of embryonic development of the zebrafish. Dev Dyn. 203 (3), 253-310 (1995).
  16. Girkin, J. M., Carvalho, M. T. The light-sheet microscopy revolution. J. Opt. 20 (5), 053002 (2018).
  17. Gualda, E. J., et al. OpenSpinMicroscopy: an open-source integrated microscopy platform. Nat. Methods. 10 (7), 599-600 (2013).
  18. Messerschmidt, V., et al. Light-sheet fluorescence microscopy to capture 4-dimensional images of the effects of modulating shear stress on the developing zebrafish heart. J Vis Exp. 138, e57763 (2018).
  19. Kaufmann, A., Mickoleit, M., Weber, M., Huisken, J. Multilayer mounting enables long-term imaging of zebrafish development in a light sheet microscope. Development. 139 (17), 3242-3247 (2012).
  20. Lee, J., et al. 4-Dimensional light-sheet microscopy to elucidate shear stress modulation of cardiac trabeculation. J Clin Invest. 126 (5), 1679-1690 (2016).
  21. Zhang, X., et al. 4D light-sheet imaging and interactive analysis of cardiac contractility in Zebrafish larvae. Zenodo. , (2023).
  22. Wen, C., et al. 3deecelltracker, a deep learning-based pipeline for segmenting and tracking cells in 3d time lapse images. Elife. 10, e59187 (2021).
  23. Yushkevich, P. A., et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage. 31 (3), 1116-1128 (2006).
  24. Fedorov, A., et al. 3D Slicer as an image computing platform for the quantitative imaging network. Magn. Reson. Imaging. 30 (9), 1323-1341 (2012).
  25. Naderi, A. M., et al. Deep learning-based framework for cardiac function assessment in embryonic zebrafish from heart beating videos. Comput. Biol. Med. 135, 104565 (2021).
  26. Mickoleit, M., et al. High-resolution reconstruction of the beating zebrafish heart. Nat. Methods. 11 (9), 919-922 (2014).
  27. Lee, J., et al. 4-Dimensional light-sheet microscopy to elucidate shear stress modulation of cardiac trabeculation. J. Clin. Invest. 126 (5), 1679-1690 (2016).
  28. Ding, Y., et al. Multiscale light-sheet for rapid imaging of cardiopulmonary system. JCI Insight. 3 (16), e121396 (2018).
  29. Choe, C. P., et al. Transgenic fluorescent zebrafish lines that have revolutionized biomedical research. Lab. Anim. Res. 37 (1), 1-29 (2021).
  30. Coelho-Filho, O. R., et al. Quantification of cardiomyocyte hypertrophy by cardiac magnetic resonance: implications on early cardiac remodeling. Circulation. 128 (11), 1225 (2013).
  31. Zhang, B., et al. Automatic segmentation and cardiac mechanics analysis of evolving zebrafish using deep learning. Front. Cardiovasc. Med. 8, 675291 (2021).
  32. Ding, Y., et al. Integrating light-sheet imaging with virtual reality to recapitulate developmental cardiac mechanics. JCI Insight. 2 (22), e97180 (2017).
  33. Koger, C. R., Hassan, S. S., Yuan, J., Ding, Y. Virtual reality for interactive medical analysis. Front. Virtual Real. 3, 782854 (2022).
  34. Yuan, J., et al. Extended reality for biomedicine. Nat. Rev. Methods Prim. 3 (1), 1-1 (2023).

Play Video

Citar este artigo
Zhang, X., Saberigarakani, A., Almasian, M., Hassan, S., Nekkanti, M., Ding, Y. 4D Light-sheet Imaging of Zebrafish Cardiac Contraction. J. Vis. Exp. (203), e66263, doi:10.3791/66263 (2024).

View Video