A Versatile Pipeline for Analyzing Dynamic Changes in Nuclear Bodies in a Variety of Cell Types

Summary

This method describes an immunofluorescence protocol and quantification pipeline for evaluating protein distribution with varied nuclear organization patterns in human T lymphocytes. This protocol provides step-by-step guidance, starting from sample preparation and continuing through the execution of semi-automated analysis in Fiji, concluding with data handling by a Google Colab notebook.

Abstract

Various nuclear processes, such as transcriptional control, occur within discrete structures known as foci that are discernable through the immunofluorescence technique. Investigating the dynamics of these foci under diverse cellular conditions via microscopy yields valuable insights into the molecular mechanisms governing cellular identity and functions. However, performing immunofluorescence assays across different cell types and assessing alterations in the assembly, diffusion, and distribution of these foci present numerous challenges. These challenges encompass complexities in sample preparation, determination of parameters for analyzing imaging data, and management of substantial data volumes. Moreover, existing imaging workflows are often tailored for proficient users, thereby limiting accessibility to a broader audience.

In this study, we introduce an optimized immunofluorescence protocol tailored for investigating nuclear proteins in different human primary T cell types that can be customized for any protein of interest and cell type. Furthermore, we present a method for unbiasedly quantifying protein staining, whether they form distinct foci or exhibit a diffuse nuclear distribution.

Our proposed method offers a comprehensive guide, from cellular staining to analysis, leveraging a semi-automated pipeline developed in Jython and executable in Fiji. Furthermore, we provide a user-friendly Python script to streamline data management, publicly accessible on a Google Colab notebook. Our approach has demonstrated efficacy in yielding highly informative immunofluorescence analyses for proteins with diverse patterns of nuclear organization across different contexts.

Introduction

The organization of the eukaryotic genome is governed by multiple layers of epigenetic modifications¹, coordinating several nuclear functions that can occur within specialized compartments called nuclear bodies or condensates². Within these structures, processes such as transcription initiation³, RNA processing^4,5,6, DNA repair^7,8, ribosome biogenesis^9,10,11, and heterochromatin regulation^12,13 take place. The regulation of nuclear bodies adjusts over both spatial and temporal dimensions to accommodate cellular requirements, guided by principles of phase separation^14,15. Consequently, these bodies function as transient factories where functional components assemble and disassemble, undergoing changes in size and spatial distribution. Hence, understanding the characteristics of nuclear proteins by microscopy, including their propensity to form bodies and their spatial arrangement in different cellular conditions, offers valuable insights into their functional roles. Fluorescence microscopy is a widely used method for studying nuclear proteins, allowing their detection through fluorescent antibodies or directly expressing targets with a fluorescent protein reporter^16,17.

In this context, nuclear bodies appear as bright foci or puncta, with a notable degree of sphericity, making them easily distinguishable from the surrounding environment^16,18. Super-resolution techniques like STORM and PALM, by providing improved resolution (up to 10 nm)¹⁹, enable more precise characterization of the structure and composition of specific condensates²⁰. However, their accessibility is limited by equipment expenses and the specialized skills needed for data analysis. Therefore, confocal microscopy remains popular due to its favorable balance between resolution and wider usage. Such popularity is facilitated by the inherent removal of out-of-focus light, which diminishes the requirement for extensive post-processing procedures for accurate segmentation, its widespread availability in research institutes, its effective acquisition time, and sample preparation that is typically efficient. However, accurately measuring protein distribution, assembly, or diffusion using immunofluorescence assays across diverse cellular conditions poses challenges, as many existing methods lack guidance on selecting suitable parameters for proteins with varying distribution patterns²¹. Moreover, handling the resulting large data volume can be daunting for users with limited experience in data analysis, potentially compromising the biological significance of the results.

To address these challenges, we introduce a detailed step-by-step protocol for immunofluorescence preparation and data analysis, aiming to provide an unbiased method for quantifying protein staining with various organization patterns (Figure 1). This semi-automated pipeline is designed for users with limited expertise in computational and imaging analysis. It combines the functionalities of two established Fiji plugins: FindFoci²² and 3D suite²³. By integrating the precise foci identification capability of FindFoci with the object identification and segmentation features in 3D space offered by 3D suite, our approach generates two CSV files per channel for each field of acquisition. These files contain complementary information that facilitates the calculation of metrics suitable for various types of signal distribution, such as the count of foci per cell, the distance of foci from the nuclear centroid, and the inhomogeneity coefficient (IC), which we have introduced for diffuse protein staining. In addition, we acknowledge that data extrapolation can be time-consuming for users with limited data handling skills. To streamline this process, we provide a Python script that automatically compiles all collected measurements into a single file for each experiment. Users can execute this script without the need to install any programming language software. We provide an executable code on Google Colab, a cloud-based platform that allows the writing of Python scripts directly in the browser. This ensures that our method is intuitive and readily accessible for immediate use.

We demonstrate the effectiveness of our protocol in analyzing and quantifying alterations in signal distribution of two nuclear proteins: Bromodomain-containing protein 4 (BRD4) and Suppressor of zeste-12 (SUZ12). BRD4 is a well-documented coactivator protein within the Mediator complex known to form condensates associated with polymerase II-dependent transcriptional initiation^24,25. SUZ12 is a protein component of the Polycomb Repressive Complex 2 (PRC2) responsible for regulating the deposition of H3K27me3 histone modification^26,27. These proteins exhibit different patterns within two distinct cell types: freshly isolated human CD4⁺ naïve T cells, which are quiescent and exhibit slow rates of transcriptional activity, and in vitro differentiated T_H1 CD4⁺ cells, which are specialized, proliferating effector cells showing increased transcription²⁸.

Protocol

The use of human samples for research purposes was approved by the Ethics Committees of the Fondazione Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Cà Granda Ospedale Maggiore Policlinico (Milan), and informed consent was obtained from all subjects (authorization numbers: 708_2020). The protocol is organized into three primary sections: immunofluorescence execution, image acquisition, and image analysis. On average, it necessitates 4 working days to be completed (Figure 1).

1. Immunofluorescence preparation

NOTE: This immunofluorescence protocol can be easily customized for various cell types and protein targets by adjusting the fixation and permeabilization conditions. Immunofluorescence preparation typically takes less than 3 days to be completed, with the duration of primary antibody incubation varying based on antibody quality and the target protein (Figure 1).

1. Sample preparation
   1. Isolate human peripheral blood mononuclear cells (PBMCs) through a density gradient centrifugation via a medium of approximately 1.077 g/mL density according to the manufacturer's instructions (Table of Materials).
   2. Isolate CD4⁺ T cells from PBMCs with magnetic beads, following the manufacturer's instructions (Table of Materials).
   3. Stain cells with antibodies for CD4, CD45RA, and CD45RO (Table of Materials) and proceed with FACS-sorting of naïve CD4⁺ T cells as CD4⁺/CD45RA⁺/CD45RO^- cells, as described elsewhere^29,30.
      NOTE: See the Table of Materials for specifications of the FACS sorter used in this protocol.
   4. Induce the differentiation of naïve CD4⁺ T cells into T helper 1 (T_H1 CD4⁺ cells) as described in ²⁹. Briefly, culture 1.5 x 10⁶ cells/mL of FACS-sorted naïve CD4⁺ T cells in T_H1 medium stimulating them with anti-CD3/anti-CD28 magnetic beads at a 1:1 ratio. Count the cells and split them when they reach 1.5 × 10⁶ cells/mL every 2-3 days (see Table 1 for T_H1 medium composition).
   5. Evaluate the secretion of effector function cytokines after 7 days of differentiation, as described in ²⁹.
2. Cell fixation and permeabilization
   NOTE: All these procedures were described in ^30,31 with minor adaptations.
   1. To ensure optimal cell adhesion, treat glass coverslips (10 mm, thickness 1.5 H) with coating solution (Table 1) as follows:
      1. Clean the glass coverslips by initially washing them with distilled water (ddH₂O), followed by a rinse with 70% ethanol (EtOH), and allowing them to air dry.
      2. Place the washed coverslips in a 24-multiwell plate for steps 1.2.1.3-1.3.
      3. Apply a drop of 200 µL of coating solution onto the glass coverslip. After 5 min, remove the drop and let it air dry.
      4. Wash the coverslips by applying a drop of 200 µL of ddH₂O. After 5 min, remove the drop and air dry.
      5. Repeat 3 x steps 1.2.1.3-1.2.1.4.
   2. Resuspend naïve CD4⁺ T cells and T_H1 CD4⁺ cells in 1x phosphate-buffered saline (PBS) at 2 × 10⁶ cells/mL concentration.
      NOTE: The indicated concentration is advised specifically for small cells such as human primary T lymphocytes. For adherent cells, glass coating treatment is unnecessary. Instead, proceed directly with growing the cells on the glass surface using the appropriate cell culture medium.
   3. Apply a 200 µL drop of the cell suspension onto the glass cover slip. Allow the cells to seed at room temperature (RT) for 30 min; then, remove the drop.
   4. Fix the cells with freshly filtered 3% paraformaldehyde (PFA solution, Table 1) for 10 min at RT.
   5. Wash the glass coverslip with TPBS (Table 1) for 3 x 5 min at RT.
   6. Remove TPBS and add permeabilization solution (Table 1) for 10 min at RT.
   7. Discard the permeabilization solution and incubate the sample in storage solution (Table 1) from 1 h to overnight (ON) at 4 °C.
      NOTE: At this stage, the protocol can be safely halted, and the glass cover slips can be preserved in storage solution in a 24-multiwell plate for 3-4 weeks.
3. Immunofluorescence
   1. (Optional) Remove the coverslip from the 24-multiwell plate and rapidly freeze on dry ice for 30 s, thaw at RT, and then wash the glass coverslip in a prefilled well with storage solution.
   2. (Optional) Repeat 3 x step 1.3.1.
   3. Wash in permeabilization solution for 5 min at RT. Next, wash for 2 x 5 min with TPBS at RT.
   4. Incubate in 0.1 N HCl for 12 min at RT.
   5. Perform two quick washes in 1x PBS.
      NOTE: The specified cell permeabilization conditions, including freeze and thaw steps and HCl treatment, are optimal for staining nuclear components in cells characterized by densely packed chromatin. However, when dealing with cells characterized by less compacted chromatin, it is advisable to reduce or avoid these steps. Additionally, for targeting cytoplasmic components, consider reducing or eliminating HCl treatment.
   6. Incubate the cells with primary antibody (BRD4, 1:500, or SUZ12, 1:100) diluted in antibody dilution buffer (Table 1) (200 µL for each glass coverslip) ON at 4 °C.
      NOTE: Immunofluorescence can be performed by multiplexing more than one primary antibody depending on experimental needs, secondary antibodies, and detection systems.
   7. Wash 3 x 5 min with PBS-T (Table 1) at RT with mild shaking.
   8. Incubate the cells with secondary antibody diluted in antibody dilution buffer (200 µL for each glass coverslip) for 1 h at RT.
      NOTE: Choose the secondary antibody that best suits the experimental needs and the available detection systems. Ensure that each secondary antibody targets the species from which the primary antibody is derived. Additionally, select fluorophores that are compatible with the microscope's filters and light source. It is crucial to avoid spectral overlap between the chosen fluorophore and the emission spectrum of the nuclear stain.
   9. Wash 3 x 5 min with PBS-T at RT with mild shaking.
   10. Stain with 1 ng/mL of 4',6-diamidino-2-phenylindole (DAPI) diluted in 1x PBS for 5 min at RT.
       NOTE: Alternative dyes can be used for nuclear staining as long as they do not overlap with the emission spectrum of secondary antibodies.
   11. Perform several quick washes in 1x PBS.
   12. Mount the glass coverslip on a microscopy slide with antifade mounting media.

2. Image acquisition

NOTE: The duration of image acquisition depends on the instrument and selected settings.

1. Confocal microscope acquisition
   1. Capture 3D images using a confocal microscope, setting a step size of 0.25 µm in z and pixel size of 0.1-0.2 µm.
      NOTE: To achieve optimal light diffraction-limited resolution with our 63x 1.4 NA Oil objective, we set the pinhole size at 0.8 AU, 2x line average, and frame size 1024 × 1024 pixels. The excitation laser as well as the channel acquisition sequence were selected deliberately to prevent interference or crosstalk between the fluorophores being used. However, the adjustment of parameters should be tailored to the specific microscope and specimen characteristics. See the Table of Materials for specifications of the confocal microscope used in this protocol.
   2. Acquire a consistent number of random fields to encompass approximately 50 cells per biological replicate.

3. Image analysis

1. Software installation
   1. Download and install the last available version of Fiji from the official Fiji Download page (https://imagej.net/software/fiji/downloads).
   2. Install 3D Suite either via the Fiji update site or manually following the instructions on the 3D Suite website (https://mcib3d.frama.io/3d-suite-imagej/#download).
   3. Install GDSC (FindFoci) either via the Fiji update site or manually by following the instructions on the GitHub repository (https://github.com/aherbert/gdsc). 
2. TIFF conversion
   1. Download the "convert_to_TIFF.py" script (Supplemental File 1).
   2. Drag and drop the script on Fiji and run the code.
   3. In the panel that appears, browse to the path where the experiment is stored. The subfolder containing the converted TIFF files is created in the same experiment folder.
3. Initialize settings on 3D Manager.
   1. Open the 3D Manager option panel by clicking on Plugins | 3DSuite | 3D Manager Options.
   2. Within the 3D Manager option window, select the checkboxes corresponding to the following measurements: Volume (unit), Mean Gray Value, Bounding box (pix), Std Dev Gray Value, Centroid (pix), and Centroid (unit).
   3. Tick the following options: Exclude objects on edges XY and Exclude objects on edges Z and click OK.
      NOTE: This step needs to be done only once to initially configure the metrics that will be stored in the spatial and quantitative information files. The selected metrics are essential for ensuring proper pipeline functioning. Optional additional metrics can be included in this step, as further described in the Discussion section.
4. Set parameters on FindFoci GUI.
   NOTE: This step needs to be done only once to configure the semi-automated pipeline with optimal parameters, which will then be incorporated into the scripts.
   1. Open the test image. Duplicate the protein channel, including the stack (Ctrl + Shift + D), check the hyperstack checkbox, and specify the appropriate channel number within the Channels (c) box, and rename it accordingly.
   2. Launch the Fiji macro recorder by navigating to Plugins | Macros | Record.
   3. Open the FindFoci plugin (click on Plugins | GDSC | FindFoci | FindFoci GUI) and select the image to be analyzed from the "Image" dropdown menu.
   4. Set the parameters as follows (Figure 2A): Gaussian blur = 1.5; background method = SD above mean; background param = 9; search method = fraction of peak - background; search param = 0.7; peak method = relative above background; peak param = 0.2; minimum size = 5; max peaks = 1,000,000.
      NOTE: The indicated parameters have been selected based on our case studies and may not be suitable for other staining procedures.
   5. To enhance foci identification, adjust the following parameters:
      1. Gaussian blur defines the extent of smoothening to better segment foci. Keep it close to the foci diameter (pixel).
      2. Background param sets a threshold to distinguish the background from the foci signal. Increase the values to impose more stringent thresholds.
      3. Search param defines the percentage of fluorescence from the peak that is included in signal recognition. Decrease the values to include areas farther away from the fluorescence peak.
      4. Peak param determines the degree to which two signal peaks are considered continuous or separated. Decreasing the value will result in the separation of peaks.
      5. Max peaks specifies the maximum number of foci identifiable. Set high numbers to include all foci in the image.
   6. Run FindFoci and copy the string that appears in the recorder window (Step 3.4.2, Figure 2B) that contains the selected parameters, excluding the quotation marks. For further information related to the settings, refer to the plugin manual instruction ²².
5. Nuclear protein quantification pipeline
   1. Download the script "nuclear_prot_q.py" (Supplemental File 2).
   2. Drag and drop the script on Fiji and click run to execute the code.
   3. Follow the instructions within the displayed dialog box to process the images.
      1. Nucleus channel: enter the number corresponding to the channel of DAPI (or any nuclear staining).
      2. Nucleus Gaussian Blur: enter the value of sigma needed to blur the image for segmentation.
         NOTE: Maintain this parameter closer to the nucleus diameter (i.e., 5-6 µm). Higher sigma values are indicated for inhomogeneous staining.
      3. Protein channel: enter the number corresponding to the channel of the staining of interest.
      4. FindFoci Parameter: paste the string obtained from the macro recording step in passage 3.4.6 (Figure 2B).
      5. (Optional) Quality control: check the quality of the segmented nuclei. This will pause the script and allow manual checking of each generated nuclear region of interest (ROI).
      6. Select an image directory: click the Browse button to navigate to the folder containing the TIFF files to be analyzed.
   4. Once all boxes are compiled, click OK to continue the execution.
   5. If a nucleus is not correctly segmented and fails to meet quality requirements, delete or modify it as indicated in Figure 2C.
      1. Check the ROI list in the ROIManager3D window; if the list is empty, select the merge window and click Quantify 3D to refresh the manager. Then, close the Quantify 3D result table. Select the nuclei channel and click Live-ROI to ON.
      2. Select the ROI belonging to the same nucleus and press Merge or press Delete in the case of undesired nuclei.
      3. Click select all and proceed with the analysis by clicking OK in the Mask check window.
      4. For results, look in the Quantification folder within the file path indicated in step 3.5.3.6, containing a txt file with records of the parameter used for the analysis.
6. Pipeline on Google Colab
   1. Download the "final_nuclear_protein_metrics.ipynb" notebook (Supplemental File 3).
   2. Open the notebook on Google Colab (https://colab.research.google.com/).
   3. Upload all the folders containing the .csv files of each image field into a folder of preference in Google Drive.
   4. Indicate in the notebook the path of the folder where the result subfolders are stored and run all cells. When the code has finished compiling, the final spreadsheet file containing all the compiled data is created in the same folder where the .csv files were uploaded.

Results

The outlined protocol in this method facilitates the visualization and quantification of alterations in nuclear protein staining within human primary T cells, and it can be customized for diverse cell types and protein targets. As case studies, we conducted and analyzed the staining of BRD4 and SUZ12 in naïve and T_H1 CD4⁺ cells.

BRD4 displays a well-dotted staining pattern in both quiescent naïve and differentiated T_H1 CD4⁺...

Discussion

In this study, we present a method for performing immunofluorescence experiments on nuclear proteins in human T lymphocytes. This method offers flexibility for use with various cell types through minor modifications in fixation and permeabilization steps, as described previously^30,31.

Our imaging workflow builds upon established techniques outlined in the literature, specifically FindFoci and 3D Suite^22,

Materials

Name	Company	Catalog Number	Comments
1.5 mL Safe-Lock Tubes	Eppendord	#0030121503	Protocol section 1
10 mL Serological pipettes	VWR	#612-3700	Protocol section 1
20 µL barrier pipette tip	Thermo Scientific	#2149P-HR	Protocol section 1
50 mL Polypropylene Conical Tube	Falcon	#352070	Protocol section 1
200 µL barrier pipette tip	Thermo Scientific	#2069-HR	Protocol section 1
antifade solution - ProlongGlass - mountingmedia	Invitrogen	#P36984	Step 1.3.12
BSA (Bovine Serum Albumin)	Sigma	#A7030	Step 1.3.6., 1.3.8.
CD4+ T Cell Isolation Kit	Miltenyi Biotec	#130-096-533	Step 1.1.2.
DAPI (4,6-diamidino-2-phenylindole)	Invitrogen	Cat#D1306	Step 1.3.10.
Dry ice			Step 1.3.1.
Dynabeads Human T-activator anti-CD3/anti-CD28 bead	Life Technologies	#1131D	magnetic beads step 1.1.4.
EtOH	Carlo Erba	#4146320	Step 1.2.1.1.
FACSAria SORP	BD Bioscences		Step 1.1.3. Equipped with BD FACSDiva Software version 8.0.3
FBS (Fetal Bovine Serum)	Life Technologies	#10270106	Step 1.1.4
FICOLL PAQUE PLUS	Euroclone	GEH17144003F32	Step 1.1.1.
FIJI Version 2.14.0	-	-	Protocol section 3
Glass coverslip (10 mm, thickness 1.5 H)	Electron Microscopy Sciences	#72298-13	Step 1.2.1.
Glycerol	Sigma	#G5516	Step 1.2.7-1.3.1.
Goat anti-Rabbit AF568 secondary antibody	Invitrogen	A11036	Step 1.3.8.
HCl	Sigma	#320331	Step 1.3.4.
human neutralizing anti-IL-4	Miltenyi Biotec	Cat#130-095-753	Step 1.1.4.
human recombinant IL-12	Miltenyi Biotec	Cat#130-096-704	Step 1.1.4.
human recombinant IL-2	Miltenyi Biotec	Cat#130-097-744	Step 1.1.4.
Leica TCS SP5 Confocal microscope	Leica Microsystems	-	Protocol section 2, Equipped with HCX PL APO 63x, 1.40 NA oil immersion objective, with an additional 3x zoom. Pinhole size : 0.8 AU. Line average 2×. Frame size 1024×1024 pixel.
MEM Non-Essential Amino Acids Solution	Life Technologies	#11140035	Step 1.1.4.
Microscope Slides	VWR	#631-1552	Step 1.3.12.
Mouse monoclonal anti-Human CD4 APC-Cy7 (RPA-T4 clone)	BD Bioscience	#557871	Step 1.1.3.
Mouse monoclonal anti-Human CD45RA PECy5 (5H9 clone)	BD Bioscience	#552888	Step 1.1.3.
Mouse monoclonal anti-Human CD45RO APC (UCHL1 clone)	Miltenyi Biotec	#130-113-546	Step 1.1.3.
Multiwell 24 well	Falcon	#353047	Protocol section 1
Normal Goat Serum	Invitrogen	PCN5000	Step 1.3.6., 1.3.8.
PBS	Life Technologies	#14190094	Protocol section 1
Penicillin/Streptomycin solution	Life Technologies	#15070063	Step 1.1.4.
PFA	Sigma	#P6148	Step 1.2.4.
poly-L-lysine	Sigma	#P8920	1.2.1.
Primary antibody - BRD4	Abcam	#ab128874	Step 1.3.6.
Primary antibody - SUZ12	Cel Signalling	mAb #3737	Step 1.3.6.
RPMI 1640 W/GLUTAMAX-I	Life Technologies	#61870010	Step 1.1.4.
Sodium Pyruvate	Life Technologies	#11360039	Step 1.1.4.
Triton X-100	Sigma	#T8787	Step 1.2., 1.3.
TWEEN 20	Sigma	#P9416	Step 1.3.
Tweezers	-	-	Protocol section 1