Jul 27, 2023
Sample preparation and imaging for large scale 3D spectral confocal imaging of tissues
- Tarek M. El-Achkar1,
- Michael Ferkowicz1
- 1Indiana University School of Medicine, Indianapolis, Indiana
Protocol Citation: Tarek M. El-Achkar, Michael Ferkowicz 2023. Sample preparation and imaging for large scale 3D spectral confocal imaging of tissues. protocols.io https://dx.doi.org/10.17504/protocols.io.bp2l6x1prlqe/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: July 27, 2023
Last Modified: July 27, 2023
Protocol Integer ID: 85596
Abstract
Frozen participant samples are first sized and preserved for high-resolution 3D imaging. Preserved biopsy sections are imaged, without staining, to understand the sample’s structure and condition. Next, the biopsy sections are stained with antibodies and fluorescent dyes to identify specific structures and cells with pathological significance. This stained sample is imaged in its entirety giving a 3D image. Finally, the 3D image is analyzed by experts to assess known and novel pathologies.
This protocol summarizes both label-free imaging and labeling for fluorescence imaging including the labeling, mounting and imaging approaches. The fluorescent labels used herein include conjugated or fluorescent small molecules and conjugated antibodies. Protocol 2 uses both directly conjugated primary antibodies for identifying specific antigens and indirect labeling with a fluorophore conjugated secondary antibody that recognizes the primary antibody.
Multiple rounds of imaging are enabled by the use of a non-hardening mounting media, low charge slides and removable rubber cement as a sealant. Details are described below.
Two imaging approaches are outlined: protocol 1 and 2. Protocol 1 involves label free imaging of auto-fluorescent species and second harmonic generation (SHG) with multiphoton excitation. These modalities are followed by protocol 2, which involves staining with fluorescent small molecules and immuno-fluorescence and subsequent large scale 3D confocal fluorescence imaging. There are three different staining strategies used in protocols, summarized in Table 1.
These protocols start with fixed 50 µm tissue sections prepared as described in the Cryopreservation protocol.
Version 3.0
| A | B | C | D | E | F | G | |
| Structure/cell type | Target | Species | Vendor (cat #) | Dilution [stock] | Direct labeling | Indirect labeling | |
| nuclei | DAPI | - | ThermoFisher (D1306) | 1:200 [1 mg/mL] | - | - | |
| vasculature | Phalloidin | - | ThermoFisher (O7466) | 1:200 [2 U/mL] | Oregon Green | - | |
| glomeruli | " | - | " | " | " | - | |
| brush borders | " | - | " | " | " | - | |
| TAL | THP | Sheep | R&D Systems (AF5144) | 1:1000 [0.2 mg/mL] | Alexa 546 | - | |
| Proximal tubules | AQP1 | Goat | Santa Cruz (sc-9878) | 1:50 [0.2 mg/mL] | - | donkey anti-goat Alexa 568 (1:200) | |
| Neutrophils | MPO | Rabbit | Abcam (ab9535) | 1:50 | - | donkey anti-rabbit DyLight 594 (1:200) | |
| macrophages | CD68 | Mouse | Dako (M0876) | 1:50 [0.3 mg/mL] | - | donkey anti-mouse Alexa 633 (1:200) | |
| T-cells | CD3 | Mouse | BD Pharmingen (557706) | 1:50 | Alexa 647 | - | |
| vasculature | CD31 | Mouse | NovusBio (NB600-562ss) | 1:50 [1 mg/mL] | Alexa 660 | - |
this panel was initiated for KPMP 2.0 (first stains October 17, 2022)
Version 4.0
| A | B | C | D | E | F | G | |
| Structure/cell type | Target | Species | Vendor (cat #) | Dilution [stock] | Direct labeling | Indirect labeling | |
| nuclei | DAPI | - | ThermoFisher (D1306) | 1:200 [1 mg/mL] | - | - | |
| vasculature | Phalloidin | - | ThermoFisher O7466 | 1:200 [2 U/mL] | Oregon Green | - | |
| glomeruli | " | - | " | " | " | - | |
| brush borders | " | - | " | " | " | - | |
| TAL | THP | Sheep | R&D Systems (AF5144) | 1:1000 [0.2 mg/mL] | Alexa 546 | - | |
| proximal tubules | AQP1 | Goat | Santa Cruz (sc-9878) | 1:50 [0.2 mg/mL] | - | donkey anti-goat Alexa 568 (1:200) | |
| neutrophils | MPO | Rabbit | Abcam (ab9535) | 1:50 | - | donkey anti-rabbit DyLight 594 (1:200) | |
| macrophages | CD68 | Mouse | Dako (M0876) | 1:50 [0.3 mg/mL] | - | donkey anti-mouse Alexa 633 (1:200) | |
| hematopoietic cells | CD45 | Mouse | Biolegend (304056) | 1:50 [0.5 mg/mL] | Alexa 647 | - | |
| vasculature | CD31 | Mouse | NovusBio (NB600-562ss) | 1:50 [1 mg/mL] | Alexa 660 | - |
This panel was initiated on May 24, 2023.
Guidelines
Excitation and filter settings
Label-free imaging was performed with femtosecond pulsed 910 nm excitation and de-scanned emissions were collected with custom band-passes of 435-465 nm for collecting SHG and 500- 550 for collecting auto-fluorescence. Spectral confocal imaging was performed with sequential scanning at each z frame for 4 excitations with 4 band-passes each giving 16 total channels. The configuration and scan settings are given in Table 2.
Protocol Quality Control metrics
1. Antibody validation: All primary and secondary antibodies are validated either by CLIA or GUDMAP citation, or by internal validation.
2. Negative controls: When tissue is available, secondary only controls will be included.
3. Tissue qualification:
a.A low resolution image of a 12 µm section of the entire biopsy labeled with PAS is collected (dissecting microscope, 6 µm resolution) and evaluated for tissue morphology and sample integrity.
b.The entire biopsy is imaged in 3D prior to labeling using multiphoton excitation to collect second harmonic and autofluorescence images (910 nm excitation, 430-460 nm and 460-660 nm emissions, respectively). Second harmonic images are used to identify/quantify the degree of fibrosis and along with tissue autofluorescence, to characterize tissue structure.
c.The entire biopsy is imaged in 2D using confocal microscopy prior to labeling.(A mosaic of confocal images of a single focal plane collected from the middle of the tissue section, using the 16 channel protocol used to image labeled tissue).These images are used to characterize levels of background tissue autofluorescence in each channel to identify potentially problematic regions and to inform the process of spectral deconvolution of the labeled tissue.
4. Microscope image collection: Prior to collection of labeled tissues, the performance of the microscope system is verified by quantitative analysis of images collected from a slide containing with 0.1 µm and 4 µm TetraSpeck fluorescent microspheres (Blue – 360 nm excitation, 430 nm emission, Green – 505 nm excitation, 515 nm emission, Orange – 560 nm excitation, 580 nm emission and Dark Red – 660 nm excitation, 680 nm emission).By exciting the fluorescence of the microspheres with each of the four laser lines, and collecting the fluorescence in the four detectors, this procedure provides a sensitive test of the integrated performance of the microscope. Performance is quantified as signal-to-noise in the images of each of the four fluorophores in the 4 µm beads. Departures in signal-to-noise or resolution exceeding 1 standard deviation of the previously established values would be grounds to delay image collection until the microscope deficiency has been remediated.
Table 4. QC metrics scan settings
| Metric | Excitation (nm) | Laser power (%) | Detector | Voltage | Gain (%) | Offset | Bandpass (nm) | Line average | Line summation | |
| PSF | 488 | 3 | PMT2 | 690 | NA | -0.18 | 493-600 | 4 | 1 | |
| SNR | 405 | 10 | HyD1 | NA | 74 | NA | 410-483 | 1 | 1 | |
| SNR | 488 | 0.5 | PMT2 | 650 | NA | -0.18 | 493-575 | 1 | 1 | |
| SNR | 552 | 4 | HyD3 | NA | 50 | NA | 575-633 | 1 | 1 | |
| SNR | 638 | 0.2 | PMT4 | 800 | NA | -0.18 | 643-776 | 1 | 1 |
5. Digital image analysis:Digital image analysis is used to (1) distinguish 8 fluorescent probes in the 16-channel confocal image volumes (spectral deconvolution), (2) segment individual nuclei in the resulting image volumes and (3) quantify the amount of fluorescence associated with each nucleus in each of the 8 channels.
a. Spectral deconvolution is validated by analysis of 16 channel images of beads labeled each of the fluorescent probes in specific proportions. The number of each type of bead is determined using spectral deconvolution, which is judged valid if the relative proportions of the different beads fall within 10% of the known proportions.
b.The accuracy of nuclear segmentation is quantified as an F1 score (composite of precision and recall) obtained from 5 random fields (quantified relative to manually annotated).Automated nuclear segmentation is judged valid if F1 scores are greater or equal than 0.8
c.Linearity in image collection is ensured by automatic elimination of regions of signal saturation (voxel values exceeding 4095) and detector dropout (voxel values of 0).
Materials
MATERIALS
DAPI Primary AntibodyThermofisherCatalog #D1306
Phalloidin Primary AntibodyThermofisherCatalog #O7466
THP Primary AntibodyR&D SystemsCatalog #AF5144
AQP1 Primary AntibodySanta Cruz BiotechnologyCatalog #sc-9878
MPO Primary AntibodyAbcamCatalog #ab9535
CD68 Primary AntibodyDakoCatalog #M0876
CD3 Primary AntibodyBD BiosciencesCatalog #557706
Siglec-8 Primary AntibodyBioLegendCatalog #347102
Secondary Antibodies
Phosphate buffered saline pH 7.4 (PBS)
PBST (PBS & 0.1% Triton X-100)MP BiomedicalsCatalog #807423
PBSTS (PBST & 10% NDS)Jackson ImmunoresearchCatalog #017-000-121
4% PFA
24 or 48 flat well bottom culture dishes
FluoromountSigmaCatalog #F4680
Glass slidesCatalog #3017
Coverglass #1.5 22X22mmThermo Fisher ScientificCatalog #12-541-B
Elmers Rubber Cement
Silicone grease
CML Latex BeadsThermofisherCatalog #C37253
STEP MATERIALS
Phosphate buffered saline pH 7.4 (PBS)
PBSTS (PBST & 10% NDS)Jackson ImmunoresearchCatalog #017-000-121
PBST (PBS & 0.1% Triton X-100)MP BiomedicalsCatalog #807423
PBSTS (PBST & 10% NDS)Jackson ImmunoresearchCatalog #017-000-121
PBST (PBS & 0.1% Triton X-100)MP BiomedicalsCatalog #807423
PBST (PBS & 0.1% Triton X-100)MP BiomedicalsCatalog #807423
FluoromountSigmaCatalog #F4680
Glass slidesCatalog #3017
Coverglass #1.5 22X22mmThermo Fisher ScientificCatalog #12-541-B
Coverglass #1.5 22X22mmThermo Fisher ScientificCatalog #12-541-B
Protocol materials
Elmers Rubber Cement
Materials
4% PFA
Materials
Siglec-8 Primary AntibodyBioLegendCatalog #347102
Materials
PBST (PBS & 0.1% Triton X-100)MP BiomedicalsCatalog #807423
In Materials, Materials, Materials, Materials and 3 steps
Glass slidesCatalog #3017
In Materials, Materials, Step 19
Coverglass #1.5 22X22mmThermo Fisher ScientificCatalog #12-541-B
In Materials, Materials, Materials and 2 steps
DAPI Primary AntibodyThermofisherCatalog #D1306
Materials
PBSTS (PBST & 10% NDS)Jackson ImmunoResearch Laboratories, Inc.Catalog #017-000-121
In Materials, Materials, Materials and 2 steps
THP Primary AntibodyR&D SystemsCatalog #AF5144
Materials
FluoromountMerck MilliporeSigma (Sigma-Aldrich)Catalog #F4680
In Materials, Materials, Step 19
Silicone grease
Materials
CD3 Primary AntibodyBecton Dickinson (BD)Catalog #557706
Materials
24 or 48 flat well bottom culture dishes
Materials
Phosphate buffered saline pH 7.4 (PBS)
In Materials, Materials, Step 9
MPO Primary AntibodyAbcamCatalog #ab9535
Materials
CML Latex BeadsThermofisherCatalog #C37253
Materials
AQP1 Primary AntibodySanta Cruz BiotechnologyCatalog #sc-9878
Materials
Secondary Antibodies
Materials
CD68 Primary AntibodyDakoCatalog #M0876
Materials
Phalloidin Primary AntibodyThermofisherCatalog #O7466
Materials
Label Free Imaging of Tissue
Label Free Imaging of Tissue
Mount sample on uncharged slide in PBS using prepositioned piece of spacer tape to support coverslip.
Gently push coverslip to the sample and seal with rubber cement dispensed from a syringe and 16 G needle.
Configure the microscope for multiphoton excitation at 910 nm for detecting SHG and auto-fluorescence (AF) (Table 2).
Equipment
SP8 confocal scan-head mounted upright DM6000 microscope
NAME
Confocal Microscope
TYPE
Leica
BRAND
n/a
SKU
Custom motorized stage coupled to a femtosecond pulsing MaiTai DeepSee laser (Spectra Physics, Santa Clara, CA) with EOM modulation for multiphoton excitation and a solid state laser launch providing 405 nm, 488 nm, 552 nm and 635 nm excitation. Imaging is performed with 25x 0.95 NA or 20x 0.75 NA Mimm objective with water immersion for label-free imaging or Leica immersion oil RI=1.51 with a 20x 0.75 NA Mimm objective for spectral confocal imaging. Detectors on the SP8 include two HyD detectors and two photo-multiplier tube (PMT) detectors ordered by wavelength, HyD1, PMT2, HyD3, PMT4. The SP8 is controlled by and linear unmixing is performed in Leica LASX software v. 3.5.2 build 18963.
SPECIFICATIONS
Configure emission path by opening confocal pinhole and adjust the bandpasses on 2 to 4 detectors to collect SHG (~455 nm) and AF as desired (refer to “Excitation and filter settings”). Use PMTs to avoid detector “overloading” and data loss.
Configure x, y and z limits to cover the entire tissue. Scan in x and y at 0.5-1 µm/pixel. Sample in z with 1-2 um steps.
Perform scans with standard galvanometers set to 400 Hz, and averaging by line 2x.
Fluorescence lifetime imaging is under development and may be included at this step.
Staining of Tissue with Fluorescent Small Modules and by Immunofluoresence
Staining of Tissue with Fluorescent Small Modules and by Immunofluoresence
Recover specimen by removing rubber cement and gently lifting coverslip.
Note
Lifting the coverslip may be easier if done after or while soaking the slide in PBS.
Washing: Sample is washed 1x in PBS >00:10:00
Phosphate buffered saline pH 7.4 (PBS)Sigma Aldrich
Blocking: Remove wash and block non-specific binding sites with PBSTS 4–8 hours at Room temperature on rocking platform.
PBSTS (PBST & 10% NDS)Sigma AldrichCatalog #017-000-121
Indirect primary antibody staining (Table 1): Incubate tissue in 200 µL staining solution for all the indirect detection primary antibodies 8-16 hours at Room temperature on rocking platform.
Washing: Wash sample with 2 mL PBST 2 times for a total of 06:00:00 .
PBST (PBS & 0.1% Triton X-100)Sigma AldrichCatalog #807423
Blocking: Remove wash and block non-specific binding sites with PBSTS 04:00:00 to 08:00:00 at Room temperature on rocking platform.
PBSTS (PBST & 10% NDS)Sigma AldrichCatalog #017-000-121
Indirect secondary antibody staining (Table 1): Incubate tissue in 200 µL staining solution for all the indirect detection secondary antibodies 08:00:00 to 16:00:00 at Room temperature on rocking platform.
Washing: Wash sample with 2 mL PBST twice for a total of 06:00:00 .
PBST (PBS & 0.1% Triton X-100)Sigma AldrichCatalog #807423
Blocking: Remove wash and block non-specific binding sites with PBSTS 04:00:00 to 08:00:00 at Room temperature on rocking platform.
Directly conjugated antibody staining (Table 1): Incubate tissue in 200 µL staining solution for all the direct detection primary antibodies 08:00:00 to 16:00:00 at Room temperature on rocking platform.
Washing: Wash sample with 2 mL PBST twice for a total of 06:00:00 .
PBST (PBS & 0.1% Triton X-100)Sigma AldrichCatalog #807423
Mount tissue with Prolong Glass or Fluoromount on pre-cleaned sides with silicone grease as needed to support coverslip.
FluoromountSigma AldrichCatalog #F4680
Glass slidesSigma AldrichCatalog #3017
Coverglass #1.5 22X22mmSigma AldrichCatalog #12-541-B
Seal cover slip to slide with rubber cement applied with a syringe.
Large-scale 3D Spectral Confocal Imaging
Large-scale 3D Spectral Confocal Imaging
Configure the microscope for spectral confocal imaging as described in Table 2 below.
! Settings need to be updated due to replacement of HYD's with PMT's.
Note
The settings for laser power, detector, voltage, gain and offset may vary based on the microscope configuration.
Configure x, y and z limits to cover the entire tissue. Scan in x and y at 0.5-1 µm/pixel. Sample in z with 1.04 µm steps.
Perform scans with standard galvanometers set to 400 Hz. The multispectral 16 channels are configured as indicated in Table 2 with 4 sequences of 4 detectors.
Preparation of Bead Slide for Reference Spectra
Preparation of Bead Slide for Reference Spectra
Incubate 20 µL of CML latex beads (ThermoFisher, cat# C37253) with 100 µg of fluorophore conjugated secondary overnight.
Note
Reference spectra for DAPI and Oregon Green are collected from tissue labeled only with DAPI or Oregon Green. The remaining reference spectra from fluorophores outlined in Table 1 are collected from single-fluorophore labeled latex beads.
Wash the beads after centrifuging at 1000 x g, Room temperature, 00:00:30 . Repeat twice. Dilute and mix individually labeled beads in MM (Prolong Glass or Fluoromount) such that 5-10 beads of each individually labeled beads is visible in a field size of 500 x 500 µm.
Beads are mounted under #1.5 coverglass and cured for 48:00:00 to 96:00:00
Coverglass #1.5 22X22mmSigma AldrichCatalog #12-541-B
Genearating Matrix for Linear Unmixing
Genearating Matrix for Linear Unmixing
Scan slide-mounted beads labeled with known fluorophores with settings given in the table below for channels 1-16.
Note
Data collected is subjected to linear unmixing with Leica LASX software
Select “Process” -> “Dye Separation” -> “Channel Dye Separation”
Select regions of interest including beads intensity and select “Add” to add to spectra.
Select Rescale to “Per Channel.”
Select “Load” to load a saved matrix. The saved matrix will be displayed near the bottom of the window.
Select “Apply”.
Applying Unmixing Matrix for Linear Unmixing
Applying Unmixing Matrix for Linear Unmixing
In the LASX software select image to unmix.
Note
Once a matrix file is generated (.sdm) it may be used again to apply collected images.
Select “Process” -> “Dye Separation” -> “Automatic Dye Separation”
Set fluorescent dyes to “8.”
Set Rescale to “Per Channel.”
Select “Load” to load a saved matrix. The saved matrix will be displayed near the bottom of the window.
Select “Apply”.
Stitiching/Merging of Unmixed Confocal Volumes
Stitiching/Merging of Unmixed Confocal Volumes
Select images to merge.
Note
Following unmixing, the confocal dataset is merged with LEICA LASX software.
Select “Process” -> “Mosaic Merge”
Check “Autostitching”, “Smooth” overlap and “Linear Blending”.
Select “Apply”.
Cytometric analysis of stitched and merged unmixed confocal volumes
Cytometric analysis of stitched and merged unmixed confocal volumes
All 3D cytometry is performed with the ImageJ plugin Volumetric Tissue Exploration and Analysis (VTEA). The settings for processing, segmentation and gating are given in analysis files. As software improves, new versions of VTEA may be used.