Stereological Estimation of Dopaminergic Neuron Number in the Mouse Substantia Nigra Using the Optical Fractionator and Standard Microscopy Equipment

Summary

This work presents a step-by-step protocol for the unbiased stereological estimation of dopaminergic neuronal cell numbers in the mouse substantia nigra using standard microscopy equipment (i.e., a light microscope, a motorized object table (x, y, z plane), and public domain software for digital image analysis.

Abstract

In pre-clinical Parkinson's disease research, analysis of the nigrostriatal tract, including quantification of dopaminergic neuron loss within the substantia nigra, is essential. To estimate the total dopaminergic neuron number, unbiased stereology using the optical fractionator method is currently considered the gold standard. Because the theory behind the optical fractionator method is complex and because stereology is difficult to achieve without specialized equipment, several commercially available complete stereology systems that include the necessary software do exist, purely for cell counting reasons. Since purchasing a specialized stereology setup is not always feasible, for many reasons, this report describes a method for the stereological estimation of dopaminergic neuronal cell counts using standard microscopy equipment, including a light microscope, a motorized object table (x, y, z plane) with imaging software, and a computer for analysis. A step-by-step explanation is given on how to perform stereological quantification using the optical fractionator method, and pre-programmed files for the calculation of estimated cell counts are provided. To assess the accuracy of this method, a comparison to data obtained from a commercially available stereology apparatus was performed. Comparable cell numbers were found using this protocol and the stereology device, thus demonstrating the precision of this protocol for unbiased stereology.

Introduction

The quantification of neuronal cell number is pivotal in pre-clinical Parkinson's disease research to determine the level of neurodegeneration within the substantia nigra (SN)^1,2. The unbiased stereological estimation of cell number in a region of interest is considered the gold standard^3,4,5.

Before the advent of unbiased stereology, the number of neurons in sections was assessed by manipulating counted cell profiles to correct for the variable probabilities that neurons come into sight in a section. One of the most commonly used methods was the correction of quantified cell counts described by Abercrombie⁶. This method attempted to take into account that cells can be quantified more than once if fragments of the same cell are found in adjacent thin sections. Therefore, Abercrombie and other authors generated equations that required assumptions about the shape, size, and orientation of the counted cells^7,8. However, these assumptions were usually not realized and therefore led to systematic errors and divergence from the actual cell number (i.e., bias). Moreover, the bias could not be reduced by additional sampling³.

For the stereological estimation of cell numbers using the optical fractionator, mathematical principles are applied to directly estimate the cell numbers directly in a defined, 3-dimensional volume. The advantage of this method is that it does not involve assumptions about the shape, size, and orientation of the cells being counted. Thus, the estimated cell numbers are closer to the true values and get closer as the sample size increases (i.e., unbiased)³. Because many rules must be followed when using stereology to keep the method unbiased, ready-to-use commercial stereology systems have been developed (for review, see Schmitz and Hof, 2005⁴). Specialized stereology systems implement design-based stereological methods with a priori defined probes and sampling schemes for stereological assessments that lead to independence from shape, size, spatial distribution, and orientation of the cells to be analyzed^4,9. However, commercially available stereology systems are expensive; this may limit implementation in new research.

The aim of this study was to develop a usable technique for the design-based stereological estimation of dopaminergic cell counts in the mouse SN, employing the optical fractionator method and using standard microscopy equipment (i.e., light microscope, standard microscope software, and a motorized x, y, z stage). For this, a step-by-step guide on how to process mouse brain tissue and how to estimate SN cell numbers using design-based unbiased stereology is presented. Moreover, templates for the calculation of the estimated cell numbers and coefficients of error (CE) are provided.

The method described here is not limited to the analysis of the SN, but can be adapted for use in other anatomically defined regions of the mouse or rat brain. For instance, unbiased stereology has been used to estimate neuronal cell numbers in the hippocampus¹⁰ and the locus coeruleus¹¹. Additionally, cell types other than neurons, such as astrocytes¹² and microglia¹³, can be assessed as well. Therefore, this method can be useful to scientists who intend to implement unbiased stereology in their research but are not willing to spend a lot of money to purchase a stereology system.

Protocol

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. The protocol was approved by local authorities at the Regierung von Unterfranken, Wuerzburg, Germany.

1. Tissue Processing and Immunohistochemistry

1. Euthanize mice with CO₂ or any other approved method.
2. Perfuse six 12-week-old C56Bl/6N male mice transcardially with 10 mL of 0.1 M phosphate-buffered saline (PBS) using a 25-G needle and an infusion pump, followed by 70 mL of 4% paraformaldehyde (PFA) in 0.1 M PBS.
   CAUTION: PFA is toxic, allergenic, and carcinogenic.
3. Dissect the mouse brain in the coronal plane with a brain matrix slicer at the region of +0.74 mm from Bregma (Figure 25, Paxinos and Franklin mouse brain atlas¹⁴).
4. Put the dorsal part of the brain, including the SN, in 4% PFA in 0.1 M PBS for 1 d at 4 °C for immersion-fixation.
5. Exchange the PFA for 30% sucrose/0.1 mol PBS solution for cryo-protection and incubate for another 2 d at 4 °C.
6. Put the dorsal part of the brain into a cryomold filled with optimal cutting temperature (OCT) compound and slowly freeze the tissue in liquid dry ice-cooled isopentane.
   CAUTION: Dry ice is extremely cold (-78 °C); use personal protective equipment (PPE), including gloves, while working with dry ice. Dry ice evaporates in CO₂; therefore, work under a fume hood.
7. Serially cut 30-µm cryo-sections in the coronal plane and collect the sections, starting at 2.46 mm from Bregma (Figure 51, Paxinos and Franklin mouse brain atlas¹⁴) and ending at 4.04 mm from Bregma (Figure 64, Paxinos and Franklin mouse brain atlas¹⁴).
8. Store four series in tubes that are filled with cryoprotectant (30% glycerol, 30% ethoxyethanol, and 40% PBS) at -20 °C. During the sectioning procedure, mark the right hemisphere by puncturing a small hole into the upper region of the brainstem.
9. For immunohistochemical staining, select one series of sections per mouse. Preincubate free-floating sections for 1 h in 10% normal goat serum (NGS)/2% BSA/0.5% detergent in 0.1 mol PBS on a shaker. Incubate the sections with rabbit anti-mouse tyrosine hydroxylase (TH; 1:1,000 dilution) antibody diluted in 2% NGS/2% BSA/0.5% detergent in 0.1 mol PBS overnight at 4 °C.
10. Apply secondary antibodies against rabbit Igs (1:100 dilution) for 2 h at room temperature, followed by avidin/biotin reagent (1:100 dilution). Incubate and stain with 3,3-diaminobenzidine-tetrahydrochloride (DAB) and H₂O₂. Mount the sections after staining them on coated object slides.
    NOTE: TH+ dopaminergic SN neurons are shown in Figure 1a. By staining one series, an average of 13 SN sections, extending from the rostral to caudal portions of the SN pars compacta (SNpc) and reticulata, are stained per animal (Figure 1b). The sections are separated by 120 µm (1/4 series) (Figure 1c).
    CAUTION: DAB is a suspected carcinogen. It is toxic by contact and inhalation. Use PPE when working with DAB.

2. Acquisition of Images

1. Capture TH immunohistochemically stained SN sections digitally using imaging software that is coupled to a microscope. Separately analyze each section of one series.
2. Use the scan slide option of the imaging software. Save in TIFF format for high-quality images (Figure 2).
   NOTE: Image resolution is 312 pixels per cm.
3. Press "Acquire" to open the acquisition window (Figure 2a).
4. Set the binning to "2" for grayscale images.
   NOTE: The acquisition of grayscale images instead of color pictures reduces the size of the final stack image file (Figure 2a).
5. Click on "Show Live" in the acquisition window to open a new window showing the live image of the section (Figure 2a and b).
6. Click on "Stage" in the acquisition window. Choose "Scan Slide" in the dropdown menu and check the box (Figure 2c).
   NOTE: This enables the acquisition and alignment of highly magnified SN images (630X magnification) in the x,y-plane, as well as the acquisition of stack images with a distance between consecutive images of 1 µm in the z-plane, covering a total range of 13 µm.
7. Select the 2.5X objective to search for the SN.
8. Change the objective to 63x.
9. Define the upper left corner (TopLeft) (Figure 2c) and lower right corner (LowRight) (Figure 2d) of the section by clicking on "Set."
10. Click on "Z-Series" and check the "Z-Plane series" box (Figure 2e).
11. Determine the actual mounted thickness of the sections at three randomly selected counting sites per section using the imaging software. For this, click on "View Top Offset" to define the top of the section (Figure 2e) and then click on "Stop View Top" (Figure 2f). Thereafter, click on "View Bottom Offset" (Figure 2f) and then click on "Stop View Bottom" (Figure 2g). Write down the thickness of the section.
12. Subtract 3 µm (guard zone) from the top offset and type the result into the top offset slot (Figure 2h). Subtract 13 µm (height of the optical disector) from the top offset number and type the result into the bottom offset slot (Figure 2i). Select the following parameters: steps, 14; size, 1.00 (Figure 2i).
    NOTE: This corresponds to a thickness in the z-plane of 13 µm, with a 1-µm distance between consecutive images.
13. Select the directory to save the file (Figure 2j).
14. Click on "Sequence" to start the acquisition (Figure 2j).
15. Click on "Processing" and select the following parameters: "All Planes," from "Sequence;" "Montage;" and the "Fast" and "Stitching" boxes, with an image overlap of 10% (Figure 2k). Start to stitch the images by clicking on "start".
    NOTE: This will generate stack images for analysis (Figure 2l, Figure 3a-e).

3. Sequence of Stereological Assessment

1. Perform the analysis of SN stack images using the NIH ImageJ software version 4.7.
2. Download the two plugins that are needed from the ImageJ.nih.gov website: the "Grid Overlay" plugin¹⁵ and the "Cell Counter" plugin¹⁶. Install both plugins as described.
3. For unbiased stereological assessment, define the counting parameters as follows: grid size, 130 x 130 µm; counting frame, 50 x 50 µm; and optical disector height, 13 µm.
4. Open the image by clicking on "File" → "Open" (Figure 4a) Set the image scale by using the "Analyze" → "Set Scale" command (Figure 4b). For this step, change the defined size from pixels to µm (Figure 4c).
   NOTE: For the analyzed images, 425 pixels corresponds to 100 µm.
5. Select the "Plugins" → "Grid" → "Grid Type: Lines" command to randomly insert a grid. Check the "Random Offset" box. Define the size of one square within the grid (grid-square) as 130 µm x 130 µm = 16,900 µm² (Figure 3b and Figure 4d and e).
6. Change image type from 16-bit to RGB Color (Click on "Image" → "Type" → "RGB Color") (Figure 4f). Locate the SNpc, as described by Baquet et al.¹⁷. Encircle the SNpc with the "Paintbrush Tool" (brush width: 11) (Figure 3c and Figure 4g-i).
7. Undo the "Set Scale" command by clicking "Analyze" → "Set Scale" → "Click to Remove Scale" (Figure 4j-l) to enable the performance of the calculations in pixels rather than µm.
8. Make a screenshot (Figure 4m), save the image file, and open the new file with ImageJ (Figure 4n). Select "Point" (Figure 4o) and a brush width of 25 (Figure 4p). Mark every grid-square that contacts the SN for placement of an optical disector (Figure 4q). Perform analysis of the cell number in all optical disectors that are localized fully or partially within the outlined SN.
9. Select one grid-square that includes parts of the SN. Measure the upper left corner of this square with the cursor in ImageJ to define the x and y coordinates to start with (Figure 4r).
   NOTE: Coordinates will be inserted for optical disector positioning using the "optical disector position" (Figure 4s), as described in step 4.
10. Calculate the optical disector position by using the spreadsheet template entitled, "Optical disector position," as described in step 4.
11. Calibrate the optical disector (macro written by Christopher S. Ward, Supplemental File) by clicking on "Plugins" → "Macros" → "Edit" (Figure 4t), open the "opt_dis_grid.txt" file (Figure 4u), and define the size of the "usergrid" in pixels that corresponds with the x,y dimension of the optical disector (Figure 4v). Select a size of 50 µm x 50 µm, so that the size of one side of the usergrid is defined as 212.5 pixels (50 x 4.25 pixels).
12. Determine the position of the optical disector in the stack image by inserting the ImageX and ImageY coordinates that define the center of the optical disector (Figure 4v). Run the macro ("Macros" → "Run Macro") (Figure 4w).
    NOTE: The grid will vanish from the stack image (Figure 3d and Figure 4x). The optical disector will persist when moving down in the z-plane (Figure 3e and Figure 4x).
13. Select "Plugins" → "Cell Counter" to start the Cell Counter plugin (Figure 4y). Initialize the Cell Counter and select a marker type (Figure 4z). Zoom in on the image (click on "+") and perform the stereological assessment of the cell numbers at a magnification of 200%.
14. Identify dopaminergic neuronal perikarya by their rounded or ovoid shape and cell size (Figure 1a). Quantify the number of cells using stereological rules.
    NOTE: Since the optical disector is a 3-dimensional cube (Figure 5a), define the exclusion sides as the front, left, and top sides of the cube (red). Therefore, don't count TH+ cells that come into contact with either the red line (left and front) or the top side. Add the results of the cell counts to the prepared spreadsheet file entitled, "Calculation of cell count" (step 5).
15. According to the SN image, calculate the next grid-square in which to place the optical disector for the next cell count, using x,y steps the size of the overlying grid (Figure 5b) and the "Optical disector position" spreadsheet template, as described in step 4.
16. Perform an estimation of the cell number using the "Calculation of cell count" spreadsheet (step 5).

4. Calculation of Optical Disector Position Using the "Optical Disector Position" Spreadsheet Template

1. Calculate the size and position of each optical disector to be placed into the overlying grid within the outline of the SN (Figure 5b) using the "Optical disector position.xlsx" spreadsheet file (Supplemental File).
2. Insert the conversion of pixel per µm, as defined by the "Set Scale" command, into the yellow marked positions of the "Optical disector position.xlsx" file (Figure 6a and Supplemental File).
3. Insert the planned size of the optical disector and the size of the grid-square of the overlying grid, defined by the length of one side (in µm), into the orange marked positions.
   NOTE: Results are depicted in the red boxes next to the orange boxes.
4. Start with the grid-square within the SN outline that was determined as the starting point (one grid-square was selected to begin with, as in step 3.9, above).
5. Insert the x and y coordinates, as measured in step 3.9, into the green boxes of the file (by using these values, the x,y coordinates for the center of the optical disector within this first square are calculated, and the results are depicted in the blue boxes). Run the opt_dis_grid.txt macro (Supplemental File).
6. Calculate the x,y coordinates of the consecutive optical disectors by inserting the grid-square distance (measured in number of squares) relative to the first grid-square (brown boxes).
   NOTE: +x: grid-square is located to the right; -x: grid-square is located to the left; +y: grid-square is located to the bottom; -y: grid-square is located to the top. Results are shown in the gray boxes.

5. Estimation of Cell Number Using the "Calculation of Cell Count" Spreadsheet

1. Calculate the estimated number of TH+ SN cells per animal (N) using the formula published by West et al., 1991³, where ∑Q^- is defined as the total number of TH+ neurons counted in all optical disectors of one brain section.
   NOTE: The height of the optical disector (h) in relation to the measured thickness of the section (t) is included in the equation. The area sampling fraction (asf) is defined as the proportion of the area (A) of the optical disector (A optical disector) frame size within the square size of the grid (A x,y step) (A optical disector/A x,y step) (Figure 3b), and the section sampling fraction (ssf) is defined as the proportion of sections of the whole serially cut brain:
   
   To ensure high-quality quantitative estimates, the coefficient of error (CE) should be lower than 0.10. Determine the CE using the formula by Keuker et al., 2001¹⁰:
   
2. For the stereological estimation of the cell number within the SN of each mouse, insert the indicated information regarding the total number of generated series per mouse, the height of the optical disector, the x,y area of the optical disector (A optical disector, in µm²), the x,y area of the grid-square (A x,y step, in µm²), and the mean measured thickness of the section (in µm) in the yellow boxes using the "Calculation of cell count.xlsx" spreadsheet file (Figure 6b and Supplemental File).
3. Analyze each SN section from the same animal, as described in step 5.
4. Insert cell counts for each optical disector in the gray boxes, each line referring to one SN section, using the "Calculation of cell count.xlsx" spreadsheet file (Supplemental File).
   NOTE: The estimated cell number and the calculated CE are depicted in the blue boxes (Figure 6b).

6. Estimation of Cell Number Using a Commercially Available Stereology System

1. Start the estimation of dopaminergic SNpc cell number by clicking on "Probes" → "Optical Fractionator Workflow" to open the Optical Fractionator Workflow window.
   NOTE: A new window will pop-up to determine if a new series of SN is going to be counted.
2. Start a new series of TH+ SN sections by clicking "Start a new subject" in the pop-up window.
3. Go through the different steps of the Optical Fractionator Workflow. Set up the subject in the first step. For this, insert the investigator's name, the subject's name (SN group), and other helpful information. Insert the numbers of sections to count for this SN. Insert the thickness of the cut tissue sections. Insert the section interval, which is four in this case. Insert the randomly determined section number to start with.
4. In the second step, set the microscope to low magnification (2X objective) and select "2X Mag Lens" from the menu.
5. In step 3, trace the region of interest with the left mouse button, which is the SNpc, as described by Baquet et al., 2009¹⁷.
6. Set the microscope to high magnification in step 4. For this, change the objective to 100X and select "100X Mag lens" from the dropdown menu.
7. Measure the mounted thickness in step 5 by focusing to the top and to the bottom of the section.
8. In step 6, define the counting frame size as 50 x 50 µm; this is the size of the optical disector (x,y).
9. In step 7, set the size of the systematic random sampling (SRS) grid as 130 x 130 µm.
10. Enter a 3-µm guard zone in step 8.
11. Save the settings in step 9.
12. Finally, count the TH+ dopaminergic cells in each of the optical disectors within the SRS grid in step 10 of the Optical Fractionator Workflow. For this, start by clicking on the "Counting" button.
13. After finishing one section, click on the "Begin Next Section" button to start the Optical Fractionator Workflow of the next SN section of the same SN series.
14. When the last SN section of one SN series has been quantified, click on "I've Finished Counting."
15. Click on the "View Results" button to display the results of the stereology sampling.

Results

Using the presented method, the estimated number of TH+ dopaminergic neurons in the right SN ranged between 7,363 and 7,987 cells and, in the left SN, between 7,446 and 7,904 cells. Thus, the mean number of dopaminergic neurons (± SEM) was 7,647 ± 83 cells for the right SN and 7,675 ± 66 for the left SN. The calculated CE for each animal was lower than 0.08 (range: 0.073-0.079) (Figure 7). To ascertain the comparability of this method with comm...

Discussion

Stereology starts with tissue processing. The serial cutting of SN tissue must be performed carefully to prevent the loss of sections during stereological analysis. Additionally, one essential step is to mark one hemisphere in order to distinguish the right from the left SN when performing stereology. Placing a tiny hole at the upper part of the brainstem generated the best results in the presented study. Moreover, since working with the optical fractionator method demands that the tissue is cut in thick sections of abou...

Materials

Name	Company	Catalog Number	Comments
Paxinos mouse atlas			The Mouse Brain George Paxinos Keith B.J.FranklinCopyright @2001 by Academic Press CD Rom designet & created by Paul Halasz
brain matrix slicer mouse	Zivic Instruments	BSMAS 001-1
paraformaldehyde	Merck	1040051000
sucrose /D(+) Saccharose	Roth	4621.1
isopentane	Roth	3927.1
glycerol	Merck	1040931000
Ethanol	Sigma Aldrich	32205-1L
Name	Company	Catalog Number	Comments
phosphate buffered saline ingredients:
sodium chloride	Sigma Aldrich	31434-1KG-R
potassium dihydrogen phosphate	Merck	1048731000
di-sodium hydrogen phosphate dihydrate	Merck	1065801000
potassium chloride	Merck	1049360500
normal goat serum	Dako	X0907
bovine serum albumin	Sigma	A4503-100G
Triton X-100	Sigma Aldrich	X100-100ml	detergent
3,3-Diaminobenzidine-tetrahydrochlorid/DAB tablets 10mg pH 7.0	Kem En Tec	4170
H2O2/ Hydrogen peroxide 30%	Merck	1072090250
avidin/biotin reagent	Thermo Scientific	32050	Standard Ultra Sensitive ABC Staining Kit, 1:100
rabbit anti mouse tyrosine hydroxylase antibody	abcam	Ab112	1:1000
biotinylated goat-anti-rabbit IgG H+L	vector laboratories	BA-1000	1:100
StereoInvestigator version 11.07	MBF
BX53 microscope	Olympus
Visiview	Visitron Systems GmbH	3.3.0.2
Axiophot2	Zeiss
ImageJ software	NIH	Version 4.7
Tissue-TEK OCT	Sakura	4583
dry ice
grid overlay plugin	Wayne Rasband		https://imagej.nih.gov/ij/plugins/graphic-overlay.html
cell counter plugin	Kurt de Vos		https://imagej.nih.gov/ij/plugins/cell-counter.html).
optical disector macro	Christopher Ward
C57Bl/6N male mice	Charles River, Germany
SuperFrost Plus coated object slides	Langenbrinck, Germany
25G needle Microlance 3	BD	300400
REGLO Analog Infusion pump	Ismatec	ISM 829
StereoInvestigator system			StereoInvestigator version 11.07
BX53 microscope			BX53 microscope
self-assorted stereology			Visiview
Axiophot2			Axiophot2