Dissociation of microdissected mouse brain tissue for artifact free single-cell RNA sequencing

Summary Single-cell RNA sequencing (scRNA-seq) provides the transcriptome of individual cells and addresses previously intractable problems including the central nervous system’s transcriptional responses during health and disease. However, dissociating brain cells is challenging and induces artificial transcriptional responses. Here, we describe an enzymatic dissociation method for mouse brain that prevents dissociation artifacts and lowers technical variations with standardized steps. We tested this protocol on microdissected brain tissue of 3-week- to 24-month-old mice and obtained high-quality scRNA-seq results. For complete details on the use and execution of this protocol, please refer to Safaiyan et al. (2021).


SUMMARY
Single-cell RNA sequencing (scRNA-seq) provides the transcriptome of individual cells and addresses previously intractable problems including the central nervous system's transcriptional responses during health and disease. However, dissociating brain cells is challenging and induces artificial transcriptional responses. Here, we describe an enzymatic dissociation method for mouse brain that prevents dissociation artifacts and lowers technical variations with standardized steps. We tested this protocol on microdissected brain tissue of 3-week-to 24-month-old mice and obtained high-quality scRNA-seq results. For complete details on the use and execution of this protocol, please refer to Safaiyan et al. (2021).

BEFORE YOU BEGIN
The protocol below was applied to mouse brains. It hasn't been tested for other species.
The workflow of the whole dissociation process is shown in Figure 1.

Preparation for tissue dissociation
Timing: 1 h 1. If gentleMACSä OctoDissociator is not available, set water bath temperature to 37 C. 2. Cool down centrifuge (applicable to 50 mL and 15 mL tubes) to 4 C. 3. Prepare the coating buffer by dissolving BSA in DPBS-CMF which might take few hours and filter it through a 20 mm filter.
Note: Store at À20 C.
3 % (w/v) BSA in DPBS-CMF 0.15 g Bovine Serum Albumin (BSA) + 5 mL DPBS-CMF (for one sample) 4. Thaw Enzyme P and Enzyme A from the ''Adult Brain Dissociation Kit'' on ice. 5. Using coating buffer, precoat all plastic ware that will be in contact with cells to avoid cells sticking to the plastic (1.5 mL tubes, 15 mL tubes, 50 mL tubes, C Tubes and pipette tips). For coating tubes, fill the tubes with coating buffer and let all the inner walls touch the buffer, afterward remove the buffer from the tube. The removed buffer can be used for coating further plastic ware. Before using tips on cells, draw coating buffer into the pipette tip once and dispense it again, then continue with the cell suspension. 6. Clean dissection tools with 70% ethanol and fill 100/17 mm petri dishes with ice.

MATERIALS AND EQUIPMENT
Buffer X, Enzyme P, Buffer Y and Enzyme A are from the Adult Brain Dissociation Kit Note: Enzyme Mix 1 can be directly prepared freshly in a C tube and stored on ice until ready to use (Figure 2A).
CRITICAL: Actinomycin D is toxic and must be handled with care. Wear gloves and avoid any skin contact.
Note: Prepare freshly in a separate tube and store on ice until ready to use ( Figure 2A). Note: A diluted solution of heparin inhibits blood clot formation and preserves the patency of the vascular system. Prepare freshly and store on ice until ready to use.
Note: Prepare freshly and store on ice until ready to use.
Note: Prepare freshly and store on ice until ready to use.
Note: Prepare freshly and store on ice until ready to use.
Note: Prepare freshly and store on ice until ready to use.

Mice perfusion and dissection
Timing: 30 min / mouse We routinely performed this protocol for young and old (3 weeks to 24 months old) mouse tissue from whole brain, or from dissected regions such as prefrontal cortex, hippocampus and white matter. While we have not tested this protocol on every age and region of the mouse brain, we believe that it is capable of isolating single cells from all regions and ages of mouse brain samples. We recommend performing a pilot experiment to determine the number of cells that can be obtained from the region of interest with one animal, then scale up accordingly. For the white matter tissue and prefrontal cortex (gray matter), we generally pool white matter dissected tissue from 2-3 mice together in one C tube. b. Perform cervical dislocation, cut the head and carefully free the brain from the skull. A proper transcardial perfusion results in white not pink colored brain. Transfer the brains in 15 mL tubes filled with ice-cold DPBS-CMF. 2. Dissection of targeted brain region. (Methods video S1: Microdissection of White and Grey matter of mouse brain, related to step 2.) a. Set up the dissection workplace to keep tissue ice cold during dissection as shown in Figures  2B and 2C. Perform the dissection in a petri dish (60/15 mm or similar), filled with cold DBPS-CMF, and on ice. Microdissected tissue are transferred to a new petri dish (60/15 mm or similar), filled with cold 1 mL of Enzyme Mix 1, and on ice. Prepare separate petri dish for each target regions (e.g one carrying white matter tissue and the other gray matter tissue Figures 2B and 2C). An explanatory Movie describing the mouse brain dissection of white and gray matter of the prefrontal cortex is provided (Methods video S1). The total dissection of all samples should be as quick as possible, and the total duration should be less than 2 hours. To prevent transcriptional responses, brains should be kept cold on ice and dissected tissue immediately placed in ice cold Enzyme Mix 1 with transcription inhibitor actinomycin D (Wu et al., 2017).
CRITICAL: All steps should be done on ice except the enzymatic digestion.

Tissue dissociation
Timing: 30 min In this step the tissue is dissociated using the automized gentleMACSä Octo Dissociator with a customized program, which decreases technical variation compared to manual mechanical dissociation.
3. Add 30 mL of Enzyme Mix 2 to each C Tube. 4. Tightly close the C Tubes and attach them upside down onto the gentleMACSä OctoDissociator.
Tap the tubes to move tissue on the walls into enzyme mix. 5. Run the ''15 min dissociation program'' (Figure 3A) 6. Upon completion of the program, detach the C Tubes from the gentleMACSä Octo Dissociator. 7. Place 70 mm filters on 50 mL tubes and pre-wet the filters with 500 mL of cold DPBS-CMF. Filter the dissociated tissue samples by pouring the dissected tissues onto the filters. 8. Add 9.5 mL of cold DPBS-CMF into each C Tubes, shake them gently and add the liquid to the respective filter to collect remaining tissue fragments and cells. 9. Discard the 70 mm filter and centrifuge cell suspension at 300 3 g for 10 min at 4 C. Aspirate supernatant slowly and avoid disturbing the cell pellet.
CRITICAL: The gentleMACSä OctoDissociator is using a custom shortened program for scRNA-seq, which may decrease cell yield, but is half of the duration which, as a result, limits transcriptional responses to dissociation.
Original gentleMACSä Program for dissecting brain tissue: First create a folder named ''GM8'' in a USB stick, and copy the program file into this folder. Then connect the USB stick to the device Click in the USB stick folder logo that appears on the left side of the screen Select the program file ''37C_cus_NT_15_1'' by clicking on the checkbox on the right side Click the Save button. A window will appear asking where to save the file. In this window click on the folder logo to find the folder where to save the file. Click ok.
Open the destination folder to ensure that the transfer was successful. For more details, please refer to gentleMACSä Octo Dissociator with Heaters Manuals section 4.5.1.
Note: The program ''37C_cus_NT_15_1'' consists of three incubations at 37 C of 4.5 min each and in between each incubation there is mechanical dissociation. First mechanical dissociation is set at 100 rpm and the last two are set at 300 rpm. The temperature is set to 37 C during the whole program. The total duration of the program is 15 min.
Note: If the gentleMACSä Octo Dissociator is not available in the lab, preheat the tissue in the Enzyme Mix 1 at 37 C for 15 min in a water bath at 37 C and proceed with manual dissociation on ice. For the manual dissociation, tissues should be triturated by three separate flame-polished Pasteur pipets with successively smaller openings (Preparation of 3 Pasteur Pipets and their application are described by (Bordt et al., 2020) and continue with Step 10.

Debris removal
Timing: 40 min In this step, cell debris are removed by a gradient centrifugation to obtain a single-cell suspension.
10. Resuspend the cell pellet gently in 1550 mL of cold DPBS-CMF and transfer the cell suspension to a 15 mL falcon. Do not vortex cell suspensions! 11. Add 450 mL of cold Debris Removal Solution from the ''Adult Brain Dissociation Kit''. 12. Mix well by using a 1000 mL pipette. 13. Overlay very gently and slowly with 2 times 1 mL cold DPBS-CMF by using a 1000 mL pipette. 14. Centrifuge at 4 C and 3000 3 g for 10 min with acceleration 9 and brake 9. 15. Three layers will form ( Figure 3B). Aspirate the two top layers completely with a vacuum pump. 16. Fill the tube up to 5 mL with cold DPBS-CMF. 17. Gently invert the tube three times. Do not vortex!

OPEN ACCESS
18. Centrifuge at 4 C and 1000 3 g for 10 min with acceleration 9 and brake 9. Aspirate supernatant completely and avoid disrupting the cell pellet.
CRITICAL: After the debris removal part, the next step depends on which method is chosen by the researchers.
Plate based scRNA-seq: please follow the tiles with ''for flow cytometry'', continue with step 32.

Myelin removal (for 103 Genomics)
Timing: 60 min The myelin removal part is necessary to remove myelin and big clumps, which can block the 103 loading chip. This step also removes most of the oligodendrocytes. A comparison of cell suspensions under light microscopy, before and after the myelin removal part, is shown in Figure 3C.
19. Resuspend the cell pellets in 270 mL of loading buffer and add 30 mL Myelin Removal Beads II. Mix well with the pipette, but do not vortex. Incubate for 15 min at 4 C.
Note: These specific volumes are sufficient to remove myelin as well as most of the oligodendrocytes from cell suspension of grey matter and white matter from three adult animals. Adjust the bead volume for the target brain region according to the data sheet.
20. Wash the cells by adding 10 times the labeling volume (in this case: 2700 mL of loading buffer) and centrifuge at 300 3 g for 10 min. 21. During the centrifugation time, place the appropriate number of LS Column (according to manufacturer of Myelin Removal Beads II) in the magnetic field of a suitable MACS Separator ( Figure 3D).
Note: From now on, preferably work in a 4 C cold room until step 26 is done.
22. Rinse the column with 3 mL loading buffer. 23. After the centrifugation, aspirate the supernatant completely. 24. Add 1000 mL of loading buffer for each LS Column and resuspend the cells. 25. Apply the cell suspension onto the LS column. Collect the unlabeled cells that pass through. This is the myelin freed sample. Do not trash! 26. Wash the column with 2 times 1 mL of loading buffer. Collect the total effluent, since it still contains the myelin freed sample. Only add new buffer for washing when the column reservoir is completely empty. 27. Filter the collected cell suspension through a 30 mm filter (prewetted with 500 mL DPBS-CMF) 28. Centrifuge for 10 min at 300 3 g at 4 C.

Cell counting (for 103 Genomics)
Timing: 10 min The cell number and viability are counted to confirm that the requirements for loading the sample on a 103 genomics chromium controller are fulfilled. After a final filtering step, the single-cell suspension is ready to be loaded on the 103 Genomics Chip.
29. After centrifugation, discard the supernatant and gently resuspend gray matter cell pellets in 120-160 mL of loading buffer (for 103 Genomics), and white matter cell pellets in 60-80 mL of loading buffer.
Note: When doing the experiment for the first time, start the resuspension with a quite low volume, so the cell density can be adjusted to 103 Genomics requirements accordingly.
30. Take 10 mL of the cell suspension and mix thoroughly with 10 mL of trypan blue. Load the mixture into a manual cell counting slide or an automated cell counter such as Bio-Rad T20 to determine the live and dead cells concentrations. Adjust the cell concentrations according to the recommendation of the 103 Genomics protocol. Troubleshooting 2 and 3 CRITICAL: If the protocol is done fast enough, the live cell ratio should be above the 70% recommended in the 103 protocol. If cell survival is low, optimize the duration of the protocol and make fresh buffers.
31. Before loading, filter the low volume cell suspension through a 45 mm filter (flow cytometry cell strainer). a. For filtering, detach the cap of a flow cytometry cell strainer from the tube and discard the tube. Pre-wet the 45 mm filter in the cap with 1000 mL of DPBS-CMF. b. Then carefully add the sample on top of the filter. On the bottom side of the filter, a drop will form. c. Carefully draw the hanging drop from the bottom of the filter with a 1000 mL pipette (Figure 4) CRITICAL: This filtering approach will free the small volume sample from debris gently with minimal loss.
Note: The data of single cell RNA-seq can be analyzed with Seurat (Stuart et al., 2019), or other available platforms.

Timing: 70 min
If the plate based scRNA-seq is required, after finishing step 18, please directly continue with step 32.

OPEN ACCESS
The single cell suspension is labeled with antibodies for flow cytometry, to finally gate microglia. 32. After the debris removal part, resuspend the cell pellets in 1 mL DPBS-CMF and filter the collected cell suspensions through a 30 mm filter (prewetted with the 500 mL DPBS-CMF) into 15 mL falcon. Wash the filter with 2 times 1 mL of DPBS-CMF and collect the total effluent. 33. Centrifuge for 10 min at 300 3 g at 4 C and discard the supernatant. 34. For each flow cytometry sample, resuspend the cell pellet (up to 10 7 cells) gently in 50 mL of blocking buffer. Incubate the samples for 10 min at 4 C. 35. Meanwhile, prepare the desired antibody staining mixture.
CRITICAL: All antibodies should be tested beforehand with unstained and single stained controls. We recommend applying the compensation process on the flow cytometry machine using control samples (unstained control samples and single stained control samples).
After the 10 min incubation, add 50 mL of the prepared antibody staining mixture to each sample. Incubate the samples for 20 min at 4 C in the dark. 36. Then add 3 mL of DPBS-CMF to each sample. 37. Centrifuge for 10 min at 300 3 g at 4 C. 38. Carefully remove the supernatant by a vacuum pump and/or by pipettes. Remove as much liquid as possible. 39. Add 500 mL of flow cytometry buffer to resuspend the pellet. 40. Put 7-AAD (final concentration: 25 mg/mL) in each sample and incubate for 5 min at 4 C in the dark (7-AAD can be used for the exclusion of non-viable cells in flow cytometry analysis). 41. Continue cell sorting in a single cell sorting mode with a flow cytometer (e.g., Sony SH800) selecting the proper gates for each antibody. 42. Standard forward scatter height versus area criteria were used to discard doublets and capture singlets. A membrane impermeant dye (7-AAD) is used for negative selection to remove dead cells. For the gating strategy for microglia, see Figure 5. Troubleshooting 2 and 3 Optional: Sort the CD45+CD11b+ single cells into 96 or 384 well plates filled with lysis buffer for a plate-based scRNA-seq method such as Smart-Seq2.

EXPECTED OUTCOMES
Using our optimized protocol, for 103 Genomics, we can achieve approximately 60,000 -120,000 live cells from white matter of three mice. However, the result can be influenced by the size of the

LIMITATIONS
For 103 Genomics, as white matter is composed of large amounts of myelin, a myelin removal step is necessary to avoid large myelin clumps that potentially block the capillaries of the 103 chromium loading chip. However, most of the oligodendrocytes are also removed as they attach to the myelin sheath. If oligodendrocytes are needed for further analyses, the myelin removal step can be replaced with sorting the live cell population.
In flow cytometry, the shear stress of sorting can activate the cells and lower cell survival rate if functional assays are planned.

TROUBLESHOOTING
Problem 1 High number of blood cells in analysis data. This could be due to insufficient transcardial perfusion or bleeding in the tissue.

Potential solution
Check color of the dissected tissue. If the tissue has abnormal blood clots or appears pinkish, improve transcardial perfusion. The red blood cell lysis solution from the Adult Brain Dissociation Kit can be used to lyse red blood cells. However, if brain resident immune cells are of interest, we recommend improving perfusion to eliminate most of the immune cells in the circulating blood.
(step 1 in step-by-step method details).

Problem 2
Low cell viability.

Potential solution
All steps of the protocol are optimized to increase viability. Maintaining all solutions fresh and conducting this protocol smoothly can improve viability. If problem persist, remove dead cells using the Dead Cell Removal Kit (step 30 in step-by-step method details).

Problem 3
Low number of cells.

Potential solution
If dissected tissue is small, the cell pellet might be barely visible after centrifugation and might be lost due to improper resuspension or accidently aspirated. Cell loss could be reduced by careful aspiration and resuspending carefully with a pipette multiple times (steps 9, 15, 18, 23, 29, 33, 38 in step-by-step method details). If those steps are not sufficient, try to increase the amount of starting material by increasing animal numbers. Also watch if the cell strainer is blocked, and prevent cells to pass through.

Problem 4
There is a limited number of animals which can be used for the experiment and they cannot be pooled together to achieve enough target tissue.

Potential solution
This can be solved by performing the dissociation on each sample separately first and then apply cell hashing (Stoeckius et al., 2018) or MULTI-seq (Gehring et al., 2020) methods to label each sample before pooling them together (step 2 in step-by-step method details).

Problem. 5
Analysis of flow cytometry isolated microglia scRNA-seq results identifies non-microglial cell types (for Flow Cytometry approach).

Potential solution
If the non-microglial cell types specifically originated from one experimental condition, this might be due to up regulation of microglial markers in other cell types. Otherwise, this indicates a problem with the flow cytometry. Before the experiment, performance of each antibody and dye should be validated. In the experiment, include all controls for flow cytometry: Negative (unstained) controls: determine background staining.
Single stained controls: validate the performance of each antibody or dye.
Fluorescence minus one (microglia) controls: discriminate target cell type by omitting the antibodies of target cells type.
If problem persist, consider enriching microglia by magnetic bead-based selection to reduce the contamination (steps 35 and 42 in step-by-step method details).

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Dr. Ozgun Gokce.

Materials availability
This study did not generate new unique reagents or mouse lines.

Data and code availability
The accession number for the single-cell RNA-seq data reported in this paper is GEO: GSE166548.