Isolation of Nuclei from Human Snap-frozen Liver Tissue for Single-nucleus RNA Sequencing

Materials and Reagents

1. 100 × 15 mm Petri dish (Corning^®, Falcon^®, catalog number: 351029)

2. Single-use scalpels No. 10 (Fisher Scientific, Feather^TM, catalog number: 08-927-5A)

3. MACS^® SmartStrainers 30 μm (Miltenyi Biotec, catalog number: 130-110-915)

4. FlowMi^TM 40 μm pipette tip strainers (Bel-Art, catalog number: H13680-0040)

5. 15 mL Falcon^TM conical centrifuge tubes (Corning^®, catalog number: 352096)

6. 50 mL Falcon^TM conical centrifuge tubes (Corning^®, catalog number: 352070)

7. DNA LoBind^® 1.5 mL microcentrifuge tubes (Eppendorf, catalog number: 022431021)

8. Fisherbrand^TM tweezers/forceps (Fisher Scientific, catalog number: 12-000-128)

9. Countess cell counting chamber slides (Thermo Fisher Scientific, catalog number: C10228)

10. Phosphate-buffered saline (PBS), 1× without calcium and magnesium, pH 7.4 (Corning^®, catalog number: 21-040-CM)

11. Nuclease-free water (not DEPC-treated) (Fisher Scientific, Invitrogen^TM, catalog number: AM9932)

12. IGEPAL^® CA-630 (Millipore Sigma, catalog number: 18896-50ML)

13. Sodium chloride (NaCl) (Millipore Sigma, catalog number: S5886-500G)

14. Magnesium chloride (MgCl₂) (Millipore Sigma, catalog number: M2393-100G)

15. Bovine serum albumin (BSA) (Millipore Sigma, catalog number: A8806-5G)

16. Hoechst 33342 10 mg/mL solution (Thermo Fisher Scientific, catalog number: H3570)

17. Protector RNase inhibitor (Millipore Sigma, catalog number: 3335402001)

18. Trizma^® hydrochloride solution (Tris-HCl) 1 M pH 7.4 (Millipore Sigma, catalog number: T2194)

19. 5 M NaCl stock (see Recipes)

20. 1 M MgCl₂ stock (see Recipes)

21. 10% IGEPAL (see Recipes)

22. 0.1% lysis buffer (see Recipes)

23. Wash and resuspension buffer (WRB) (see Recipes)

24. Hoechst stain buffer (see Recipes)

Equipment

1. Refrigerated benchtop centrifuge

   Note: Using a swinging bucket rotor for centrifugation helps to improve nuclei yields when compared with a fixed-angle rotor centrifuge.

2. Countess II FL automated cell counter, or equivalent cell counting method

3. Agilent Bioanalyzer

Procedure

1. Prepare materials and reagents

   1. Fill a rectangular ice pan and ice bucket with ice.

   2. Prepare the lysis buffer, wash and resuspension buffer (WRB), Hoechst stain buffer, and PBS. Place these reagents on ice.

   3. Set the benchtop centrifuge to 4°C and allow it to cool before use.

   4. Label the 15 mL Falcon tubes for samples and supernatant waste.

   5. Keep the forceps and scalpels chilled.

   6. Keep the tissue frozen on dry ice until lysis to avoid thawing, as this will lead to RNA degradation.

      Note: Ideally, liver biospecimens should be snap frozen and cryopreserved immediately after resection to prevent degradation. Tissue can be stored at -80°C or in liquid nitrogen until use.




2. Lysis

   1. Perform all lysis steps in a biosafety cabinet.

   2. Add 2 mL of ice-cold lysis buffer per 100 mg of tissue onto a Petri dish over ice in an ice pan.

   3. Using a clean set of forceps, transfer the snap-frozen biopsy into the lysis buffer on the dish. Do not allow the biopsy to thaw before placing it into the lysis buffer (Figure 2A).

      
      

      
      Figure 2. Scalpel homogenization of liver tissue biopsies. (A) Incubation of liver biopsies in lysis buffer. (B) Liver biopsies after a 5 min incubation and manual dissociation with a scalpel.
      

      
      

   4. Allow the tissue to thaw in the lysis buffer for 5 min.

      Note: If significant clumping of nuclei is observed, the lysis time may be reduced to prevent damage to the nuclear membrane.

   5. Mince the tissue in the lysis buffer using a pair of forceps and a scalpel. The tissue should be minced into 2–4 mm pieces (Figure 2B).

   6. Using a 5 mL serological pipette, add 2 mL of ice-cold PBS to the tissue lysate on the dish. Mix by gently pipetting up and down 10 times.

   7. Aspirate the lysate and filter through a 30 μm smart strainer into a 15 mL conical centrifuge tube.




3. Wash

   1. Centrifuge the tissue lysate at 500 × g for 5 min at 4°C. Remove the supernatant without disturbing the pellet.

      Note: The pellet may not be visible during this and subsequent wash steps (Figure 3). Care must be taken not to disturb and aspirate the pellet. A 5 mL serological pipette can be used to remove most of the supernatant, followed by a 1,000 μL pipette to carefully remove the remaining lysate.

      
      

      
      Figure 3. A small pellet of nuclei is present after lysis. The image shows a small pellet of nuclei after lysis, filtering, and centrifugation. The pellet may not be visible after centrifugation.

   
   

   2. Add 1 mL of ice-cold WRB and re-suspend the pellet by gently pipetting up and down 10–20 times.

   3. Filter the suspension through a 30 μm smart strainer into a new 15 mL conical centrifuge tube.

      Note: Tilt the strainer at a slight angle and pipette at the corner to allow for airflow and to reduce volume loss. If the suspension does not pass through, gently tap the strainer.

   4. Centrifuge at 500 × g for 5 min at 4°C. Using a 1,000 μL pipette, carefully remove the supernatant without disturbing the pellet.

   5. Add 70 μL of ice-cold WRB. Re-suspend the pellet by gently pipetting up and down 10–20 times. Transfer the nucleus suspension to a 1.5 mL LoBind microcentrifuge tube.

      Note: Compared to the 10× nuclei isolation protocol, our protocol includes only one wash step to accommodate samples with low yields of nuclei and to decrease run time. Additional wash steps can be included if the sample is contaminated and has a high enough yield (see below).




4. Nucleus counting

   1. Mix 10 μL of the nucleus suspension with 10 μL of Hoechst stain buffer in a PCR tube.

   2. Transfer 10 μL of the stained nucleus suspension to a Countess cell counting chamber slide.

   3. Assess the concentration and quality of the suspension (Figure 4) using a Countess II FL automated cell counter.

      Note: Cell counting and quality assessment of nuclei can be done with an automated cell counter or manually. We recommend using a fluorescent DNA stain, such as Hoechst 33342 or DAPI, to discern nuclei from tissue debris. Alternatively, trypan blue staining and visualization under brightfield can be used. However, this makes reliable estimation of nucleus concentration challenging, especially if the nuclei are not intact.

      
      

      
      Figure 4. Hoechst staining and imaging of isolated nuclei. Microscope images of nuclei in the middle square of a hemocytometer under (A) brightfield and (B) fluorescent Hoechst staining are shown to provide a higher resolution and scale compared to a Countess. In (B), examples of singlet, doublet, and degraded nuclei are labeled.

      
      

   4. If the concentration of nuclei is too high, dilute with WRB to achieve a target concentration of 1,000 nuclei per microliter. For loading nuclei into the 10× Chromium Controller, the concentration range limit is 100–2,000 nuclei per microliter, and the recommended target concentration is 700–1,200 nuclei per microliter. Dilute with WRB to reduce the concentration if necessary.

   5. If the nucleus suspension contains high amounts of debris or doublets and aggregates, perform an additional wash step by repeating steps C2–C5.

      Note: Higher levels of debris and nuclei aggregates, especially those larger than 50 µm in diameter, lead to a higher chance of clogging the microfluidic channel, resulting in a wetting failure. However, note that wash steps can result in a significant loss in yield of approximately 50%. Using a 40 µm pipette tip strainer in place of a 30 µm smart strainer can help prevent volume loss, although clogging may occur. The centrifugation speed and spin time can also be reduced, for example to 400 × g at 4 min, to help prevent the formation of nuclei aggregates.




5. Single-nucleus RNA-seq

   1. Prepare the suspension to load a predetermined number of nuclei. Suspensions can be diluted in nuclease-free water to achieve the target number of nuclei to load. For 10× solutions, consult the appropriate 10× user guide for nuclei recovery numbers given suspension concentrations.

      Note: Loading higher numbers of nuclei will result in a higher doublet rate. Doublet rates approximately follow a Poisson process (McGinnis et al., 2019) and the rate can be predicted from the number of nuclei loaded.

   2. For 10× solutions, perform the single-cell protocol detailed in the appropriate 10× user guide. The protocol should be adjusted accordingly if another platform, such as drop-seq (Macosko et al., 2015), is used. Briefly, these steps include 1) droplet/bead formation, 2) reverse transcription, 3) cDNA amplification, and 4) library construction. The details of each depend on the specific method and technology.

      Note: Begin the single-cell protocol as soon as possible. Droplet emulsification and reverse transcription should begin no more than a few hours after isolation.

      The number of PCR cycles to amplify cDNA is experiment- and sample-specific, but we recommend higher numbers to accommodate the lower RNA yields from nuclei compared to cells. Consult the single-cell protocol for appropriate ranges. For 10×, either 12 or 13 cycles are recommended.

   3. Assess the quality of the cDNA and sequencing libraries using an Agilent Bioanalyzer.

      Note: Ideally, there should be a smooth peak centered between 1,000 and 1,500 base pairs. However, RNA from human liver tissue is susceptible to degradation, thus cDNA traces can show significant peaks below 1,000 base pairs (Figure 5).

   
   

   
   Figure 5. Representative bioanalyzer trace of snRNA-seq library from frozen liver tissue. Figure shows a representative cDNA trace from an Agilent Bioanalyzer tape station. This frozen liver tissue sample shows RNA degradation with most cDNA fragments below a length of 1,000 base pairs.

   
   

6. Sequencing

   1. Sequence the libraries using the appropriate platform. For Illumina sequencing, select the number of lanes to achieve a read depth of 20,000–50,000 reads per nucleus.

      Note: The target read depth per nucleus will depend on the desired coverage and the amount of RNA in a sample. Generally, higher sequencing depths will have a higher sensitivity for more lowly expressed genes. Samples with high amounts of RNA will require higher sequencing depths to profile lowly expressed genes. In contrast, samples with low amounts of RNA will reach saturation at lower sequencing depths.

Data analysis

For alignment and gene quantification, we recommend using STARsolo (Dobin et al., 2013; Kaminow et al., 2021). If there is significant RNA degradation, then adapters, polyA tails, and template switch oligos can be trimmed. Raw snRNA-seq data from frozen tissues typically require quality control filtering and processing. A barcode-rank plot can help assess the quality and extent of nucleus and RNA degradation. If an elbow point and count threshold are not easily discerned, droplets can be filtered based on their gene count distributions using DIEM (Alvarez et al., 2020) or EmptyDrops (Lun et al., 2019). Our publications on snRNA-seq of liver (Alvarez et al., 2022; Rao et al., 2021) provide details on filtering droplets using DIEM (Alvarez et al., 2020). After droplet filtering, nuclei can be clustered using Seurat (Stuart et al., 2019) or another approach (Yu et al., 2022).

Adapting to different tissue samples

The steps in this protocol were optimized for snap-frozen liver tissue samples. However, our protocol can be optimized for other sample types as well. Two key areas for optimization are homogenization and washing. Mincing with a scalpel helps to preserve the integrity of the nuclei and reduce ambient RNA contamination. However, other homogenization methods, such as a dounce grinder, can be tested to help increase yields. The second area of optimization includes the number and volume of wash steps. We include only one wash step with 1 mL of buffer to minimize a loss in concentration. If a higher number of nuclei is present (>1,000 nuclei per microliter) from a sample type, an additional wash step can be included to reduce tissue debris. Further optimizations to this approach will involve balancing yield and quality, where greater homogenization and less washing improves yields and decreases the quality.

Cryopreservation and snap freezing of liver tissue

Proper cryopreservation and snap freezing of liver tissue is critical to ensure high quality RNA and intact nuclei. For snRNA-seq, it is critical to avoid RNA degradation. Therefore, ischemic time must be minimized, and tissue should be snap frozen as soon as possible. It is also necessary to reduce ice crystal formation, as this can potentially damage nuclei and reduce the yield. Suspensions of intact, single nuclei will generate the highest quality snRNA-seq data. Therefore, snap-freezing methods that rapidly and evenly freeze the tissue to preserve morphology are preferred. Once frozen, tissues can be stored in -80 °C, or, for long-term storage, in the vapor phase of liquid nitrogen. The tissue should then be thawed in lysis buffer only once the protocol begins.

Recipes

1. 5 M NaCl stock

   Add 2.922 g of NaCl to 10 mL of water.




2. 1 M MgCl₂ stock

   Add 2.0331 g of MgCl₂·6H₂O to 10 mL of water.




3. 10% IGEPAL

   Add 100 μL of IGEPAL to 900 μL of PBS.




4. 0.1% lysis buffer

   To 9.77 mL of nuclease-free water, add

   100 μL of 1 M Tris-HCl

   20 μL of 5 M NaCl stock

   10 μL of 1 M MgCl₂ stock per 10 mL

   100 μL of 10% IGEPAL

   75 μL of 40 U/μL RNase inhibitor.

   Note: RNase inhibitor in the lysis buffer helps to prevent RNA degradation during the lysis step.




5. Wash and resuspension buffer (WRB)

   1. Add 200 mg of BSA to a 15 mL conical centrifuge tube.

   2. Add up to 10 mL of 1× PBS.

   3. Add 50 μL of 40 U/μL RNase inhibitor.




6. Hoechst stain buffer

   1. Add 10 μL of 10 mg/mL Hoechst stain to 990 μL to create stock A.

   2. Add 2 μL of stock A to 98 μL of PBS to get a working solution.

Acknowledgements

We thank the individuals who participated in the liver HCC cohort (Alvarez et al., 2022). The work was supported by NIH grants R01HG010505 and R01DK132775. M.A. was supported by an HHMI Gilliam Fellowship.

References

1. Aizarani, N., Saviano, A., Sagar, Mailly, L., Durand, S., Herman, J. S., Pessaux, P., Baumert, T. F. and Grun, D. (2019). A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 572(7768): 199-204.
2. Alvarez, M., Rahmani, E., Jew, B., Garske, K. M., Miao, Z., Benhammou, J. N., Ye, C. J., Pisegna, J. R., Pietiläinen, K. H., Halperin, E., et al. (2020). Enhancing droplet-based single-nucleus RNA-seq resolution using the semi-supervised machine learning classifier DIEM. Sci Rep 10(1): 11019.
3. Alvarez, M., Benhammou, J. N., Darci-Maher, N., French, S. W., Han, S. B., Sinsheimer, J. S., Agopian, V. G., Pisegna, J. R. and Pajukanta, P. (2022). Human liver single nucleus and single cell RNA sequencing identify a hepatocellular carcinoma-associated cell-type affecting survival. Genome Med 14(1): 50.
4. Andrews, T. S., Atif, J., Liu, J. C., Perciani, C. T., Ma, X. Z., Thoeni, C., Slyper, M., Eraslan, G., Segerstolpe, A., Manuel, J., et al. (2022). Single-Cell, Single-Nucleus, and Spatial RNA Sequencing of the Human Liver Identifies Cholangiocyte and Mesenchymal Heterogeneity. Hepatol Commun 6(4): 821-840.
5. Ben-Moshe, S. and Itzkovitz, S. (2019). Spatial heterogeneity in the mammalian liver. Nat Rev Gastroenterol Hepatol 16(7): 395-410.
6. Dobin, A., Davis, C. A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M. and Gingeras, T. R. (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29(1): 15-21.
7. Drissen, R., Buza-Vidas, N., Woll, P., Thongjuea, S., Gambardella, A., Giustacchini, A., Mancini, E., Zriwil, A., Lutteropp, M., Grover, A., et al. (2016). Distinct myeloid progenitor-differentiation pathways identified through single-cell RNA sequencing. Nat Immunol 17(6): 666-676.
8. Habib, N., Avraham-Davidi, I., Basu, A., Burks, T., Shekhar, K., Hofree, M., Choudhury, S. R., Aguet, F., Gelfand, E., Ardlie, K., et al. (2017). Massively parallel single-nucleus RNA-seq with DroNc-seq. Nat Methods 14(10): 955-958.
9. Halpern, K. B., Shenhav, R., Matcovitch-Natan, O., Toth, B., Lemze, D., Golan, M., Massasa, E. E., Baydatch, S., Landen, S., Moor, A. E., et al. (2017). Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature 542(7641): 352-356.
10. Kaminow, B., Yunusov, D. and Dobin, A. (2021). STARsolo: accurate, fast and versatile mapping/quantification of single-cell and single-nucleus RNA-seq data. bioRxiv: 2021.2005.2005.442755.
11. Kuppe, C., Ibrahim, M. M., Kranz, J., Zhang, X., Ziegler, S., Perales-Paton, J., Jansen, J., Reimer, K. C., Smith, J. R., Dobie, R., et al. (2021). Decoding myofibroblast origins in human kidney fibrosis. Nature 589(7841): 281-286.
12. Larson, A. and Chin, M. T. (2021). A method for cryopreservation and single nucleus RNA-sequencing of normal adult human interventricular septum heart tissue reveals cellular diversity and function. BMC Med Genomics 14(1): 161.
13. Lun, A. T. L., Riesenfeld, S., Andrews, T., Dao, T. P., Gomes, T., participants in the 1st Human Cell Atlas, J. and Marioni, J. C. (2019). EmptyDrops: distinguishing cells from empty droplets in droplet-based single-cell RNA sequencing data. Genome Biol 20(1): 63.
14. Macosko, E. Z., Basu, A., Satija, R., Nemesh, J., Shekhar, K., Goldman, M., Tirosh, I., Bialas, A. R., Kamitaki, N., Martersteck, E. M., et al. (2015). Highly Parallel Genome-wide Expression Profiling of Individual Cells Using Nanoliter Droplets. Cell 161(5): 1202-1214.
15. MacParland, S. A., Liu, J. C., Ma, X. Z., Innes, B. T., Bartczak, A. M., Gage, B. K., Manuel, J., Khuu, N., Echeverri, J., Linares, I., et al. (2018). Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat Commun 9(1): 4383.
16. McGinnis, C. S., Murrow, L. M. and Gartner, Z. J. (2019). DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors. Cell Syst 8(4): 329-337 e324.
17. Rao, S., Yang, X., Ohshiro, K., Zaidi, S., Wang, Z., Shetty, K., Xiang, X., Hassan, M. I., Mohammad, T., Latham, P. S., et al. (2021). beta2-spectrin (SPTBN1) as a therapeutic target for diet-induced liver disease and preventing cancer development. Sci Transl Med 13(624): eabk2267.
18. Stuart, T., Butler, A., Hoffman, P., Hafemeister, C., Papalexi, E., Mauck, W. M., 3rd, Hao, Y., Stoeckius, M., Smibert, P. and Satija, R. (2019). Comprehensive Integration of Single-Cell Data. Cell 177(7): 1888-1902 e1821.
19. Trefts, E., Gannon, M. and Wasserman, D. H. (2017). The liver. Curr Biol 27(21): R1147-R1151.
20. Villani, A. C., Satija, R., Reynolds, G., Sarkizova, S., Shekhar, K., Fletcher, J., Griesbeck, M., Butler, A., Zheng, S., Lazo, S., et al. (2017). Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356(6335).
21. Xu, Y., Mizuno, T., Sridharan, A., Du, Y., Guo, M., Tang, J., Wikenheiser-Brokamp, K. A., Perl, A. T., Funari, V. A., Gokey, J. J., et al. (2016). Single-cell RNA sequencing identifies diverse roles of epithelial cells in idiopathic pulmonary fibrosis. JCI Insight 1(20): e90558.
22. Yu, L., Cao, Y., Yang, J. Y. H. and Yang, P. (2022). Benchmarking clustering algorithms on estimating the number of cell types from single-cell RNA-sequencing data. Genome Biol 23(1): 49.
23. Zhang, Q., He, Y., Luo, N., Patel, S. J., Han, Y., Gao, R., Modak, M., Carotta, S., Haslinger, C., Kind, D., et al. (2019). Landscape and Dynamics of Single Immune Cells in Hepatocellular Carcinoma. Cell 179(4): 829-845 e820.