Gene Expression Center
University of Wisconsin-Madison Biotechnology Center

Sample Prep for Single Cell/Nuclei RNA-seq

Posted on

With rare exception, the single most important factor in determining the success of a single cell RNA-seq library preparation is the condition and health of the single cell/nuclei suspension at the time their transcripts are captured. Please consider the following aspects of sample preparation to give your samples the best chances of success in these assays. Note that much of this information is covered and expanded upon in the 10X Genomics “Cell Preparation for Single Cell Protocols” handbook; we have tried to pull out specific points for emphasis here, but we strongly encourage you to review this document directly.

General Sample Prep

  • Preparation of high quality single cell/nuclei suspensions is the responsibility of the submitting lab; the GEC does not offer this as a service. We also do not perform any antibody labeling steps for users interested in cell surface protein applications (either CITE-seq or “hashtagging” for sample multiplexing). Please let us know if you are considering these applications so that we can ensure you have all of the information you need to get started.
  • Handle your cell/nuclei samples as if you are handling isolated RNA – maintain the samples on ice (as long as the cold temperature is not deleterious for your specific sample type); wear gloves at all times and use approved nuclease-free plastic consumables (tubes and filter-tip pipet tips). The GEC uses Eppendorf DNA LoBind tubes and Rainin LTS pipettes and filtered tips.
  • When trying to disperse aggregated cells or a centrifuged cell pellet, limited mixing with regular-bore pipet tips is acceptable. When working with mostly dissociated cells, any pipet mixing of the sample should be performed gently with factory-made wide-bore pipet tips. Do not cut your own wide-bore tips at the bench.
  • If you are preparing cells that are inherently RNase-rich (e.g. granulocytes or cells from pancreas, lung, or spleen), include an RNase inhibitor in your wash and resuspension buffers (the choice of which RNase inhibitor to use is important!). Preparations of single nuclei suspensions should always use an RNase inhibitor.
  • If your samples contain excessive amounts of red blood cells, consider adding an RBC lysis step. Since RBCs generally (but not always) lack a nucleus, they will usually not be counted with the AO/PI nuclear stain we use to count cells/assess viability and will therefore not be factored into the number of cells we load to target your cells of interest. However, the RBCs will still be captured and could potentially impact sequencing costs if they end up soaking up a significant proportion of the sequencing reads.
  • Low sample viability (1) will almost always dramatically reduce the number of cells we can capture; (2) may cause the experiment to fail outright if there is high RNA degradation; and (3) may complicate interpretation of the data if there is excessive “background” RNA from dead/lysed cells. If cell viability is an issue in your samples, consider trying magnetic bead-based cleanup (e.g. Miltenyi’s Dead Cell Removal Kit) or flow sorting with a live/dead marker like DAPI or 7-AAD.
  • Cryopreservation of samples is possible with caveats. If you are working with snap frozen tissue, isolating nuclei is recommended rather than trying to recover intact cells. If you are using a cryopreserved cell suspension, significant cell death is expected upon thawing the samples, and one of the viability enrichment steps described above is strongly recommended. One of the options for fixed cells (either 10X Genomics Flex or Parse Evercode) may be a suitable alternative to cryopreservation of fresh samples if you are working with samples that you are unable to immediately process into a viable single cell suspension.
  • If you are experiencing large aggregates or abundant debris in your cell suspensions, consider filtering the samples. Unless you have a very large amount of sample, the general recommendation would be to use 40 um Flowmi tip strainers. GEC staff can do this once the samples arrive in our lab if necessary; please note that some loss of sample volume is expected.
  • When the sample is not limited, we ask for ~100,000+ cells per sample to start with. Our preferred cell concentration for most 10X Genomics assays is 1,000 – 1,600 cells/uL depending on the desired cell capture target. If you are able to deliver the cells to the GEC in this range, it is always preferable to minimize hands-on time before we can load the sample.

Sample Buffer Options (10X Genomics fresh sample assays)

  • Please click here for the relevant pages from the 10X Genomics “Cell Preparation for Single Cell Protocols” handbook. When possible, the recommended buffer to use for the final resuspension of your samples is calcium- and magnesium-free PBS with 0.04% BSA. Many other buffers (e.g. many common types of cell media) are acceptable, and it is generally okay use FBS up to 10% or BSA up to 2% if it helps maintain the health of your samples.
  • In most cases, the samples will go into the cell partitioning and reverse transcription reactions in the same buffer in which you resuspend them, so it is important to ensure that buffer does not contain excessive amounts of components that will interfere with either the generation of the oil droplet emulsions (e.g. detergent/surfactants) or with the reverse transcription reaction (e.g. EDTA, excess Mg++ ions).

Considerations for Flow Sorting of Samples

  • Please click here for the relevant pages from the 10X Genomics handbook with guidance on sorting cells. We frequently work with staff from the UW Flow Cytometry Lab, which has a lab down the hall from ours in the Biotechnology Center.
  • In the majority of cases, sorted samples will need to be concentrated after the sort. GEC staff can do this, but it is helpful for you to know safe/preferred centrifugation parameters (speed in rcf/x g and time) for your cells so that we can make sure we are minimizing cell loss as part of this process.
  • We generally recommend using a larger/lower pressure nozzle in order to minimize stress on the cells, but samples can still be significantly impacted by these stresses. Please be aware of the risks of sorting larger numbers of samples in one batch, as the early sample(s) may potentially experience larger quality drops while sitting on ice after the sort.

Considerations for Nuclei

  • The limited resolution of our automated cell counter makes it difficult to evaluate the quality of isolated nuclei in our lab. We strongly recommend you evaluate your protocol by looking at some test nuclei preps under a microscope at 40x or 60x magnification. Please visit this link for example images of high quality nuclei and other general advice from 10X Genomics. If you do not have easy access to a microscope in your lab, the Biochemistry Optical Core next door to UWBC has facilities and equipment that you can use.
  • As mentioned above, nuclei preparations for RNA-seq assays should always include RNase inhibitor in the wash/resuspension buffers.
  • If you are preparing nuclei for the 10X Genomics Multiome ATAC + Gene Expression assay or the standalone ATAC assay, the 10X Genomics library preparation kits (typically ordered by the GEC) include a concentrated buffer that you must prepare and use for the final resuspension of the samples. In the Multiome version, preparation of the 1X buffer requires DTT and RNase inhibitor which you must obtain separately.

Considerations for Fixed Cells (Parse Evercode or 10X Genomics Flex)

  • As with general sample prep, the fixation of cells/nuclei for either Parse Evercode or 10X Genomics Flex is the responsibility of the submitting lab. Both vendors sell reagent kits for fixing samples for their assays. Attempting your own “homebrew” fixation is not recommended.
  • Since fixed cells will always stain as dead when using fluorescent dyes, we cannot rely on AO/PI measurements of the samples on our cell counter to assess the quality of the sample at the start of the library preparation. In case any troubleshooting is necessary, it is very important for you to have some measurement of the sample viability (preferably with images, e.g. from an automated cell counter) prior to fixation.

Please contact us (gecinfo@biotech.wisc.edu) if you are considering a single cell RNA-seq project. We are happy to connect you with additional resources or vendor-specific technical support to help you get started.