Guidelines for droplet-based single-cell | nuclei mRNA-seq (10x Genomics)
GENERAL GUIDELINES FOR SAMPLE PREPARATION
Single-cell or single-nuclei mRNA sequencing gives valuable information about the heterogeneity in gene expression present within a tissue or cell type. 10x offers 3’- and 5’-based sequencing of the transcriptome; the first one (3’ assay) being the most widely used and the latter (5’ assay) being used when profiling of the TCR/BCR receptors is also desired.
CHECKLIST CRITICAL POINTS
A successful droplet-based single-cell|nuclei sequencing experiment in the 10x Chromium system is highly dependent on the quality of the cells and/or nuclei. Therefore, it is important to be aware of some critical aspects of sample preparation to ensure a proper and good preparation of the sample:
1. High cell viability rate: ideally greater than 90% of viable cells (minimum of 70%)
To increase the rate of viable cells in the cell suspension, we recommend using the Miltenyi Dead Cell Removal Kit. FACS-sorting of living cells is also an option. Additionally, please check for cell viability with automated cell counters or manual count following trypan blue staining or equivalent.
2. Cells|nuclei should be intact
Avoid long sorting times (longer than 1h) as this may lead to loss of viability or stability. Maximal sorting time depends on cell type (sensitivity or robustness of cells to experimental procedures and conditions). Disrupted cells|nuclei will increase background and result in noisy data, as free-floating RNAs can be encapsulated and sequenced. In case you work with nuclei, or sensitive cells, you should add 0.2 U/µl RNase Inhibitor to your media. Please see specific recommendations from 10x Genomics about tested RNase inhibitors: https://kb.10xgenomics.com/hc/en-us/articles/360000087632 & https://kb.10xgenomics.com/hc/en-us/articles/360049543672 :::
3. Cell|nuclei suspension should be clean of debris and aggregates
It is crucial that the cell|nuclei suspension is clean of debris, and that cells or nuclei are not clumped together. These are frequent causes of clogging of the 10x chip, which leads to loss of the sample.
4. Observations
Approximately half of the cells loaded into the 10x microfluidic chip are encapsulated. There is a small portion of multiplets, i.e. more than one cell encapsulated in a single bead. The estimated rate of multiplets is 0.8% for every 1.000 cells|nuclei loaded, and these cannot be distinguished with the standard scRNA-seq assay.
INPUT MATERIAL
Type & media
Cells|nuclei suspension, with no or hardly any aggregates. If no sorting is done, we recommend filtering the cell|nuclei suspension with a 30-40 µm strainer; alternatively doing a density gradient centrifugation (e.g. Percoll, sucrose, Iodixanol, etc).
If you are working with nuclei, please see a suggested protocol for nuclei isolation at the end of these guidelines. In addition, we recommend going through the demonstrated protocols for nuclei isolation from 10X Genomics (link in page 4).
Cells|nuclei can be delivered in most commonly used media, such as DMEM, HBSS, PBS+BSA (0.04%), or up to 10% FCS. The medium needs to be EDTA-free (or at very low concentration!).
*medium should be calcium- and magnesium-free; these influence on the stability of the reverse transcriptase.
Amount & volume
The optimal sample concentration is 800-1200 cells|nuclei/µl. The maximum sample volume that can be loaded on the chip is ~40µl, thus sufficient cell density is needed.
To obtain 10k cells|nuclei encapsulated and sequenced, at least 22k should be delivered. However, if material is not limiting, we recommend delivering a higher number of cells|nuclei, e.g. 50-100k cells|nuclei in total within the optimal concentration of 800-1200 cells|nuclei/µl. In this case, we will take an aliquot of the sample to get the maximum number of cells|nuclei encapsulated.
We recommend using LoBind or BSA-coated tubes during sample preparation (and sorting) to reduce cell|nuclei loss by preventing cells or nuclei from sticking to the tube wall. We can provide BSA-coated tubes for you, or you can prepare them yourself following our recipe.
In few sporadic cases, we experienced low quality in GEM formation. Thus, if possible, please deliver anough cells for a second repetition of the encapsulation procedure, in case it is needed. This avoids the need to repeat the experiment completely.
NOTES
Sample quality control
We perform a visual inspection of the cells|nuclei under the microscope, to check the quality of the material and assess viability and concentration/number.
Pooling of multiple samples in a single run or sequencing more than 10k cells (valid for single-cell mRNA sequencing only)
The barcoding strategy allows poolimg of different samples in a single 10X run, sequencing more cells than the maximum 10k that can be done in a standard 10x experiment.
Cells labeling can be done with antibody- or lipid-based oligos compatible with the 10X workflow. The antibody-based approach is only suitable for mouse and human samples; the lipid-based approach is species-agnostic.
Labelling cells with antibody-conjugated oligos:
TotalSeq tagged antibodies from BioLegend (https://www.biolegend.com/en-us/totalseq) should be used for this barcoding strategy. The products fully compatible with 10x are TotalSeq B and TotalSeq C, the former used for the 3’ scRNA-seq and the latter used for 5’ scRNA-seq workflow, which is used for TCR/BCR profiling (https://kb.10xgenomics.com/hc/en-us/articles/360019665352-What-is-the-difference-between-TotalSeq-A-B-and-C-).
We can provide aliquots of TotalSeq antibodies upon request. Currently, we have 8 different barcoded TotalSeq A antibodies available that are compatible with the 3’ scRNA-seq. Mouse antibodies are directed against CD45 and H-2 MHC Class I (product numbers Cat#155801, 155819, 155821, 155823, 155825, 155827, 155829); human antibodies are directed against CD298 and ß2-Microglobulin (product numbers Cat#394601, 394619, 394623, 394625, 394627, 394629). Please make sure these proteins are expressed in your experimental system.
More information about the labeling protocol can be found under these links:
Labelling cells with lipid-conjugated oligos:
10x Genomics provides lipid-conjugated oligos (Cell Multiplexing Oligos, or CMOs) that bind to the lipids in the cell or nuclei membranes - and thus can be used for any cell type and species.
We can provide aliquots of CMOs for cell/nuclei barcoding. A cell labeling protocol can be found here
Please note that 10x has limited support for labeling nuclei with this approach.
Labelling should be performed by the user. We recommend to start with higher number of cells (500 k minimum, ideally 1-2 million) to compensate for cell loss during washing steps. Thorough washing of unbound antibodies/CMOs is critical for proper assignment of cells during data analysis.
SEQUENCING
Paired-end sequencing, 2x100 bp reads (different read lengths can be requested). Sequencing depth generally depends on cell type, but a good starting point is 30 k reads/cell or nuclei. In general, we observed that cells|nuclei derived from cell lines and organoids have a higher RNA content compared with those from primary culture. Therefore, we recommend a deeper sequencing of material from cultured cell lines and organoids (e.g. 50 k reads/cell), and a shallower sequencing depth for RNA from cells from primary culture (e.g. 25-30 k reads/cell).
Due to the design of the 10x Genomics system, only the 3’ ends of mRNAs are sequenced. This implies that we can only derive count information, and not other features about the transcripts. If such an approach is desired, we recommend using the SmartSeq2 approach, which sequences the full transcript.
Technical/biological replicates are dependent on experimental design and research question. We are glad to provide guidance with this.
BIOINFORMATICS ANALYSIS
Our single-cell RNA analysis pipeline gives information about the number of cells that were sequenced, estimation of background noise, average number of genes expressed per cell|nuclei, clustering analysis, and a list of genes that characterizes each cluster. More information can be found on this link: https://support.10xgenomics.com/single-cell-gene-expression/software/overview/welcome
In case you need further support, or any other type of information, we can arrange a meeting with one of our bioinformatics experts to discuss the experiment.
SAMPLE SUBMISSION
Because cells have to be loaded fresh into the microfluidic system, we need to process the samples immediately upon receiving. Therefore, we ask users to contact us 1-2 weeks in advance (or earlier, if you are planning a large experiment) so we can schedule a date and time for the immediate processing of the samples.
Please bring samples in the scheduled time, since our planning of downstream processes is done accordingly. Usually, samples should be delivered by 13h latest, preferably between Tuesday and Thursday. If your biological system/experimental setup does not allow to schedule the delivery earlier, please speak to us beforehand to agree on a date/time that suits both sides.
We strongly recommend using LoBind tubes or BSA-coated tubes for low cell numbers (<100 k cells in total). Use of non LoBind tubes will drastically reduce yields.
Providing project and sample information via the Rosalind Client Portal
Please log in to Rosalind to enter all necessary project information making use of the “Project Description” field during the project set-up dialogue. Give all relevant details about project design, sample preparation method and how the samples are eluted and/or preserved. Sample-specific information such as concentration, volume, used preparation protocols etc. should be provided when the samples are added to the created project.
Please also indicate if you require a bioinformatics analysis during project setup.
To submit samples, you must be registered in our Rosalind client portal. Therefore, please send an e-mail to genomecenter@tu-dresden.de with your contact details (name, working group, working group leader, your e-mail address and telephone number). This will enable us to set up an account for you. After you have received an automatic e-mail, please log in to Rosalind with your login details. If needed, we can reset your login details.
Following up project progress
You can accompany the status of your project in our Rosalind Client Portal. Documents with information about the preparation method and quality control of samples and libraries are provided for your project.