Single-cell RNA Sequencing and Analysis of Human Pancreatic Islets

This protocol generates high throughput transcription data of individual human islet cells, which can be used to study anochrome cell heterogeneity, identify rare cell type, and gene regulation in disease. This technique generates larger scale single cell data, is easy to use, and has fairly short processing time. This method can be applied to islets from other species and to profile other freshly isolated tissues.

However, the protocol needs to be optimized to suit tissue-specific dissociation procedures. It is critical to prepare high quality dissociated cells. QC of the cell suspension before single cell partitioning is key, as is handling the emulsion carefully and quickly.

Some quality checkpoints like knowing if a chip has clogged are better seen than read. After obtaining human islets and incubating overnight, count and handpick 200 to 300 islets using a P200 pipette. Transfer the islets to a 15-millimeter conical tube containing five millimeters of complete islet media, pre-warmed to 37 degrees Celsius.

Place the tube in a centrifuge at 200 times G for two minutes. Gently aspirate the supernatant without disturbing the pellet on the bottom. Add one millimeter of pre-warmed cell dissociation solution and disrupt the pellet by pipetting gently up and down.

Incubate the islets at 37 degrees Celsius for nine to 11 minutes. Every three minutes, pipette up and down slowly for 10 seconds to dissociate the cells into single cells. Once the islet cells are well-dissociated and the solution becomes cloudy, add nine milliliters of complete islet media and filter through a 30-micrometer cell strainer into a new 15-milliliter conical tube.

Collect the cells by centrifugation at 400 times G for five minutes. Aspirate the supernatant and re-suspend the cell pellet in 200 to 300 microliters of 1X PBS 04%BSA solution. Now, mix 10 microliters of the cell culture with 0.5 microliters of AO/DAPI.

Pipette up and down to mix thoroughly. Load 10.5 microliters onto a slide and run the cell count assay on a fluorescence-based automated cell counter to determine count and viability. Place a microfluidic chip into a chip case.

Orient the chip case, ensuring oil wells are closest to the person performing the experiment. Use a pipette to add the calculated volume of cells into the prepared tube strips. Pipette to mix five times.

Without discarding the pipette tips, transfer 90 microliters of cell mixture to row one of the chip. Wait for 30 seconds, then pipette very slowly to load 40 microliters of gel beads to row two. Dispense 270 microliters of partitioning oil to the wells of row three.

Hook the chip gasket onto the tabs of the chip holder. Place the assembled chip holder into the single cell partitioning device and press the run button. Upon run completion, remove the chip gasket from the holder, open the chip case at a 45-degree angle, and transfer 100 microliters of the emulsion from the chip into a well in a blue plastic 65 well plate.

Dispense 125 microliters of pink emulsion breaking reagent into each emulsion. Wait for one minute, then transfer the entire volume of each well into each 0.2-milliliter tube. Ensure that there is a layer of clear and a layer of pink in the tube strip.

With a pipette, remove 125 microliters of the pink layer from the bottom of the tube strip without disturbing the clear layer. It is normal for a small volume of approximate 15 microliters of the pink layer to remain in the tube. Then add 200 microliters of clean up mix to each of the tube strips and incubate at room temperature for 10 minutes.

Transfer the tube strips to a magnetic stand and wait for two minutes to allow the solution to clear. Remove the supernatant and discard. Then wash the beads with 80%ethanol twice.

Allow the beads to dry for one minute. Remove the tube strips from the magnet and add 35.5 microliters of elution solution to the beads. Pipette to re-suspend the beads in the solution and incubate for two minutes at room temperature.

Transfer the tube strips to a magnetic stand and allow the solution to clear. Transfer the purified complementary DNA from the tube strips to clean 0.2-milliliter tube strips. To amplify the complementary DNA, first add 65 microliters of complementary DNA amplification master mix to each sample.

Place the tube strips in a thermocycler and run the program according to the manuscript. Samples can be stored at four degrees Celsius for up to 72 hours. After normalizing the complementary DNA to 15 anagrams and 20 microliters, aliquot 30 microliters of tagmentation mix to each complementary DNA sample on ice.

Put the samples in the thermocycler and run the tagmentation protocol at 55 degrees Celsius for five minutes, decreasing to 10 degrees Celsius and hold. Then add 60 microliters of sample index PCR Master Mix and 10 microliters of 20-micromolar forolegol sample index to the 30-microliter purified complementary DNA sample. Return the tubes to the thermocycler to amplify the final library product.

Normalize each sample with water to two nanograms per microliter and pull three microliters of each normalized sample together in a 1.5-milliliter tube. Dilute pool with elution buffer to 0.25 nanograms per microliter. Denature the pool by combining 12 microliters of the diluted pool sample, one microliter of one animal or DNA control, two microliters of elution buffer, and five microliters of 0.4 normal sodium hydroxide.

Incubate the mixture for five minutes, then add 10 microliters of 200-millimolar Tris at pH8, and load 4.05 microliters of the mixture into 1345.95 microliters of HTA-1. Next, load 1.3 milliliters into the sequencer's cartridge and run according to the manufacturer's guidelines using a sequencing recipe. On the computer, run Cell Ranger to de-multiplex raw base call files generated by sequencing into FASTQ files.

Align FASTQ files to Human b37.3 genome assembly and use CSC gene model to obtain expression quantification. Examine the barcode rank plot to be sure of the separation of the cell-associated barcodes and the background. To control cell quality, exclude cells with less than 500 detected genes, less than 3, 000 total number of unique molecular identifiers and greater than 0.2 viability score.

Adjust the cut-offs according to tissue and cell types. Next, use R package mclust to remove cells that express more than one hormone gene. Use R package seurat to normalize gene expression by the total unique molecular identifiers and multiply by a scale factor of 10, 000 at the cell level.

Then detect variable genes using the average expression and dispersion of all cells. Adjust the cut-offs according to the tissue and cell types. Perform the principal component analysis with the variable genes.

Cluster cells with a selected number of principal components. Derive cell cluster-enriched genes by comparing one cell cluster with the rest of the cells. In this single cell RNA sequencing protocol, the acquired human islets were first dissociated as validated by the alpha and beta cells in RNA fluorescence and C2 hybridization.

A successful example of emulsion following the partitioning step shows a uniform pale, cloudy solution with minimal partitioning oil separated from the gel beads. In contrast, a poor quality emulsion with clear phase separation between the gel beads and oil could be due to a clog during the chip run. After complementary DNA amplification, a representative fragment-sized distribution shows the typical peak for a good quality complementary DNA sample resided near 1, 000 to 2, 000 base pairs.

The spike near 600 base pairs was specific to islet-complementary DNA. The fragment-sized distribution for the RNA sequencing libraries was between 300 and 500 base pairs. The clustering analysis revealed 12 cell types in the space of T-distributed Stochastic Neighbor Embedding dimensions.

Three sub-populations in alpha cells and four in beta cells were revealed. This technique allows researchers to build a comprehensive cell islets for human islets. With more data from non-diabetic and diabetic donors, it is used for resource for therapeutic target discovery.