Single-cell RNA-seq and cold tissue dissociation

Single-cell RNA sequencing (scRNA-seq) enables access to the transcriptome at single-cell resolution, generating expression profiles for individual cells. As a result, scRNA-seq can provide detailed insight into cell-to-cell gene expression variability, helping researchers make meaningful discoveries in many different fields including immunology, oncology, and neuroscience.

Discover the recent advances in single-cell sequencing methodologies, their workflows, and applications

One of the first and most critical steps in an scRNA-seq protocol is the dissociation of tissues to yield fully-dissociated but intact and viable cells. Different options for this step exist, each with advantages and limitations that can influence your experimental success. For example, cold-temperature tissue dissociation is popular because it is thought to better preserve the natural transcriptome.

So, how should you approach tissue dissociation for your scRNA-seq experiment?

Like so much in science, the answer depends on your unique experiment and your desired outcomes. In this blog, we look at why tissue dissociation is such an integral part of scRNA-seq study and whether cold tissue dissociation could improve your results.

Why tissue dissociation matters for single-cell RNA sequencing

Many biological tissues are complex structures composed of a mixture of cell types that may have specialized and/or local functions and may be unevenly distributed within the tissue architecture. The gut mucosa, for example, is a complex organ that is stratified into the epithelial and lamina propria layers. In turn, each layer is comprised of various cell types that perform specific roles (1).

Similarly, a tumor is made up of numerous cell types — including infiltrating immune cells, inflammatory cells, cancer-associated fibroblasts, vascular cells, stromal cells, extracellular matrix, and tumor cells — that uniquely contribute to the development of cancer. ScRNA-seq provides a means to study this cell heterogeneity, supporting a greater understanding of cell-to-cell gene expression variability and, ultimately, informs the development of more effective diagnostics and treatments for diseases.

To profile single cells, the tissue needs to be dissociated. If this isn’t completed effectively, you could end up with a sample containing clumps, dead or dying cells, extracellular debris, and extracellular RNA. These would influence your data quality, leading to erroneous sequencing data.

There are several methods for tissue dissociation, each with unique advantages and limitations. So, what’s the best approach?

Warm tissue versus cold tissue dissociation

Tissue dissociation is usually carried out enzymatically, mechanically, or using a combination of the two methods. Whichever approach you opt for, isolating single cells from tissues requires a careful balance between thoroughly separating cells and obtaining sufficient yield without compromising cellular integrity or destroying more fragile cell types. If you attempt to maximize yield by exerting excessive chemical or mechanical pressure on cells, you might induce a stress response in the cells, which alters the expression levels of genes during sample preparation.

Furthermore, the process of tissue dissociation can lead to differences in cell-type composition and gene expression. For example, storing human peripheral blood cells at room temperature overnight induced the expression of thousands of genes (2). And major transcriptomic changes have been directly linked to the type of tissue dissociation used (3).

So, how can you minimize the impact on your transcriptome data?

Because most enzymes used in tissue dissociation — including most serine proteases, collagenase, dispase, and/or hyaluridonase — operate optimally at body temperatures, protocols usually set temperatures at approximately 37°C. Tissue dissociation at these temperatures is known as warm tissue dissociation.

But, because these are viable cells and their transcriptional machinery remains active at 37°C, extended incubation for tissue dissociation can introduce gene expression artifacts. An alternative approach that mitigates dissociation-induced transcription responses is to use cold-active enzymes and incubating at much lower temperatures.

For instance, the protease from the psychrophilic soil bacterium, Bacillus licheniformis, is one of a handful identified and tested that demonstrates enzymatic activity at temperatures as low as 6°C (4). It creates the possibility of performing tissue dissociation on ice, which is predicted to decrease cellular activity and arrest transcription thereby minimizing transcriptional artifacts.

The advantages and limitations of cold tissue dissociation

Various researchers have used the B. licheniformis protease and other cold-active proteases to compare cold and warm tissue dissociation of different tissue types. Overall, these studies have drawn two major conclusions, which are that the temperature of tissue dissociation can:

• induce changes in the transcriptome

• impact the composition of cell types within a sample

Generally, the experiments show that warm dissociation tends to produce transcriptional profiles with a higher level of artifacts, mainly associated with a stress response. For example, one study found a significant increase in the expression of apoptosis genes in adult mouse kidney when enzymatic dissociations took place at 37°C compared to 4°C (3).

Another experiment on mouse brain tissue dissociation compared enzymatic digestion at 37°C and mechanical dissociation at 4°C. The experiment highlighted the profound and consistent alterations in the transcriptome of neuronal and glial cells. These finding were mirrored in various tissues including retina (6), gut mucosa (1), and inner-ear tissues (7).

Some studies have highlighted the change in cell type composition caused by warm tissue dissociation. Denisenko et al. (2020) identified eight cell populations that were less abundant in warm-dissociated samples in comparison to cold-dissociated ones. Indeed, one cell type was almost completely eliminated from the warm-dissociated samples.

So, if you want a more accurate picture of the transcriptome, you may think that cold tissue dissociation is the way to go.

However, cold dissociation may also be linked to changes in cell type constitution. Indeed, Denisenko and colleagues also noted that certain cell types were more abundant in warm-dissociated samples, potentially indicating their inefficient dissociation by cold-active proteases (3).

An additional limitation of cold dissociation is that it can still induce cellular stress, like warm tissue dissociation does. Given that cold-shock responses are well-established in the literature, cold-shock stress artifacts may affect transcriptional data. However, at least in mammalian cells, transcriptional machinery is largely inactive, so it is unlikely that there would be transcript-level changes.

The future of single-cell RNA-seq and cold tissue dissociation

The method and conditions of tissue dissociation will influence cell yield, cell type constitution, and transcription state in scRNA-seq experiments. You can reduce transcriptional artifacts with cold tissue dissociation and, ultimately, produce a better representation of the ‘real life’ transcriptome.

Cold tissue dissociation is still limited by the availability of cold-active enzymes and their lower efficiencies. But the continued mining of psychrophilic organisms for better cold-active enzymes may well identify new enzymes. The optimization of these enzymes could lead to improvements in cold tissue dissociation. In turn, improved protocols will help realize the full potential of scRNA-seq, facilitating cutting-edge research and discovery across many scientific disciplines.