3.1 Single cell transcriptomics
Analysis of the transcriptome profile for a single cell is done using
single cell RNA sequencing (scRNA-seq). Most scRNA-seq workflows involve
the capture of mRNA based upon separation via the 3’polyadenylated
region of the transcript. However, many different protocols exist to
isolate other forms of RNA, such as ribosomal RNA (rRNA), or long
non-coding RNA (lncRNA) and are available in both low and high
throughput manners. Generally, library construction for scRNA-seq is
relatively similar regardless of the type of RNA to be isolated. Cells
must be lysed with the appropriate lysis buffer concentration and volume
as to expel the contents of the cell without denaturing the RNA to be
isolated followed by the “pulling down” of the RNA to be isolated.
Traditionally, this is done by using an oligonucleotide sequence which
is complementary to the poly adenylated tail of the transcript. Since
the polyadenylated region is not protein coding, this essentially
preserves the region of the transcript which would later be translated
into protein. Some of the highest throughput methods for capture of
polyadenylated RNAs involve bead-based separation, where the capture
bead is coated with the complementary oligonucleotide sequence, a
barcode that tags the transcript with a cell-identifier, and a second
barcode which serves as a unique molecular identifier (UMI) for each
transcript within the cell. After the library preparation, a third
barcode is enzymatically added that allows for differentiation of
different sample groups; this is especially important when multiple
samples are sequenced at the same time. This method is employed in
“droplet-based platforms” where the entire workflow is carried out
within an oil droplet.45 Droplet based workflows allow
for accurate identification and quantification of transcripts from a
single cell and elucidate heterogeneity within a given cell population.
A study by Kong et al., investigating the changes in the single cell
transcriptomic profiles within monocytes after induction with the widely
used Bacillus Calmette-Guerin (BCG) vaccine, found that the amount of
systemic inflammation and differential response to LPS upon secondary
immune stimulation were markedly reduced.46 BCG has
been used for decades to confer protection against the deadly
mycobacterium tuberculosis (TB). However, recent literature has
supported the idea that BCG confers immunity against other pathogens as
well, including viruses, and the phenomena has been dubbed “trained
immunity.” It has been hypothesized that the decline in the rate of BCG
vaccination (owing to the lack of TB infections worldwide), might have
contributed to the increase in COVID-19 mortality. In the study, an
elevated production of key pro-inflammatory cytokines such as IL-1β,
TNFα, and IL-6 was observed upon secondary challenge with LPS, following
induction with the BCG vaccine. Differential expression of inflammatory
mediating chemokines such as CCL3 and CCL4 was also observed, along with
a strong correlation between IL-1β and CCL3 and CCL4 before training.
Both CCL3 and CCL4 are regulators of inflammation in an inter-cellular
signaling dependent manner, highlighting the ability of scRNA-seq to
shed light upon the spatio-temporal indications provided in the
data.47 After training, or transcriptional
reprogramming induced by BCG vaccination, this correlation dampened or
disappeared, suggesting that the reprogramming event lowered sensitivity
of the monocytes to pro-inflammatory signals. It should be noted that
this trend was found for both IL-6 and TNFα along with their correlated
but differentially responding chemokine pairs. These observations are in
concordance with the hypothesis that BCG induction can lower systemic
inflammation upon secondary challenge with LPS. This study represents an
effective use of scRNA-seq to evaluate the transcriptional reprogramming
of monocytes upon BCG induction and touches on the possible indications
of spatial dynamics within the analyzed monocyte populations.