Case studies
Often overlooked, the microbiome constitutes an important mechanism
involved in both gut-mucosal and systemic immunity. 70-80% of all
immune cells are present in the gut. There are intricate interplays
associated with the intestinal microbiota and local mucosal immune
system which shape the immune response to invasive
pathogens.88 The gut microbiota influences many
aspects of host physiology such as the development of the immune system,
drug metabolism, regulation of inflammatory diseases, overall
nutritional status, and drug metabolism. For assessing the impact of the
microbiota on host physiology, it is important to have a model that
facilitates drawing meaningful biological conclusions. One such model is
the use of germfree (GF) mice which have dramatic impairments on their
metabolism and immune system.89 In the study put forth
by Manes et al., comparisons are presented between the proteomic and
transcriptomic expression profiles within the terminal ileum (a part of
the small intestine with a high concentration of commensal microbes) of
GF and conventional mice. The germfree status was treated as the
perturbed state. Upregulated (up) corresponded to GF/C ratio values
>1 and downregulated (down) corresponded to GF/C ratios
<1. The first part of their analysis involved the global
transcriptomic data of the ilea (total RNA-Sequencing) to assess
transcripts affected by germ status. Hierarchical cluster analysis (HCA)
of the data revealed two distinct clusters for up and down regulated
genes. It should be noted that different strains of mice were used, and
the germ status had a greater effect on the C57BL/10A mice than on the
BALB/c mice. Ileum proteome data was then analyzed to assess the impact
of germ status on protein expression. Like the trend in the
transcriptome data, mouse strains also had a stronger effect on
C57BL/10A mice than BALB/c, but this observation was milder in the
proteomics data set. Using HCA on the data, 63 proteins were found to be
affected by mouse strain. An HCA of these proteins was performed, and
they were partitioned into up and down regulated groups. Highly
expressed genes were identified more at the proteome level, however
those that were more effected by germ status were classified at both the
transcriptomic and proteomic levels. Next the group compared
significantly affected genes based on germ status in both datasets to
assess their statistically significant correlations. This was done by
tallying the list of genes and analyzing them with HCA. Most of the
genes which were significantly affected within one dataset were
unaffected in the other, showing a level of underlying discordance
between the two omics layers. However, those genes which were
significantly affected in both showed concordance more so than
discordance. Thus, the concordant and discordant genes between the two
omics layers were used to construct subsets for each. These subsets were
then put through ingenuity pathway analysis to uncover the enriched
biological pathways associated with them. Presence of the microbiota in
conventional mice was linked top upregulation of immune system pathways
for the gene subsets which were transcriptome-proteome concordant. The
group hypothesized that this was due to migration of immune cells to the
ileum as shown in previous studies.90 Also, metabolic
pathways were downregulated in conventional mice, and this was
representative for both transcriptome-proteome concordant
(concordant[T,P]) and discordant (discordant[T,P]) gene sets.
For pathway analysis on the transcriptome enriched and proteome enriched
genes, 22 pathways were downregulated in both and almost all of them
were immune related such as antigen presentation. 43 pathways were
upregulated in either or and they were almost all metabolism related.
Only three pathways were upregulated in both, and they were the
glutathione mediated detoxification, serotonin degradation, and
xenobiotic metabolism signaling pathways. Network analysis was performed
on the proteome subset and concordant[T,P] subset showing
downregulation of immune system pathways and upregulated metabolic
pathways, respectively. These pathways may be representative to those
that of which are normally resistant to perturbation but affected by
microbiota presence. Four pathways were upregulated in the
discordant[T,P] gene set and two of them highly overlapped, mTOR and
eukaryotic initiation factor 2 (eIF2) both of which contain protein
translation machinery. Using ingenuity biofunctions, genes annotated
with “RNA processing” or “Translation” and showed
intra-transcriptome concordance and transcriptome-proteome discordance
indicating mechanisms involved in posttranscriptional regulation. In
fact, all the ribosomal proteins (RP) and eIF genes are known to be post
transcriptionally regulated. However, this example is representative of
only a subset of the multiomics workflows, and reveals the need to
interrogate multiple omics layers to draw meaningful biological
conclusions. Due to transcriptomic-proteomic discordance caused by the
microbiota, neither of the layers alone is sufficient to produce a
complete picture of the complex influence the microbiota has on the
host’s physiology, emphasizing the need for integrative omics-based
approaches to fully elucidate the host-microbiota interactions and their
molecular underpinnings. An overview of the discussed “omics”
approaches and their application in analyzing various biomolecules
(transcripts, proteins, metabolites) within a cell is presented in
Figure 2.
The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)
has caused the disease COVID-19, a global health crisis. Mutated
variants of the SARS-CoV-2 genome are responsible for enhancing the
pathogenicity of the virus and thus the overall impact of the pandemic.
Using a multiomics approach, Thorne et al. demonstrated that SARS-CoV-2
affects more than just the adaptive immune response; the innate immune
system is implicated as well in the severity and transmission of the
virus.91 They first assessed the multiplicity of
infection for the first wave isolates (early lineage) and Alpha variant
isolates in Calu-3 human epithelial cells by measuring intracellular
copies of envelope (E), Nuclear capsid (N), and virion production. It
should be noted that viral dsRNA is classified as a PAMP which are
recognized by RNA sensing adaptors such as Mitochondrial
antiviral-signaling protein (MAVS). Interestingly, they noticed 6 hours
post infection (hpi) that the total area of dsRNA decreased for the
Alpha isolates even though replication was comparable between alpha and
first wave isolates, using single cell immunofluorescence. They
hypothesize that this could be due to two factors. One being that the
Alpha N protein can contribute to innate immune evasion by sequestering
dsRNA and thus induce epitope masking, the other being the transposable
elements contributing to the reduction of endogenous dsRNA production.
It has been shown that Alpha infection leads to lower expression and
secretion of interferon-β (IFNβ) - a prominent marker for innate immune
activation.92 At early time points (24 hpi), Alpha was
shown to induce less expression of IFNβ and interferon stimulated genes
(ISGs) when compared to an early lineage variant B.1.13
hCoV-19/England/IC19/2020 (IC19), using RNA-Seq. This suggested enhanced
innate immune evasion potential of the Alpha variant. To further
characterize innate immune antagonism by Alpha, they compared global
host responses through mass spectrometry-based protein abundance and
phosphopeptide enrichment as well as total RNA-seq in Calu-3 cells (10
and 24 hpi). Notably, changes in RNA abundance and protein
phosphorylation seem to be infection driven yet the changes in overall
protein abundance are less drastic. It was also noticed that there is no
clear correlation between protein/mRNA abundance and the levels of
protein phosphorylation, indicating that enhanced phosphorylation may be
driven by another mechanism independent of the levels of protein
abundance. Gene set enrichment analysis comparing Alpha to early lineage
variants highlights differences in pathways related to the innate immune
system. The highest scoring terms, namely IFNα, IFNβ, and
cytokine/chemokine signaling are the most enriched in the RNA and
phosphorylation datasets. Also, the RNA-seq and protein abundance
datasets show reduced expression of ISGs driven by Alpha infection which
agrees with Alpha driven reduction of IFNβ production. They discovered
that the innate immune activation is antagonized through up and
downstream modulation of TBK1 (a kinase involved in nucleic acid
sensing) related pathways, and due to decreased phosphorylation at early
time points, suggesting that delayed activation of viral recognition
signaling pathways is characteristic of the Alpha variant and not the
variants of earlier lineage. To further assess the differences in host
response between the different variants, viral RNA-seq and proteomics
data were analyzed in parallel, and it was uncovered that the innate
immune antagonist Orf9b was substantially increased in Alpha relative to
earlier lineage variants. Looking further into the impact of Orf9b on
the host response to the Alpha variant, RNA-seq data was used to map
target genes to their corresponding transcriptional regulators to
estimate transcription factor (TF) activities. By extracting
significantly regulated transcription factors from the enriched terms in
the RNA-seq pathway analysis, it became clear that IRF and STAT family
TFs are less activated in Alpha infected cells than by early lineage
variants. To corroborate this finding, single-cell immunofluorescence
showed reduced IRF3 nuclear translocation after infection with Alpha in
comparison to VIC. STAT1, STAT2, and IRF9 are downstream of the type I
IFN receptor, and they are inhibited by Orf6 in Alpha infected cells.
This inhibition leads to the inability of STAT1 and IRF3 to undergo
nuclear translocation. In conclusion, Orf9b is regulated by host
phosphorylation and suppresses MAVS downstream signaling by targeting
TOM70 binding pocket (Ser50 and Ser53). This study demonstrates how
multiomics approaches can be used to study changes within SARS-COV-2
variants and that the enhanced innate immune evasion is an important
mechanism for enhanced pathogenicity of the Alpha variant.
INNATE IMMUNE PATHWAYS IN COVID-19 PATHOGENESIS:
The COVID-19 pandemic caused by the SARS-CoV-2 virus has been known to
lead to immune dysregulation and potentially lead to acute respiratory
distress syndrome (ARDS). ARDS can be caused by hypercytokinemia,
otherwise known as a “cytokine storm.” This is characteristic of a
hyperactivated and unregulated pro-inflammatory cytokine response, which
can lead to inflammatory induced tissue damage for the host. Typically,
viral infection is mediated by type-I IFN gene induction leading to the
expression of interferon stimulated genes (ISG) that exhibit anti-viral
mechanisms of action. SARS-CoV-2 has developed mutations that are
structural changes in the proteome and can antagonize the host IFN
response. Thus, studying the host immune response to SARS-CoV-2
represents a fundamental step for diagnosing phenotypes indicative of a
dysregulated immune response and may provide insight on how to treat the
disease. In the study put forth by Zhou et al., bronchoalveolar lavage
fluid (BALF) was collected from 8 COVID-19 patients (SARS2), 146
community-acquired pneumonia patients (CAP), and 20 healthy controls
(Healthy).93 Meta-transcriptomic analysis revealed
56% alignment to the human genome among all samples, also captures the
transcriptomic information for microbes found in the samples. The
differentially expressed gene (DEG) population for the SARS2 vs healthy
(SARS2-H) comparison was markedly higher than the rest, showing that
SARS-CoV-2 infection causes significant perturbations from homeostasis
in the host lung tissue. Some key upregulated DEGs in SARS2-H included
pro-inflammatory cytokine and chemokine genes (IL-1β, CXCL17, CXCL8,
CCL2), anti-viral ISGs (IFIT, IFITM family genes), and calgranulin genes
(S100A8, S100A9, S100A12). Downregulated DEGs included genes involved in
morphogenesis and cellular migration (NCKAP1L, DOCK2, SPN, DOCK10).
Among the cell signaling pathways, “interferon signaling” was most
enriched in the SARS2 group and was also enriched to a lesser extent in
the Virus-like CAP samples. This indicates an IFN-driven response to
SARS-CoV-2 infection. Many of the enriched pathways for the SARS2 group
are classified as innate immune pathways such as NF-κB, TNF, and the
IL-17 pathways. Network analysis showed a densely connected subnetwork
between the ribosome and chemokine signaling pathway. Upon assessment of
the cytokine profiles, it was observed that the pro-inflammatory
cytokine expression dissipates over time, suggesting that inflammation
during COVID-19 is resolved over time and unquenched inflammation may
lead to detrimental outcomes regarding disease progression. CXCL17,
which has a role in neutrophil recruitment in the lung was the most
upregulated chemokine for SARS2. Monocyte attractors, such as CCL2 and
CCL7 were upregulated as well. Data indicated a correlation between
viral load and chemokine production, where higher viral load corresponds
to higher chemokine gene expression. IL1RN and ILB interleukin genes
were enriched specific to SARS2, validated by their quantification of
cognate protein products (IL-1Ra and IL-1β respectively) in COVID-19
patient plasma. IL-1Ra is the inhibitor to IL-1β, and IL-1β has been
previously reported to be the driving factor of the proinflammatory
response during ARDS, this suggests a potential biomarker for COVID-19
severity based upon the ratio of IL-1Ra and IL-1β. The IFN response was
then examined where SARS2 showed elevated expression of ISGs (83
significantly upregulated) compared to CAP subgroups which diminished
over time for each patient. These included IFIT and IFITM genes with
broad antiviral functionality (IFIT1,2,3 and IFITM2,3); these have also
been shown to inhibit viral cellular entry of SARS-CoV-2. Also, key
innate immunity associated transcription factors, namely IRF7 and STAT1
were markedly upregulated which could further potentiate the IFN
response. SARS2 also displays higher neutrophil population compared to
the pneumonia group, and less diversity in the T and B cell populations
when compared to the innate cell populations. In a previous study,
neutrophil to lymphocyte ratio was characteristic of disease severity in
COVID-19 cases, indicating yet another biomarker for disease
severity.94 This study showcases the capacity of
transcriptomics to detect phonotypes associated with disease severity at
different expression levels, such as DGE analysis of cytokine/chemokine,
ISG, and even broad morphogenic cell population determination. Several
potential crosstalk points between the innate and adaptive immune system
also bring to light the innate immune pathways which affect the
pro-inflammatory response to SARS-CoV-2 pathogenesis. Study of
host-pathogen interactions at both the innate and adaptive immune levels
is required to decode the early and late biomarkers associated with
pathogenicity and virulence and to develop effective treatment options
in the clinical setting.
CONCLUSIONS:
Multiomics approaches are powerful tools for studying host-pathogen
interactions and provide a comprehensive view of the innate immune
system. Transcriptomics and proteomics offer valuable information in
several areas such as differential gene expression, protein levels, site
of modification, proteoform profiling, higher order structure, and
protein interactions, both qualitatively and quantitatively, and with
high accuracy. Considering the role of metabolites in the regulation of
immune cell signaling; immunometabolism has gained more interest as
well. It can predict how precise changes in the metabolites involved in
the signaling pathways shape the immune response. Single cell
measurements on the other hand, have made it possible to observe the
heterogeneity between cells of the same population, whether it be the
heterogeneity between transcriptome, proteome or metabolome or the
architecture of a population of cells. Single cell omics has completely
revolutionized the field, as a tool to study systems biology. As is the
case with any technique, limitations regarding cost, highly specialized
instrumentation, trained personnel, or availability remain for omics
methods. Table 1 summarizes prominent features, limitations and
executability of various “omics” technologies.
The development of advance instrumentation has truly empowered the omics
approaches and has reduced our reliance on high affinity antibodies for
exploring the immune system. High throughput imaging is highly
informative in exploring cellular morphology and has been used to
profile large compound libraries for drug discovery. Assessing molecular
interactions over time in order to predict disease pathogenesis has
become easier with the use of computational models and molecular
dynamics simulations. Latest deep learning-based tools such as AlphaFold
have established their value in structural predictions, investigating
protein-protein interactions, and simulating pathway dynamics more
accurately. Multi-omics analysis platforms have also revolutionized the
diagnosis and pathogenesis of Covid-19, the biggest pandemic of our era,
as well as aided in the identification of potential therapeutic
treatments. Taken together, or even in part, omics tools are highly
powerful in investigating systems biology closely, holistically, and
multidimensionally, providing a larger and more complete picture of the
innate immune landscape.