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.