An in silico pipeline approach uncovers a potentially intricate network involving spike SARS-CoV-2 RNA, RNA vaccines, host RNA-binding proteins (RBPs), and host miRNAs at the cellular level – Journal of Genetic Engineering and Biotechnology

The results of an integrated system biology analysis on RBPs

The use of our pipeline in the analysis of all SARS-CoV-2 S protein-encoding mRNA from reported variants in literature as well as the mRNA vaccines recognize by the WHO, revealed the presence of RBP-targeted motifs that are conserved. Our method identified 22 possible RBPs motifs, 18 of which interact closely, as shown in Fig. 2.

STRING output showing the possible connection between RBPs. The specific RBPs capable of binding the identified target are highlighted in red, light blue, and green

Some RBPs can bind the spike RNA multiple times. The protein Y-box-binding protein 1 (YBX1), in particular, binds to all three RBP motifs identified (sequences motifs are shown in Fig. 1). YBX1 is involved in a variety of cellular functions, including transcription and translation control, pre-mRNA splicing, DNA repair, and mRNA packaging. This protein is also found in mRNP complexes and may play a role in microRNA processing. The possibility of YBX1-binding to multiple motifs distributed across the mRNA coding Spike could reduce cytoplasmic protein availability [25, 26].

Knocking down YBX1 expression inhibits K562 leukemic cells proliferation and myeloid differentiation, according to scientific literature. Through the GSK3B/Cyclin D1/Cyclin E1 pathway, YBX1 depletion induces apoptosis and decreases cyclin D1 and cyclin E1 protein expression, resulting in a reduction in Pancreatic Ductal Adenocarcinoma growth [27].

MATR3 (Matrin 3), ZNF638 (Zinc finger protein 638) and SNRNP70 (Small Nuclear Ribonucleoprotein U1 Subunit 70) proteins may also bind multiple spike mRNA motifs. The second and third RBP motifs are binded by MATR3, the first and third RBP motifs are binded by ZNF638, and the first and second RBP motifs are binded by SNRNP70 (Fig. 2). Matrin 3 is an RNA/DNA-binding protein with multiple functions in gene expression regulation, including stabilizing target RNAs, supporting transcription factor activity, and modulating chromatin architecture [28].

Matrin 3 is also involved in the regulation of the transcriptional and translational networks, which control cell growth and proliferation, most likely through the formation of nuclear protein complexes that modulate pro and antiapoptotic signaling pathways. Reduced cell proliferation, which causes endothelial necrosis, is caused by matrin 3 downregulation [29].

ZNF638 inhibition reduces adipocyte differentiation and expression of adipocyte-specific genes. ZNF638 promotes adipogenesis by acting as a transcriptional co-factor of CCAAT/enhancer-binding protein (C/EBP), which leads to the expression of the peroxisome proliferator-activated receptor (PPARG), which controls adipocyte differentiation [30].

SNRNP70 is a nuclear ribonucleoprotein that binds to spliceosomal RNA U1 and is a key component of the spliceosome (U1snRNP). SNRNP70’s cytoplasmic pool plays an important role in nerve-dependent acetylcholine receptor (AChR) clustering and neuromuscular synaptogenesis. SNRNP70 can modulate the transcriptome composition independently of its nuclear activities, according to current evidence [31].

The interaction of HNRNPA1 (heterogeneous nuclear ribonucleoprotein A1), HNRNPA1L2 (heterogeneous nuclear ribonucleoprotein A1 like 2), and HNRNPA2B1 (heterogeneous nuclear ribonucleoprotein A2/B1) is a key component of the network. The first motif can be binded by HNRNPA1, while the second motif can be binded by HNRNPH2 (Heterogeneous Nuclear Ribonucleoprotein H2). These proteins are part of a family of heterogeneous nuclear ribonucleoproteins (hnRNPs) that are involved in nucleoplasmic pre-mRNA processing, as well as mRNA maturation and transport [32].

These proteins’ ability to recognize and bind other RNA molecules, such as hairpin-containing primary transcripts (pri-miRNA), appears to be very intriguing. By increasing the affinity of DGCR8 for primary miRNA transcripts, this specific bind process promotes pri-miRNA processing [33].

HNRNPA1 is also an RNA-binding protein that plays a role in the life cycle of many DNA and RNA viruses. During viral infection and cellular stress, activation of the innate immune response causes a redistribution of the cytoplasmic HNRNPA1 process. HNRNPA1 is a trans-acting factor that binds to the 5′ untranslated RNA single strand of human rhinoviruses and influences virus translation [34].

The RNA-binding proteins RBM5, RBM28, RBM42, and RBM6, which are all members of the same protein family, as well play a key role in the network. Despite the lack of a direct link between RBM28 and the core network, these proteins appear to play a role in a variety of cellular mechanisms, including cell cycle arrest and apoptosis induction via pre-mRNA splicing of multiple target genes, including TP53. They appear to be critical for cancer cell transformation and progression prevention in a variety of cancers, including lung cancer [35].

In 75% of primary lung cancers, as well as prostate and breast cancer, a possible reduction in RBM5 availability, mostly localized in the cytoplasm, has already been described [36].

Reduced cytoplasmic availability of the Strubbelig-receptor family protein group (SRSF1, SRSF7, SRSF9) can determine autophagy suppression regulated by SRSF1 and apoptosis of colon and lung cancer cells induced by SRSF7 and SRSF9 reduction [37,38,39].

Other factors, such as LIN28A (Lin-28 homolog A), FXR2 (FMR1 autosomal homolog 2), CELF6 (CUGBP Elav-like family member 6), and RBFOX1 (RNA-binding Fox-1 homolog 1), are also reported in the core network (Fig. 2).

FXR2 is involved in RNA metabolism, neuronal plasticity, and muscle development, and LIN28A suppression has been shown to inhibit osteogenic differentiation and decrease the expression of several osteogenic genes [40].

CELF6 is associated with the pathogenesis of myotonic dystrophy by regulating muscle-specific alternative splicing (DM). Furthermore, CELF6 depletion promotes cell cycle progression, proliferation, and colony formation by modulating p21 gene expression [41]. Finally, RBFOX1 belongs to the Rbfox protein family, which are master regulators of gene networks involved in neurogenesis as well as mature neuronal functions. RBFOX1 has been linked to the regulation of gene networks that aid cell survival in stressful conditions [42, 43].

The central network is not directly correlated with PABPC5 (polyadenylate-binding protein 5), YBX2 (Y-box-binding protein 2), SAMD4A (sterile alpha motif domain containing 4A).

PABPC5 is involved in the regulation of mRNA metabolic processes in the cytoplasm, whereas YBX2 is a major constituent of messenger ribonucleoprotein particles and is involved in the regulation of the stability of germ cell mRNAs (mRNPs) [44, 45].

Finally, SAMD4A is a conserved RBP found in a variety of species, and it regulates gene translation and stability. SAMD4A has been linked to angiogenesis and tumor progression in breast cancer, and its low expression in human breast tumor tissues/cells has been linked to a poor prognosis and survival [46].

miRNA signature enrichment analysis

The in silico analysis was extended using the TomTom tool to analyze the presence of RNA fragment sequences generated by the degradation of RBP-binding motifs, taking into account a possible delay in physiological degradation of these motifs, such as a protective influence of RBPs-binding on motifs. With RBP target motif fragment sequences in common, we detected 39 candidate host miRNAs. (The complete list of miRNA is shown in Table 1.). It is well recognized that a single miRNA can regulate the expression of multiple target mRNAs (up to 100 at once) and that each mRNA can be regulated by multiple miRNAs [47]. Only miRNAs with more than two possible mRNA targets, cytoplasmic localization, and experimental evidence of regulatory effect in protein expression were selected from the miRNA list using GeneCodis4 and Mienturnet.

Only three miRNA (has-miR-455-3p, has-miR-31-3p, and has-miR-15b-5b) interact with each other with a cytoplasmic localization, in particular as part of cytoplasmic ribonucleoprotein granules in processing bodies (P-bodies) primarily composed of translationally repressed mRNAs (Figs. 3 and 4), indicating a regulatory effect in protein expression.

GeneCodis4 provided the network plots and bar chart results. Co-annotation and enrichment network plot (A). Only three miRNA interact with each other with a primary cytoplasmic localization as part of cytoplasmic ribonucleoprotein granules processing bodies, according to GeneCodis4 analysis of all 39 miRNA interactions (P-bodies). Each bar’s length corresponds to the −log10 (Adj. Pval) and the number of significant annotations found to agree with the color intensity (B)

The output of MIENTURNET. The miRNAs that are similar to each motif are separated into boxes. The three miRNAs identified by GeneCodis4 analysis are highlighted in the networks

We discovered that the same three miRNA generated by GeneCodis4 are characterized by the greatest number of possible interactors, and thus can modify the expression of different mRNA, using the software Mienturnet on the same subset of 39 miRNA.

Finally, we looked into the possibility of has-miR-455-3p, has-miR-31-3p, and has-miR-15b-5b becoming more available in the cytoplasm.

In some cancers, Hsa-mir-455-3p plays an important role. Reduced levels of Hsa-mir-455-3p in pancreatic cancer cells reduce cell apoptosis while increasing cell migration, invasion, and EMT (epithelial-mesenchymal transition). Surprisingly, an increase in small RNA that mimics the effect of has-mir-455-3p may have the opposite and protective effect [48].

Furthermore, increased levels of hsa-miR-21-3p in hepatocellular carcinoma were found to be significantly elevated, indicating poor overall survival [49].

Finally, hsa-miR-15b-5p has been shown to either promote or inhibit tumor progression in a variety of tumor types. In breast cancer, hsa-miR-15b-5p expression was upregulated compared to normal breast tissue, and it was associated with poor overall survival in patients. In breast cancer-promoting breast cancer cell proliferation, migration, and invasion, hsa-miR-15b-5p availability significantly decreased HPSE2 expression at both the mRNA and protein levels [50].