Proteomic investigation reveals dominant alterations of neutrophil degranulation and mRNA translation pathways in patients with COVID-19

Abstract

The altered molecular proteins and pathways in response to COVID-19 infection are still unclear. Here, we performed a comprehensive proteomics-based investigation of nasopharyngeal swab samples from patients with COVID-19 to study the host response by employing simple extraction strategies. Few of the host proteins such as interleukin-6, L-lactate dehydrogenase, C-reactive protein, Ferritin, and aspartate aminotransferase were found to be upregulated only in COVID-19-positive patients using targeted multiple reaction monitoring studies. The most important pathways identified by enrichment analysis were neutrophil degranulation, interleukin-12 signaling pathways, and mRNA translation of proteins thus providing the detailed investigation of host response in COVID-19 infection. Thus, we conclude that mass spectrometry-detected host proteins have a potential for disease severity progression; however, suitable validation strategies should be deployed for the clinical translation. Furthermore, the in silico docking of potential drugs with host proteins involved in the interleukin-12 signaling pathway might aid in COVID-19 therapeutic interventions.

Keywords: molecular biology; proteomics; specimen.

Graphical abstract

Figure 1

The workflow of sample preparation for detection of viral load and mass spectrometry analysis Swab samples were collected in viral transfer media and RNA was extracted followed by cDNA preparation which was analyzed by RT-PCR for determining the viral load. For MS analysis, proteins in swab samples were precipitated using three organic solvents namely acetone, ethanol and isopropanol. The proteins were digested, desalted and subjected to mass spectrometry analysis. Raw files from Mass Spectrometry were analyzed using Maxquant and significant proteins were selected for MRM analysis. (A) Schematic representation for RT-PCR and mass spectrometry-based detection of swab proteins. (B) Comparison of various extraction methods. The sample pool used in this study was prepared from all three solvents, which yielded the maximum average number of unique viral peptides as compared to the other extraction methods. (C) Representative MRM spectra for three viral peptides as observed in COVID-19 positive swab samples. The peptides belong to Replicase 1ab, Nucleocapsid and Spike proteins respectively.

Figure 2

Alteration of host proteome in response to COVID-19 infection To study the host proteome profile the mass spectrometry generated raw datasets were processed with MaxQuant software against the Human Swiss-Prot database. (A) Map of the Segregation of Positive, True Negative, and Recovered samples using PLSDA. Analysis of 18 positive, 11 recovered and 7 true negative samples showed segregation into three clusters. True negative clusters distinctly classified from recovered and positive samples. Sample 30, a recovered sample found to be placed within the positive sample cluster. (B) HeatMap of Positive, True Negative and Recovered samples. A list of 164 significant proteins found to be common between Positive vs Negative, Positive vs Recovered and Positive vs True Negative has been used to draw a hierarchical clustering based heatmap. The figure depicts the top 25 significant proteins found to segregate the groups using the ward clustering algorithm. (C) The unsupervised heatmap of 25 significant protein shows a perturbation between the non-severe and severe group. (D) Clustering analysis to segregate the Positive severe and non-severe samples. 12 severe positives and 11 non-severe positive samples were found to be segregated into two clusters in PLSDA. (E) Violin plot showing the expressional difference of clinical protein markers of COVID diagnosis. Violin plot of LDH-A and LDH-B – the subunits of Lactate dehydrogenase found to be significantly upregulated in COVID-19-positive samples when compared with COVID-19-negative samples. STAT1, a key regulator of interleukin also found to be significantly upregulated in COVID-19-positive, having biological connections with D-dimers and creatine phosphokinase which are clinical markers. HBA and HBB subunits of hemoglobin are also found to be significantly upregulated in the COVID-19-positive samples. (Unpaired Welch's T test; ns: 5.00e-02 < p <= 1.00e+00; ∗: 1.00e-02 < p <= 5.00e-02; ∗∗: 1.00e-03 < p <= 1.00e-02; ∗∗∗: 1.00e-04 

Figure 3

MRM validation of the host proteins upregulated in COVID-19-positive samples The MRM analysis was performed on 6 COVID-negative and 16 COVID-positive patient swab samples. 1 μg peptide from each sample was injected into a Vanquish HPLC coupled to a TSQ Altis™ triple quadrupole and run against a transition list of peptides belonging to important clinical markers. The list of transitions was prepared for unique peptides of these selected proteins using Skyline (Version 20.2.1.286). This list included a spiked-in synthetic heavy peptide (THCLYTHVCDAIK) used for monitoring the consistency of the mass spectrometry runs. For identification of the sensitivity of the peptide detection, we performed serial dilution of two crude synthetic peptides (heavy and light). The concentration of peptide was calculated using the Scopes method from its O.D. value at 205 nm and 280 nm. Different concentrations of the peptides starting from 25 to 125 ng were run in TSQ Altis™ Triple Quadrupole Mass Spectrometer. (A) Peak shapes of representative peptides for clinical marker proteins validated using MRM. The statistically significant change in the expression of the proteins was observed between COVID-19-positive and COVID-19-negative patient samples (T-test, ∗∗p < 0.05; Fold change > 1.5 at a confidence interval of 99% - determined by Skyline). (B) Box plots of representative peptides for clinical marker proteins validated using MRM. (C) Standard curve of heavy synthetic peptide HSGFEDELSEVLENQSSQAELK. The crude heavy synthetic peptide was diluted in the range of 25 to 125 ng concentration. The standard curve for this peptide was plotted using the peak area against the concentration of the peptide. The intensity of the peak area was proportional to the amount of the synthetic peptide. The lowest amount of synthetic peptide detected was at 30.9 ng. (D) Standard curve of light synthetic peptide HSGFEDELSEVLENQSSQAELK. The crude light synthetic peptide was diluted in the range of 25 to 125 ng concentration. The standard curve for this peptide was plotted using the peak area against the concentration of the peptide. The intensity of the peak area was proportional to the amount of the synthetic peptide. The lowest amount of synthetic peptide detected was at 19.8 ng.

Figure 4

Molecular pathway perturbations with the progression of COVID-19 infection The figure represents the enriched GO biological processes with their co-expressed proteins in the form a bipartite network where few proteins has been shown in the form of violin plot for COVID-19-positive (high and low viral load patients) and COVID-19-negative controls. (Unpaired Welch’s T-test, p-value annotation legends: ns: 5.00e-02 < p <= 1.00e+00; : 1.00e-02 < p <= 5.00e-02; ∗: 1.00e-03 < p <= 1.00e-02; ∗: 1.00e-04 < p <= 1.00e-03; ∗∗: p <= 1.00e-04). The representative images from the Reactome depicting the key pathways perturbed in the host are also shown.

Figure 5

Molecular docking studies of drugs binding to the host proteins involved in interleukin-12 signaling pathways Autodock Vina (1.1.2) was used to identify the drugs binding to the host proteins involved in the neutrophil degranulation, translation, and interleukin pathway. A screening of 29 FDA-approved, 9 clinical, and 20 pre-clinical trial drugs against the host protein identified several potential drug candidates targeting the interleukin-12 signaling pathway. (A) shows the drug Loratadine (green) docked with three proteins from the interleukin pathway; ADP-ribosylation factor 1 (binding energy or BE -9.8 kcal/mol), carbonic anhydrase 1 (BE -8.3 kcal/mol) and macrophage migration inhibitory factor (MIF) (BE -8 kcal/mol). (B) shows the same proteins docked with the drug UCPH-101 (cyan); ADP-ribosylation factor 1 (BE -9.9 Kcal/mol), carbonic anhydrase 1 (BE -8.5 kcal/mol) and MIF (BE -8.8 kcal/mol). Both of the drugs bind to all three proteins with negative binding energy greater than their respective control inhibitor. The interacting amino acid residues, which are present on the ligand-binding pocket are labeled. Almost all of the interacting residues belong to hydrophobic amino acids.