SARS-CoV-2: The Self-Nonself Issue and Diagnostic Tests

Introduction

Canonical immunology lays on the concept that the human immune system evolved to attack and destroy extraneous entities such as infectious pathogens (that is, the “nonself”), in this way defending the human host (that is, the “self”) from harmful infections.^[1] Accordingly, self-reactive lymphocytes that might react with peptides/structures present in the human host are selectively deleted from the immune repertoire to protect the host from self-reactivity.^[1,2]

Hence, identification and mathematical definition of self and nonself entities are a conditio sine qua non for furthering our still incomplete understanding of the immune system,^[3] exploiting the immunogenic potential of vaccines in fighting infectious agents, and formulating specific diagnostic tools.^[4] Indeed, discriminating self from nonself currently is all the more necessary at the molecular level because in silico comparative sequence analyses^[5,6] have documented that a high level of peptide sharing exists between pathogens and the human host. Immunologically, this peptide sharing highlights the risk that serological immunoassays for measuring patients' immune responses against a pathogen might actually reflect the extent of cross-reactivity phenomena targeting human proteins.^[7]

In this scientific framework, taking the clue from recent data on the peptide sharing between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and Homo sapiens,^[8-12] the present study used SARS-CoV-2 spike glycoprotein (gp) as a research model and mathematically quantified the phenetic and genetic sequence differences that characterize the viral antigen as nonself when compared with the human host. That is, the entire human proteome was searched for peptide sequences shared with the viral gp. Pentapeptides were used as scanning probes to determine the exact viral versus human peptide sharing because a five amino acid grouping is the minimal antigenic and immunogenic space sufficient to specify an immune reaction, that is five amino acids represent the immune measurement unit.^[13-20] Specifically, the research was addressed to find pentapeptide identities, i.e., perfect matches, between SARS-CoV-2 gp and human proteins. This is because for a perfect peptide match (i.e., 5/5 identities and no gaps allowed) there is one and only one corresponding nucleotide sequence while for a homologous peptide (i.e., a peptide where four out of five amino acids are identical but there is a gap) there would be more corresponding nucleotide sequences depending on the gap position.^[20]

Following whole human proteome analyses, data were obtained showing that only a handful of pentapeptides (exactly 107 out of 1,269 pentapeptides) are uniquely present in the viral protein antigen and absent in the human host, in this way specifying the SARS-CoV-2 spike gp as nonself when compared with the human proteome. However, this phenetic difference disappears at the genetic level. Indeed, furthering at the nucleotide level the sequence analyses revealed that the pentadecameric oligonucleotides coding for the 107 pentapeptides present in the SARS-CoV-2 spike gp and not expressed in the human proteome actually are present in human nucleic acids too.

Materials and Methods

SARS-CoV-2 spike amino acid and nucleotide sequences were retrieved from isolate Wuhan-Hu-1, GenBank: MN908947.3. The viral protein antigen, which is 1,273 amino acids long, was dissected into 1,269 pentapeptides offset by 1 residue, that is, overlapped each other by 4 residues (i.e., MFVFL, FVFLV, VFLVL, and so forth). Next, for each viral pentapeptide, the entire human proteome was searched for occurrences of the same pentapeptide match by using PIR Peptide match (https://research.bioinformatics.udel.edu/peptidematch/index.jsp) and Peptide Search program (https://www.uniprot.org/peptidesearch/).^[21]

Viral pentapeptides that are absent in the human proteome were further investigated at the genetic level using the nucleotide Basic Local Alignment Search Tool (BLASTn) program (http://blast.ncbi.nlm.nih.gov).^[22,23] That is, for each pentapeptide unique to SARS-CoV-2 spike gp, the corresponding coding pentadecameric oligonucleotide sequence was used as a probe to scan the entire human NCBI (National Center for Biotechnology Information) nucleotide collection searching for instances of the same identical oligonucleotide sequence (i.e., 15/15 identities and no gaps allowed).

Results

Discussion

This study analyzes the pentapeptide sharing between SARS-CoV-2 spike gp and the human proteome, and mathematically defines the identity of the viral antigen as a set of 107 pentapeptides uniquely present in the spike gp and absent in the human proteins. Crucially, this viral versus human phenetic specificity disappears at the genetic level. The data have relevant implications in SARS-CoV-2 immunology, vaccinology, and clinical diagnostics.

Indeed, in the immunological context described under Introduction, the presence in human nucleic acids of the oligonucleotide sequences coding for the 107 pentapeptides that phenetically specify the viral antigen fails to support the deterministic hypothesis according to which the immune system evolved to discriminate infectious nonself from noninfectious self.^[1-3] Rather, ►Tables 1 and 2 suggest that SARS-CoV-2 and humans derived their genetic information from common ancestral templates. In this regard, this study supports the viral eukaryogenesis hypothesis, according to which the primordial eukaryotic cell was a consortium consisting of a viral ancestor of the nucleus, an archaeal ancestor of the eukaryotic cytoplasm, and a bacterial ancestor of mitochondria.^[24-26]

Moreover, the present data confirm and strengthen the concept^[27-32] that only vaccine formulations based on peptide sequences uniquely present in infectious pathogens and absent in the host proteins have the potential to selectively hit the pathogens and halt infections. In the case in point, by being absent in the human proteome, the unique SARS-CoV-2 spike gp peptides described in ►Table 1 represent an ideal basic peptidome platform that could result in effective and highly specific anti-SARS-CoV-2 vaccines exempt from harmful cross-reactivity.

Clinically and of utmost importance in diagnostics, the viral versus human pentapeptide and oligonucleotide sharing shown in ►Table 1 and 2, respectively, could have a severe impact on the validity of the current polymerase chain reaction (PCR) tests for SARS-CoV-2 spike detection. In fact, claims have been reported about the rates of false negatives/positives in SARS-CoV-2 detection by means of serological and PCR tests,^[33-38] in this way raising numerous concerns. As observed by Viswanathan et al,^[39] healthy individuals may be falsely identified as positive, requiring confirmatory testing and potentially leading to the unnecessary isolation of these individuals. In agreement, Gubbay et al^[40] suggested that large-scale SARS-CoV-2 screening testing initiatives among low pretest probability populations should be evaluated thoroughly prior to implementation given the risk of false positives and consequent potential for harm at the individual and population levels, and this not to mention the enormous waste of economic resources that might be caused by unreliable large-scale SARS-CoV-2 tests. As a matter of fact, data from ►Table 1 clearly suggest that serological immunoassays for measuring antipathogen antibody response might actually be indicative of cross-reactions with human proteins. In line with ►Table 1, ►Table 2 indicates that SARS-CoV-2 spike detection by PCR might be affected by the risk that human nucleotide sequences can be amplified, thus generating false-positive results with consequent wrong medical diagnoses. Such a risk is real in light of the fact that oligonucleotide sequences have been shown to be shared between the human genome and primers that have been proposed/used for SARS-CoV-2 detection by PCR.^[41] Therefore, this study might be of help not only to understand cross-reactivity phenomena and to address new specific peptide-based approaches in anti-SARS-CoV-2 vaccinal protocols but also to define a specific and precise diagnostics of SARS-CoV-2 infection and disease.

In closing, it is also worth noting that, in agreement with reports from this laboratory,^{[8,10,11,42-44]} Khavinson et al^[45] have recently described an intense peptide sharing between almost all the SARS-CoV-2 proteins and human proteins, with hepta- and octamers scattered along the entire length of the SARS-CoV-2 spike protein molecule, thus furtherly supporting the possibility of cross-reactivity and consequent autoimmunity between SARS-CoV-2 and the human host.^[7]