by Barbara Illi
Institute of Molecular Biology and Pathology, National Research Council (IBPM-CNR), c/o
Department of Biology and Biotechnology “Charles Darwin”, Sapienza University, Rome, Italy

Table of Contents

• Introduction
• Mediators of inflammation: acute phase reactants and the cytokine storm
• Interferons: beneficial or detrimental?
• CoV-2 and the Human Leukocyte Antigen system
• T cells response to CoV-2
• CoV-2 specific antibodies
• Immune dysregulation in Covid-19: lymphopenia and eosinopenia
• Conclusions

For non-experts

• Cytokines
• Transcriptome analysis
• Interferons
• The HLA system
• Respiratory viruses and immune response
• Antibodies

References

Introduction

The need to understand how to contain COVID-19 pandemic, through the development of vaccines and treatments, and to unveil how our self-defense machine, i.e. our immune system, responds to SARS-CoV-2 (hereafter CoV-2) infection, appears extremely urgent. Related to this last point, one issue is that, to date, immunological studies have been performed in hospitalized patients, which, frequently, underwent critical conditions. Noteworthy, an exacerbated immune response characterizes COVID-19 severe clinical cases, with lymphopenia and eosinopenia, activation of cytokine-secreting cells, consequent cytokine storm, leading to acute respiratory distress syndrome (ARDS), tissues damage, disseminated intravascular coagulation and multiorgan failure. However, it is now clear that our immune system employs all its army and weapons to fight CoV-2 and eradicate COVID-19¹.

Mediators of inflammation: acute phase reactants and the cytokine storm

Acute phase proteins, raising at early stages of the inflammatory process, due to infectious and non-infectious agents, may be diagnostic and prognostic markers of human disease². They include: C-reactive protein (CRP), alanine transaminase (ALT), lactate dehydorgenase (LDH), creatin kinase (CK).

In COVID-19 severe patients, acute phase proteins, together with cardiac troponin I, procalcitonin, increased prothrombin time, ferritin, D-dimer (which is a degradation product of fibrin, digested by the plasmin enzyme), fibrinogen, aspartate transaminase (AST), increased erythrocyte sedimentation rate, were constantly reported at diagnosis and correlated with the severity of the disease. Importantly, patients carrying these indicators and with dismal prognosis were older than less severe patients and likely affected by other pathologies (cardiovascular disease, diabetes, hypertension, cancer etc.)³.

COVID-19 severe patients recurrently presents what is called a cytokine storm syndrome, with clinical manifestations overlapping the secondary haemophagocytic lymphohisiocytosis (sHL), a hyperinflammatory syndrome, usually triggered by viral infections, characterized by unremitting fever, hyperferritinaemia, cytopenias⁵ and ARDS, occurring in the 50% of cases⁶. A typical cytokine profile detected in COVID-19 patients shows high levels of IL2, IL6, IL7, granulocyte colony stimulating factor (GCSF), IFNγ inducible protein 10 (IP10), monocyte chemoattractant protein 1 (MCP1), macrophage inflammatory protein 1α (MIP 1α), and TNFα^7,8. A recent transcriptome analysis from bronchoalveolar lavage fluid (BALF) and peripheral blood mononuclear cells (PBMCs) in a number of COVID-19 patients specimens has revealed that, in BALF, pathways related to viral replication are activated while in PBMCs the major transcripts output is represented by immune related pathways⁹.

Interferons: beneficial or detrimental?

Interferons are classified into three subtypes, alpha (α), beta (β) and gamma (γ). Interferons α and β belong to the type 1 subclass, whereas interferon γ represents the type 2 subclass. Recently, another interferon subclass has been identified, named type 3, called interferon lamba (λ) which, however, resembles type 1 interferons. CoV-2 has been demonstrated to weakly induce interferons¹⁰.

Given their antiviral properties, interferons-based prophylaxis and therapies have been exploited. In particular, interferon α and β have been used to treat hepatitis C and B, and preliminary data show that they could be efficient in prevent CoV-2 infection¹¹. Nevertheless, type 1 interferons-based therapy has considerable side effects, due to the ubiquitous expression of their receptors. Conversely, interferon λ is expressed on epithelial and immune cells and it has been demonstrated to prevent viral dissemination from nasal epithelial cells to the upper respiratory tract¹². Furthermore, interferon λ is a tissue-protective molecule rather than pro-inflammatory and stimulates both adaptive immunity and antibody production, essential for long term immunity. Therefore it may have beneficial effects on COVID-19 patients. Regarding the use of interferon λ in the clinics, this issue will be discussed in the episode about COVID-19 therapies and vaccines. Interestingly, it has been proven that both type 1 and type 2 interferons may induce ACE2 expression in human upper airway epithelial cells and in human primary basal cells from nasal scraping of human donors¹³. This finding suggests that CoV-2 may use interferons to enhance its infection ability. However, interferon λ-stimulated cells upregulate the guanylate binding protein 5 (GBP5), a protein that inhibits furin activity and viral spread from infected cells¹⁴. Whether interferons may be beneficial or detrimental for COVID-19 patients is a still debating issue and may depend on factors such as the time of infection, age, gender, co-morbidities.

CoV-2 and the Human Leukocyte Antigen system

Different HLA alleles have been demonstrated to confer different viral susceptibility to CoV-2 and disease outcomes, as occurs for SARS-CoV¹⁵. Recently, an in silico analysis (i.e. based on mathematical predictive models, not on experimental evidences) of the binding affinity of 145 HLA types for viral peptides of the entire CoV-2 proteome, revealed that HLA-A alleles show the best presentation capacity, whereas HLA-C the fewest.

The comparison with proteomes of other human CoVs, revealed that the best presenters of conserved peptides, and potentially capable to confer cross-protective immunity, are the HLA-A*02:02, HLA-B*15:03 and HLA-C*12:03 alleles (where A, B and C are the genes, the first number is the allele and the second number is the specific protein, according to the WHO Nomenclature Committee for Factors of the HLA System). Conversely, the HLA-A*25:01, HLA-C*01:02 and HLA-B*46:01 alleles are the least presenters, where the latter is also associated with a more severe disease due to SARS-CoV infection¹⁶. However, this study has the limit to be based on predictive models; therefore, unless validated, it has not any clinical value.

Regarding HLA class II proteins, HLA-DR deserves a special mention. HLA-DR is expressed on the surface of monocytes, such as dendritic and B cells and macrophages. HLA-DR presentation of pathogens-derived peptides to CD4⁺ and CD8⁺ T-cells initiates a specific immune response. Further, HLA-DR activates T-helper lymphocytes, playing a central role in the immune response against viral agents. HLA-DR molecules are underrepresented on the surface of CD14⁺ monocytes upon CoV-2 infection, as occurs during sepsis. When the number of HLA-DR⁺ monocytes drops, pneumonia evolves into severe respiratory failure, requiring mechanical ventilation¹⁷. This phenomenon may depend, at least in part, on the high quantity of IL6 found in critically ill COVID-19 patients, as IL6 downregulates the expression of HLA-DR¹⁸.

IL6 family if cytokines exerts many functions, including B cell activation, but are also involved in metabolic and neurotrophic control. In COVID-19 patients, IL6 is produced mainly by CD14⁺ monocytes and CD4⁺ T lymphocytes¹⁷. A negative correlation exists between IL6 levels and the number of HLA-DR molecules on the surface of CD14⁺ monocytes and a correlation also exists between the absolute number of HLA-DR molecules on CD14⁺ monocytes and the absolute number lymphocytes, which are frequently found decreased in whole blood cell counts (see below). Indeed, Tocilizumab, an IL6 blocker, restores HLA-DR expression on monocytes and even increases the number of lymphocytes in whole blood¹⁷.

T cells response to CoV-2

Very recently, it has been demonstrated that a robust CD4⁺ and CD8⁺ T cells response occurs in 70–100% of cultured PBMCs from COVID-19 patients, with respect to CoV-2 unexposed individuals. Specifically, CD4⁺ T lymphocytes were mainly activated when PBMCs were exposed to Spike (S) epitopes, less when Membrane (M) and Nucleocapsid (N) epitopes were used to stimulate COVID-19 PBMCs. CD8⁺ cells were activated by M and S epitopes, with at least other 8 CoV-2 Open Reading Frames (ORFs) targeted¹⁹.

Furthermore, COVID-19 patients makes anti-S antibodies at a level corresponding to the magnitude of the activation of S-specific CD4⁺ T cells and also CD4⁺ and CD8⁺ T cell response were well correlated. This observation perfectly fits with the role of CD4⁺ T cells in helping B cells to produce antibodies and to orchestrate a CD8⁺ T cell response.

Finally, in line and supporting the results exposed in the previous paragraph, related to a cross-protective immunity, in the 40–60% of unexposed individuals, CoV-2 responding CD4⁺ T cells were detected¹⁹. Interestingly, in contrast with other Coronaviruses, including SARS-CoV and MERS-CoV, which elicit mostly S, M and N-specific CD4⁺ and CD8⁺ responses, CoV-2 presents a broad panel of antigens able to stimulate CD4⁺ and CD8⁺ T cells response, including a variety of ORFs¹⁹.

CoV-2 specific antibodies

It is now well established that CoV-2 elicits an antibody response in COVID-19 patients, with M immunoglobulins (IgM, usually the first class of immunoglobulins appearing during a humoral response) emerging in the acute phase of the disease and IgG at later time points¹.

Antibodies raise from 5 to 10 days after the emergence of symptoms. However, different timing of seroconversion has been reported, with IgM and IgG peaks at 17–19 and 20–22 days after symptoms, respectively, in some cases. IgA, which mediate mucosal immunity and may reduce the viral attachment to the mucosa epithelium, have been also reported in CoV-2 infected patients, with an onset intermediate to that of IgM and IgG.

Interestingly and in support of the great importance of humoral response to fight COVID-19, children, who seem to exhibit low susceptibility to infection and milder symptoms, present high IgG levels within 1 week after the onset of the disease, with basically no IgM, indicating a very rapid seroconvesion from IgM to IgG. This may also account for an under-estimated number of asymptomatic pediatric patients²⁰.

Very recently, using S Receptor Binding Domain (RBD) of CoV-2 as a bait, B memory cells from recovered COVID-19 patients have been isolated and four antibodies produced by these cells have been demonstrated to specifically bind CoV-2 RBD, but not SARS-CoV RBD. Among them, two (named H4 and B38) demonstrated specific neutralizing activity, that is, they impaired S binding to ACE2. Specifically, at the RBD-B38 interface 18 of the 21 aminoacids involved in RBD-ACE2 interaction are employed, explaining why B38 blocks the binding of CoV-2 RBD to ACE2²¹. Importantly, B38 and H4 antibodies have been proven to be protective against COVID-19 in a mouse model of the disease²¹.

[UPD 2020/09/16] The entity of the humoral response is dependent on the disease severity. The more is the severity f the disease, the more is the antibody titer. This latter is 3000 fold higher than in hospitalized patients than in outpatients or convalescent plasma donors. The neutralization activity of the antibodies is also highest in hospitalized patients versus ooutpatients²⁷. It has also been reported that in mild cases a low level of antibodies corresponds to a long-lasting durantion of the potitivity to the viral RNA²⁸. Regarding the duration of humoral immunity, the majority of the literature assess that this last 2–3 months²⁹. However, the induction of B and T lymphocytes guarantees the acquisition of a cellular immunity¹⁹. Noteworthy, it has been recently discovered that unifected individuals or even individuals who had any contact with COVID-19 patients or SARS-CoV-2 infected people, possess memory T cells, stimulated by and specific for SARS-CoV-2, which recognize the nuclecpsid (N) protein of SARS-CoV and the non-streuctural proteins (nsp) 7 and 13 and N of other betacoronaviruses³⁰. This means that the incidental contact with other betacoronaviruses, including SARS-CoV, may induce a SARS-CoV-2 specific cellular immunity³⁰.[UPD end]

Immune dysregulation in COVID-19: lymphopenia and eosinopenia

Many literature reports describe lymphopenia to occur in critically ill COVID-19 patients, with T lymphocytes, including T regulatory lymphocytes (T_regs), which play a pivotal role in limiting excessive immune responses to pathogens, being the most affected. In some cases, also NK cells decrease and B lymphocytes are reduced at the lower limit of the reference values²².

The decrease in lymphocytes count, may be used as a diagnostic and prognostic marker for COVID-19. Indeed, patients with T lymphocytes greater than 20% at 10–12 days after the onset of symptoms were classified as mild-moderate; conversely, at that time point, patients with less than 20% T lymphocytes were classified as severe. At about 20 days after diagnosis, patients with more than 20% T lymphocytes were accepted as recovering; patients with a percentage of T lymphocytes between 5 and 20% were classified at still high risk, whereas individuals with a percentage of T cells less than 5% were critically ill. The molecular mechanism underlying this phenomenon is still unknown.

Interestingly, in PBMCs from COVID-19 patients an activation of the p53 pathway has been reported⁹, suggesting that T lymphocytes decrease may be due to apoptosis. This possibility is also supported by the evidence that T lymphocytes may be susceptible to CoV-2 infection²³.

Lymphopenia may be also dependent on the viral load to which an individual is exposed. Low viral loads may induce an appropriate T and B cells response and appearance of neutralizing antibodies, leading to virus clearance. High viral loads may provoke an excessive immune response, with cytokines storm, decreased T lymphocytes levels and dismal prognosis¹.

Another parameter to be taken into account is the eosinophils count. Eosinophils act in adaptive immunity and produce antiviral molecules. Further, they serve as APCs against respiratory viruses. In COVID-19 patients, a constantly reduced level of eosinophils has been detected, when compared to pneumonia-affected individuals^24,25. This phenomenon may depend on different factors: immune exhaustion, loss of eosinophilopoiesis (i.e. block of eosinophils production from the bone marrow) or recruitment to the site of infection, despite no accumulation of eosinophils has been found in autoptic specimens of CoV-2 infected lungs²⁶.

Conclusions

Basically, critically ill individuals shows high levels of cytokines, especially IL6, which lowers CD14⁺ monocytes and lymphocytes count in a negative feedback loop, as CD14⁺ monocytes and CD4⁺ T lymphocytes are the major producer of IL6 in COVID-19 affected individuals¹⁷.

The low number of T_regs is not sufficient to counteract an excessive immune response, enhancing this immune dysregulation. The magnitude of the viral load is important to determine the severity of the disease, against which, however, we are able to elicit a robust humoral and T-cell mediated response^19,21.  Altogether, these findings establish strong premises for the set up of efficient therapies and vaccines.