Approach to a highly-virulent emerging viral epidemic: A thought experiment and literature review

Mohamed Amgad1, Yousef A Fouad2, Maha AT Elsebaie2

1. Department of Biomedical Informatics, Emory University, Atlanta, GA, USA 2. Faculty of Medicine, Ain Shams University, Cairo, Egypt

Corresponding Author: Mohamed Amgad1 Email address: mtageld@emory.edu

1

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

Table of contents ____________________________________________________________________________

Abstract 3

Introduction 3

Survey Methodology 4

Thematic analysis 4 1a. Thematic analysis of key findings 4 1b. Potential causes of tissue damage and mortality 5

Table 2: Potential mediators of tissue damage and lethality. 5 1c. Possible viral virulence and immune evasion tactics 6 1d. The peculiar finding of lymphopenia 10

Table 4: Potential mechanisms behind lymphopenia and immune suppression. 10 1e. Why serum is non-protective 11

Table 5: Potential mechanisms why serum is non-protective. 11

2. Interrogation of viral biology 12 2a. Detailed clinical workup 12 2b. In-depth assessment of immune response 13 2c. Determination of virus structure and classification 15 2d. Investigating viral tissue tropism and host binding targets 16 2e. Investigating viral antigenic determinants and natural antibody targets 16 2f. Establishment of in-vitro and in-vivo experimental models 17

3. Management and vaccination strategy 17 3a. Non-vaccine management strategies 17 3b. Dynamics of infection spread and herd immunity threshold 18 3c. Vaccine development strategy 18 3d. Identification of peptide/subunit vaccine candidates 20 3e. Testing vaccine candidates 20

4. Acknowledgements 21

5. References 21

2

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

Abstract

What are the immunological facets of a highly successful viral epidemic, and what are likely successful strategies that can be used to counter its spread? Unlike many challenges in biology and public policy, viral epidemics are unique in that they require swift response, and quick application of existing knowledge to infer the underlying biology of a new pathological agent. This essay contextualizes the experience and findings from the viral immunology literature to respond to a hypothetical (yet likely) emerging viral epidemic. It begins with a review of the causes of some defining features of highly virulent viral epidemics, including causes of mortality, viral virulence, and immune evasion and suppression tactics. We provide an overview of lines of investigation to characterize emerging viral epidemics, including a brief survey of clinical and biological assays for immune surveillance and in-depth interrogation of viral biology. Finally, we provide a broad overview of management and vaccine development response strategies.

Introduction

This essay provides a bird’s eye view of various approaches to investigate and respond to a new viral epidemic. To this end, let’s assume that we are responsible for determining the national research budget and assigning funding priorities; what factors should we consider to determine which basic, translational and clinical research approaches receive our support? A thought experiment, in the form of a case study, is used to provide clearer context, and the background literature in virology and immunology is surveyed for factors that explain different findings or provide context to disparate management and research strategies. The thought experiment is summarized as follows:

A new pathogen is detected after an unknown tube in the back of a research lab freezer spills. After a few weeks, many people are infected. Patients who become infected have difficulty breathing. After about one year patients undergo complete respiratory failure and die. In less than 1% of patients there is clearance of the pathogen within about a month. One year after the pathogen is discovered there are cases in every country around the world. The pathogen is slowly spreading and it is estimated in the next 10 years 95% of the population will be infected. All infected patients have a detectable B-cell response to the pathogen. Serum from infected patients does not protect those uninfected. Also, serum from the patients that control the virus is non-protective. All patients mount a T-cell response to the virus in the initial stages, however this declines within about 1 month. In the patients that clear the disease, the T-cell response remains strong until the virus clears. T and B cells are not infected by the pathogen.

The case study highlights some key findings that have been reported in some of the most virulent viral epidemics in recent years like SARS-CoV, H5N1 Influenza, and Ebola. We begin our analysis by a qualitative, broad hypothesis of what might be the the underlying explanation for some of the observed features and what these might imply and how they affect our approach.

3

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

Survey Methodology

The literature was surveyed using PubMed and Google Scholar search queries. References of included articles and reviews were also surveyed for relevant content.

Thematic analysis

1a. Thematic analysis of key findings

The virus appears to be airborne, primarily affecting the respiratory system (Table 1). Its success seems to be attributed to direct or indirect tissue damage, cell-mediated immune suppression, and inadequacy of humoral immune response. Each of these findings warrants careful investigation, although of particular interest is the finding of lymphopenia, as this is not only an uncommon finding in typical viral infections (which generally cause lymphocytosis [1]), but is probably a key contributor to viral persistence and lethality. Moreover, research into the genetic and immune profile in viremic non-progressors and survivors should help elucidate natural mechanisms of viral protection and hence better inform drug and vaccine development [2,3].

Table 1: Thematic analysis of clinical and laboratory findings.

Finding(s) Likely explanation Likely implications Literature

- Infected individuals have difficulty breathing. After one year, death from respiratory failure . - After one year, there are cases in every country

- This virus is likely airborne, transmitted via droplets such as from coughing. The virus primarily affects the respiratory system. - Virus has high virulence, and has evades immune system (Table 3).

- Droplet infection control measures important, including awareness campaigns (wearing masks, avoiding crowds), strict quarantine and travel restriction. - Research needs to focus on the virus’s likely primary site, the lungs and airways.

[4]

- In <1% of patients there is clearance within a month. - Serum from survivors is non-protective.

One or a combination of: 1- different viral strains/phenotypes, 2- specific genetic polymorphisms to viral receptors or HLA alleles, 3- favorable cytokine response.

The genetic and immune profile in these survivors, including a detailed account of their immune cell populations and cytokine profile, should receive significant attention and funding.

[2,3,5]

All infected patients have a detectable B-cell response

As is typical of many viral infections, patients develop a short IgM response, followed by a longer-lasting IgG response against specific viral proteins.

- Specific anti-viral IgM and IgG can be used for diagnosis of infection. - The specific viral proteins targeted by the antibodies are potential vaccination and/or drug targeting candidates.

[6,7]

- All patients mount T cell response in initial stages, declines within a month.

Decline in T-cells could result from less production, immune exhaustion, apoptosis, or local pooling of lymphocytes in lungs or lymph nodes. This is

- Modulation of aberrant/suppressed cytokine response may help control the disease. - Targeting of apoptosis and immune exhaustion pathways may be relevant for

[2,3,5,8–10]

4

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

- In survivors, T cell response remains till virus clears. - T and B cells not infected.

discussed in Table 4.

therapeutic targeting. - Research needs to examine bone marrow (for suppression) and lymph nodes (for sequestration of T-cells or T-cell subsets).

Serum from infected patients does not protect those uninfected.

Antibody-mediated immunity plays a secondary role in protection. More potential explanations discussed in Table 5.

- Antibody titre should be measured as a confounder. - Work should be focused on vaccine and cytokine response targeting as opposed to direct targeting of viral components.

[5,9]

1b. Potential causes of tissue damage and mortality

Viral mortality is likely attributed to damage to respiratory tissue, leading to hypoxemia and respiratory failure (Table 2). Some of this damage may be caused directly by the virus replication cycle through infection and lysis of cells [4]. Radiologic findings from chest CT-scans can confirm viral pattern of pneumonia in infected victims, and the presence of virus or viral particles in sputum or bronchoalveolar lavage would provide stronger confirmation. Like many respiratory viral infections, the weakened immune system opens the door for opportunistic bacterial and fungal infections, leading to compound infections that resist treatment with single antibiotics or antifungal agents [11]. Some additional damage may be caused by induction of noxious substances or due to direct cytopathic effect by some viral proteins [12,13]. Moreover, the observed pattern of lymphopenia suggests cytokine-mediated damage also plays an important role (“cytokine storm”), caused by a vicious cycle of T-cell recruitment, local inflammation, neutrophil and macrophage recruitment, degranulation, and release of cytopathic substances like myeloperoxidase, collagenase, gelatinase C, etc. [8,13–15].

Table 2: Potential mediators of tissue damage and lethality.

Mediator Potential mechanisms Literature

Opportunistic infections

Opportunistic infections can result from the lymphopenia and immune suppression, such as what happens in AIDS after prolonged HIV infection. Mycobacterial infections (like tuberculosis) or fungal infections may be the immediate causes of death.

[11,13,16,17]

Extreme production of proinflammatory cytokines (“cytokine storm”)

- A vicious cycle of activated T-cell recruitment and release of attack molecules like perforin, leading to Diffuse Alveolar Damage (DAD) (SARS) - Infected pneumocytes secrete IL-1β, IL-6, IL-8, IP-10, MCP-1 which recruit and activate macrophages. These secrete more cyto/chemokines like MIG, IP-10, and CXCR-3 causing T-cell recruitment and tissue damage (SARS)

[8,13–15]

Direct cytopathic effect

It is possible that viral proteins have a direct cytopathic effect, similar to the EBOV GP protein. (EBV)

[12,13]

Noxious substances Induction of nitric oxide production, resulting in cell apoptosis or necrosis. (EBV)

[13]

5

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

Ischemic necrosis of tissue

Activation of extrinsic coagulation pathway, leading to fibrin thrombi formation and coagulation. (EBV)

[12,13]

1c. Possible viral virulence and immune evasion tactics

Viruses employ a huge array of mechanisms to increase their virulence and evade the host immune system. Some viruses, including CMV, HIV, HCV, Influenza and HSV employ multiple mechanisms to antagonise various arms of the immune system, while others employ a more restricted set of virulence tactics [8,18–22]. Virtually all arms of the immune system may be antagonized by viruses, including innate immunity and both types of adaptive immunity (humoral and cell-mediated). Given the rapid rate of progression and spread of the virus in this case study, we may assume that it employs multiple virulence mechanisms (Table 3). A very basic, yet extremely effective, modulator of immune evasion is high antigenic variation through antigenic drift and, occasionally, antigenic shift [18,22]. Antigenic drift is the simple evolutionary selective pressure caused by error-prone viral replication inside host cells, causing a heterogeneous viral population, a subset of which evades recognition by humoral and cell-mediated immune responses. Antigenic shift occurs when multiple viral strains exchange RNA inside a common host, a process which caused the emergence of the H1N1 Flu strain responsible for the 2009 outbreak, and quite possibly one of the contributors to the viral strain in this case study [23]; the origin of the samples inside the broken container are unknown and may have been from wild animal species like swine to be used to microbiological or ecological research.

One of the most important defenses against viral infections are interferons. The interferon system is activated by a convergent set of signals, including activation of pattern recognition signals like endosomal and plasma membrane toll-like receptors (TLR), intracellular/intranuclear DNA sensing pathway (STING) and RIG-I-like viral dsRNA sensing pathway, among others [5,15,19,24]. Various TLRs recognize different types of viruses; endosomal TLR-3, TLR-7 and TLR-9 recognize dsRNA, ssRNA and CpG DNA, respectively. We will be performing various structural experiments to determine the type of genomic material our virus contains (and hence its family), which would determine which of these pathways is potentially activated and/or antagonized by the virus. Viruses have been reported to block or antagonize all of these innate sensing pathways, hence reducing NFkB and IRF3/7 mediated production of interferon by virally-infected cells and, most importantly, plasmacytoid dendritic cells [25,26]. Viruses employ other means of blocking type I and type II interferon production, including antagonism to dendritic cell functions (through direct infection or induction of secretion of inhibitors like TGF𝛽), reduction of CD4+ T-cell activation, and expression of INF𝛾 that act as “interferon sinks” to prevent its binding with receptors [18]. A more direct way of inhibiting interferon response is achieved by viruses like HCV, perhaps the best studied model in this regard due to the pivotal role of INF therapy in combating HCV-mediated liver cirrhosis [19,27]. Viruses like HCV and HCMV are known to directly block various element of the interferon response pathway, including blockade of STAT1 phosphorylation and nuclear translocation, inhibition of protein

6

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

kinase R, blockade of 2-5(A) synthetase activation and induction of secretion of Jak/STAT inhibitors, among other mechanisms [18,19]. Viruses like HSV, VV, CMV and EBV (among others) inhibit the inflammatory response and complement systems, thus preventing critical early events required for well-orchestrated immune activation [28–31].

Table 3: Potential mechanisms of viral immune evasion or antagonism.

Mechanism Explanation Literature

Antigenic variation

Constant changing of the envelope or other epitopes, resulting in reduced or diminished recognition by antibodies. Two processes can cause this:

● Antigenic drift - successive point mutations leading to lost recognition by antibodies or CTL (“CTL escape”) (influenza, HIV, HCV, LCMV)

● Antigenic shift - RNA exchange between viral strains in a secondary host (such as with the 2009 H1N1 outbreak).

[18,22,32]

Inhibition of antibody and complement systems

Mechanisms include: ● Expressing inactivating Fc receptors on infected cells (HSV, CMV) ● Expressing inactivating complement receptors on infected cells (HSV) ● Expression of complement inhibitory proteins (VV)

[17,28–30]

Inhibition of inflammatory response

This leads to reduced availability of immune cells at sites of infection. Mechanisms include:

● Secretion of soluble cytokines receptors, including IL-1, TNF, INF𝛾. These bind soluble cytokines (“cytokine sink”) (VV, MCV)

● Inhibition of expression of adhesion molecules like LFA-3 and ICAM-1, hence blocking adhesion of CTL to infected cells (EBV)

● Encoding TLR-like molecules that compete with endosomal TLR receptors and prevent NF𝜅B pathway activation. (VV)

[17,28–30]

Interference with MHC class I presentation of viral peptides

This would cause a reduction in CTL mediated killing of infected cells, and could be achieved through a variety of mechanisms:

● Downregulation of MHC class I presentation pathway proteins like TAP2, LMP2/7 (AdV-12).

● Reduced proteasomal degradation though viral protein structural changes (murine leukemia virus, EBV, HCMV)

● Reduced TAP-mediated transport to ER by blockade (HCMV) or binding-site competition (HSV, AdV).

● Intracellular retention of MHC class I molecules in the ER (HCMV, by formation of a complex with β2m) or Golgi (VZV), inhibiting MHC class I terminal glycosylation (AdV), or tethering to clathrin-coated pits (HIV-1).

● Lysosome-mediated degradation of MHC class I molecules (murine CMV, HHV-7)

[17,28–30]

Antagonism to dendritic cell functions

This would reduce antigen presentation and secretion of key cytokines (like IL-2, IL-12, type I INF), resulting in a suboptimal adaptive immune response. Mechanisms include:

● Direct infection of immature DCs (MCMV, HCMV, HIV-1) ● Secretion if TGFβ, blocking maturation of DCs (HCMV)

[18,33]

Downregulatio This would prevent antigen presentation by DCs and macrophages to TH cells. [17,18]

7

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

n of MHC class II expression

Various mechanisms are involved: ● STAT or JAK protein downregulation/degradation, resulting in lack of

IFN𝛾-mediated induction (HCMV, VZV) ● Soluble inhibitors (other than IL-10/TGFβ), HCMV. ● Degradation of MHC class II pathway components like HLA-DM-α and

HLA-DR-α (HCMV) ● Prevention of AP-1 dependent clathrin-mediated transport of antigens

(BPV).

Evasion of NK cell-mediated apoptosis

Possible mechanisms include: ● Secretion of MHC class I heavy-chain homologous that bind to inhibitory

receptors on NK cells (HCMV, MCMV) ● Downregulation of expression of NKG2D activating ligands, resulting in

less NK cell activation (MCMV) ● Upregulation of non-classical MHC class I molecule HLA-E, an inhibitor

of NK cells.

[17,18]

Reduced CD4+ cell activation

This causes overall reduction of adaptive immune response. Mechanisms include: ● Antagonism to dendritic cell function (see above) ● Downregulation of MHC class II expression (see above) ● Downregulation of CD4 by enhanced degradation or retention in the ER

(HIV) ● Downregulation of the T-cell receptor transcription (HHV)

[17,18]

Antagonism to interferon production

Various mechanisms are implicated, including: ● Less pDC activation, hence less type I INF production (see above) ● Reduced CD4+ activation (see above) ● Expression of IFN𝛾 receptor homologues, acting as viral decoys that bind

soluble interferon, preventing its action. ● Blockade of innate viral sensing mechanisms like Toll-like, RIG-like and

cGAS/STING pathways (see below) The reduced production results in less blockade of viral transcription, less cleavage of viral RNA, less MHC class I and II expression, less inhibition of viral uncoating and release by molecules like tetherin, and overall blunting of innate immune response.

[13,15,17,19,24]

Antagonism to interferon pathway

Effects would be same as when interferon production is blocked, but mechanism is different:

● Induction of SOCS2, a negative regulator of Jak/STAT signaling (HCV) ● Blockade of STAT1 phosphorylation and nuclear translocation (HCV,

DenV,RSV) ● Inhibition of protein kinase R activation (HCV, HCMV) ● Blockage of 2-5(A) Synthetase activation (HCV) ● Encoding 2’-O methyltransferases to cap viral RNA and prevent

IFIT1-mediated inhibition of translation (MHV, SARS, VACV, WNV)

[13,15,19]

Blockade of RIG-I-like viral dsRNA sensing pathway

Almost all elements of this pathway have been reported to be antagonized, including:

● Blockade of PP1𝛼/𝛾 dephosphorylation of MDA5 (MCV) ● Blockade of TRIM25/RIPLET mediated activation of RIG-I (Influenza,

HCV) ● Direct inhibition of MDA5/RIG-I (CBV3, PV, EV-D71) ● Blockade of RIG-I interaction with MAVS (DenV, RSV)

[15,19,25,26]

8

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

● Direct inhibition of MAVS (HCV, PRRSV, CVB) ● Inhibition of TRAF-3 complex (HIV-1) ● Inhibition of IRF3/7 (EV-D68, SARS, MERS, RSV) ● Inhibition of NF𝜅B/I𝜅B complex (MCV)

Blockade of viral DNA sensing pathway (STING)

Several mechanisms are implicated: ● Inhibition of cGAS cytoplasmic recognition of viral DNA (HIV-1) ● Inhibition of IFI16 nuclear recognition of viral DNA (HSV-1) ● Inhibition of STING (HPV-18, AdV, KSHV, DENV, SARS, HCV)

[25,34]

Blockade of TLR pathway

Potential mechanisms include: ● Inhibition of MyD88 (VACV, HCV) ● Inhibition of TRIF (VACV, HCV) ● Inhibition of IRAK-2 (VACV)

[25,35]

Infection of immune privileged sites

Despite the fact that the cause of death is respiratory failure, the virus may persist (and be shielded) in immunologically privileged sites like the CNS (blood brain barrier) and kidney (glomerular basement membrane). (HIV-1, polyomaviruses)

[33,36]

Lymphopenia and/or immune suppression

Discussed in Table 4. Lymphopenia results in an ineffective immune response due to quantitative decrease in B- and T- cell availability to fight the infection. (HIV, SARS, Influenza)

See Table 4

To achieve this, viruses have evolved mechanisms including: induction of expression of inactivating Fc or complement receptors; expression of soluble complement inhibitory proteins or cytokine receptors; inhibition of expression of adhesion molecules like LFA-3 and ICAM-1, preventing attachment to and killing by CTL cells. Viruses antagonize another key facet of the immune system, cell-mediated immunity, using a diverse repertoire of evasion tactics. One of the key events involved in CTL-mediated killing is recognition of foreign viral peptides that have been cleaved by the proteasome and presented on MHC class I molecules on the plasma membrane of virally-infected cells [18]. Adenoviruses, HSV, CMV and other viral strains antagonize this system by downregulation of viral proteasomal degradation, reduced MHC class I expression, intracellular retention of MHC class I molecules or, quite commonly, inhibition of TAP-mediated transport of viral peptides to the ER. MHC class II presentation of viral peptides on antigen-presenting cells like dendritic cells is also antagonized by some viruses, especially CMV, through blockade of IFN-𝛾 response and degradation of MHC class II pathway components, among other mechanisms. This, coupled with the inhibition of dendritic cells, dampening of interferon response, and downregulation of T-cell receptors and co-stimulatory molecules, results in reduced CD4+ helper T-cell activation, and subsequent blunting of cell-mediated and humoral immunity [18,33]. Some viruses like HCMV and VZV have even evolved mechanisms to evade NK cell-mediated killing, which normally would be triggered by downregulation of MHC class I molecules (hence reduced inhibition of NK cells) and expression of stress-response signals (triggering activating receptors like NKG2D on NK cells) [18].

9

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

1d. The peculiar finding of lymphopenia

Despite immune evasion tactics, a well-functioning immune system is often capable of compensating the loss of some of its components. The virus in this thought experiment, much like SARS-CoV, EBOV and H5N1 Influenza, induces an immunosuppression phenotype without direct infection of lymphocytes (Table 4) [9,14,16,37]. The obvious first guess is that the virus directly affects production from the bone marrow by direct infection of hematopoietic precursors or by indirect cytokine-mediated marrow suppression [38]. Reduction of neutrophil count and a hypocellular bone marrow aspirate would confirm this. Excluding bone marrow pathology, apoptosis of lymphocytes is the likely immediate mechanism of lymphocyte depletion [16,39–41]. Apoptosis may be triggered by a range of mechanisms. Interferon, paradoxically, is well-known for induction of immunopathology, both acutely and on chronic exposure [5,21,42]. Acute interferonopathies result from upregulation of death receptors and apoptosis-inducing ligands on monocytes, airway epithelial cells, and lymphocytes, while chronic type I interferonopathy results from induction of immunosuppressive cytokines like IL-10 and overexpression of PD-1 and PDL-1. Generally-speaking, prolongation of infection and viral chronicity result in immune exhaustion, mediated through receptors such as PD-1 or CTLA-4 [14,21,31,43,44]. One cannot exclude other plausible mechanisms of lymphopenia like glucocorticoids (as a physiological stress response or therapeutically), altered lymphocyte differentiation, or sequestration in lymphoid tissue [20,44–47].

Table 4: Potential mechanisms behind lymphopenia and immune suppression.

Mechanism Explanation Literature

Acute type I IFN mediated immunopathology

- Upregulation of TRAIL (TNF-related apoptosis-inducing ligand) on inflammatory monocytes. - Upregulation of DR5, a TRAIL receptor, inducing airway epithelial cell apoptosis. - Direct receptor-mediated lymphopenia (IFNAR receptor on B and T cells).

[5,21,42]

Chronic type I IFN mediated immunopathology

- Induction of IL-10 and other immunosuppressive molecules. - Induction of death ligand PD-1 on macrophages and DCs. These bind to PDL-1 receptors on T cells, leading to immune exhaustion and suppression.

[5]

Apoptosis of T-cells

- Induction of IL-12p40 in the peribronchial lymph nodes, causing plasmacytoid DC’s to express FasL and induce apoptosis in CTL cells. (H5N1) - Direction interaction and induction of apoptosis on T-cells. (EBOV, SARS)

[16,39–41]

Immune exhaustion

- Overproduction of inflammatory cytokines like type I IFN, IL-6, IL-27. (HIV, Ebola, H1N1). - Immunosuppressive cytokines like IL10 and TGFβ, or IL-10 homologues. (HCMV,

[14,17,21,31,43,44]

10

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

EBV). - Induction of expression of CTL inhibitory receptors like PD-1, CTLA-4, TIGIT, TIM-3, etc.

Bone marrow suppression

If the virus also targets the bone-marrow, through direct infection of hematopoietic progenitors (Parvovirus B19) or indirectly through cytokines (CMV, EBV, VZV, HCV, Dengue).

[38,48]

Glucocorticoid mediated

- Direct or indirect activation (through stress response) of hypothalamic-pituitary-adrenal access. Glucocorticoids induce lymphopenia and downregulate chemokines. - Management strategies involving combined high-dose steroids (to reduce inflammatory damage) under cover of antivirals/antibiotics may exacerbate this.

[18]

Altered T-cell or ILC differentiation

- Inhibited differentiation of naive T-cells to memory cells (central/effector), or terminal effectors. - Promotion of differentiation into the Treg phenotype, resulting in immune suppression.

[20,45,47]

Sequestration of lymphocytes

It is theoretically plausible that the virus mimics the drug fingolimod, by modulating the S1PR receptor, sequestering lymphocytes in lymph nodes.

[46]

1e. Why serum is non-protective

A robust adaptive immune response requires the cooperation of the humoral and cell-mediated arms. After initial encounter with the antigen, B-cells undergo affinity maturation and class switching in lymph node germinal centers, with help from dendritic cells and TFH cells that recognize the same epitopes as B-cells (a process called “linked recognition”) [17]. Failure of this process may in part be responsible for non-protectiveness of serum, since the quantity, affinity and avidity of antibody response in infected individuals (perhaps even in survivors) may be inadequate (Table 5). The importance of humoral immune response also varies between different viral infections due to structural factors (viral peptide antigenicity), factors related to the viral target tissues and replication cycle (latency, infection of immune privileged sites), or factors related to viral virulence (evasion of humoral immune response). In our case, it is likely that the cell-mediated immune response is of greater importance and high concentrations of antibody are needed for passive transfer of immunity.

Table 5: Potential mechanisms why serum is non-protective.

Mechanism Explanation Literature

Antibody alone is non-protective.

- Much of the viral cycle is spent intracellularly, sequestered from extracellular antibodies (integration and latency). - Innate immunity (TLR/TNF/RLR/cGAS), NK cell mediated immunity, and CD8+ mediated adaptive immunity may be more important.

[49]

Failure of For an effective immune response to take place, antigen-specific T- and B- cells [17]

11

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

“linked recognition” and affinity maturation

much cooperate, with resulting affinity maturation and class switching. This means that the viral epitope recognized by TCR (when bound to MHC class II) must be contained within the epitope recognized by the BCR. In RSV infection, for example, this requirement is not met, leading to failure of induction of effective antibody response.

Variable antibody titre in infected/survivor individuals.

- Antibody levels are variable between survivors and individuals and at various stages of infection. This may be an uncontrolled confounder. - Very high antibody titres may be needed to achieve protection due to:

● Low antibody affinity ● Poor delivery to sites of viral sequestration. Virus may persist in

immune-privileged sites. ● Direct binding by soluble viral antigens, acting as decoy to prevent

binding to virus. (Ebola virus sGP has been hypothesized to serve this role, for example).

● Viral envelope structural features that block or minimize antibody access, such as variable loops and glycosylation.

[13,50]

High Antigenic variation

The antigenic variation may be so high such that the antibodies produced in one individual are unlikely to bind to the virus in another infected individual. This could result from high mutation rate in viral RNA/DNA (“antigenic escape”).

[18,22,32]

Damage mostly by immune response

Even with modest antibody levels, survivors may be protected by genetically-determined factors that favorably reduce cytokine storm or other contributors to immunosuppression and tissue damage.

See Table 4.

2. Interrogation of viral biology

2a. Detailed clinical workup

One of the challenges in tackling respiratory viral outbreaks is vague clinical symptoms and overlap with clinical picture of common cold, seasonal Flu, etc. We begin our analysis by a detailed clinical workup in infected individuals at various stages (early, late, and when applicable, convalescence) (Table 6).

Table 6: A non-exhaustive list of relatively simple clinical and laboratory investigations. [6,17,51–53].

Test/Modality Explanation

CBC + differential blood count

Are there findings indicative of bone marrow suppression or, for example, RBC hemolysis? What does differential WBC count show, beside the lymphopenia? Are blood components including neutrophils, monocytes and RBCs affected?

“Standard” Flow cytometry and cell sorting (FCS)

FCS of common markers and cell subpopulations, including Pan-leukocyte (CD45+), Pan-T cells (CD3+), TH cells (CD3+CD4+), CTL cells (CD3+CD8+), B cells (CD3-CD19+), NK cells (CD16+CD56+), and Monocytes (CD3-CD14+). Memory T cells (CD45RO+ and CD4+/CD8+) and naive T cells (CD45RA+CD62L+ and CD4+/CD8+) may be assessed.

Serum albumin and

Determined using protein electrophoresis (albumin, overall immunoglobulin), single radial immunodiffusion/nephelometry (total Ig levels, Ig classes, and IgG subclasses). Albumin

12

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

immunoglobulin composition

levels probe synthetic liver function, while Ig classes determine type of antiviral immune response.

Anti-viral Ig levels

Specific anti-viral IgG and IgM levels for disease screening and response monitoring.

Virus detection and quantification

Specific detection of viral DNA/RNA by PCR, and quantification of viral load by RT-PCR. This step would only be possible after sequence of viral DNA/RNA is established.

Blood and urine chemistry workup

Determination of key parameters like serum electrolytes, BUN, creatinine, liver function enzymes etc is critical to understand associations with viral virulence and lethality.

Comprehensive radiologic assessment

Chest CT scan can reveal immediate cause of death and show different patterns associated with viral infection (primary) or bacterial/fungal infections (opportunistic). CT-scanning of head, abdomen, and other anatomic locations can help reveal anatomic locations infected. MRI and other imaging modalities like PET-CT may also be needed.

Bone marrow biopsy

This would help determine whether lymphopenia results from bone marrow suppression either directly (hypocellularity, apoptosis, etc) or indirectly (just hypocellularity).

Lumbar puncture Whether the virus infection and homing homing to the CNS is a contributor to viral virulence.

Not only would this help us form a detailed account of the natural history of infection in non-survivors in comparison to survivors, but it would also be critical in establishing reliable diagnostic (high specificity) and screening (high sensitivity) criteria for the disease [54]. Of particular interest is the investigation of anatomical sites potentially targeted by viral infection (lung, liver, bone marrow, CNS, etc), detailed analysis lymphocyte subpopulations using flow cytometry, and blood/urine chemistry workup. Studies should investigate the sensitivity and specificity of antiviral monoclonal antibodies (IgM or IgG) and how they correspond to various stages of the infection and natural disease history. Importantly, we should investigate the presence of a “window period” (akin to HBV infection) where the virus is undetectable through any antibody response, as this has clear implications on diagnostic and quarantine measures [55]. After preliminary basic studies determine the sequence of viral genetic material, the sensitivity and specificity of viral PCR and quantitative RT-PCR diagnostic assays should also be determined. Additionally, efforts should be made to develop and validate rapid diagnostic test (RDT) assays for use in low-income, remote, and high clinical flow settings [7].

2b. In-depth assessment of immune response

In addition to the clinical tests above, which could potentially be performed in a large fraction of the infected population, we should perform in-depth analysis of immune system functions in smaller subsets of the infected population as well as the survivors (Table 7). We begin by detailed analysis of the cytokine profile in infected individuals, healthy controls, and disease survivors. Like the clinical workup, this analysis should be done at various stages (early, late,

13

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

convalescence), and at various anatomic sites (especially the respiratory system, through bronchoalveolar lavage or lung biopsy/autopsy).

Table 7: Assays for in-depth assessment of immune function

Function Explanation Literature

Cytokine profile (in-vivo)

Plasma level of cytokines like INF𝛼/𝛽/𝛾, IL-2, TGF-β, IL-10, IL-12, IL-17, TNF-𝛼, among others. Typically measured using Sandwich ELISA.

[17]

TH response (in-vivo)

Delayed-type hypersensitivity test (DTH) involving intradermal injection of viral antigen and measuring erythema or induration.

[56,57]

Antigen-specific T-cell quantification

- FCS using MHC class I Peptide:MHC tetramers (standard) using known viral epitopes - FCS using MHC class II Peptide:MHC tetramers (non-standard, difficult) - TCR CD3 PCR (qualitative measure of skewing of repertoire; inferior to MHC tetramers)

[17,56,57]

Lymphocyte proliferation

- Ex-vivo stimulation/proliferation of T and B cells by a polyclonal mitogen (PWM or S aureus) - Viral antigen-specific T-cell lymphocyte proliferation assay

[17,57]

TH and CTL cytokine production

- Either mitogen (non-specific) or specific antigen induced in-vitro cytokine production - ELISA assay for Bulk cytokine production - ELISPOT for cell-level cytokine production (IFN-𝛾 capture) - Intracellular cytokine detection using FCM (cytokine FCM) - Intracellular cytokine mRNA level by RT-PCR (not well established, but sometimes used)

[17,57]

Cytotoxicity assays

Chromium release assay for CTL (against antigen-expressing target cells) and for NK cells (non-specific lysis of NK-sensitive tumor target cells).

[56,57]

Neutrophil function

- Respiratory burst activity by NBT dye reduction test or chemiluminescence - Phagocytic, chemotactic, and bactericidal function (more complex tests)

[57]

Complement activity

- CH50 assay of total complement activity - Immunochemical quantification of complement by nephelometry or immunodiffusion

[57]

Cytokines are detected using standard antigen capture assays like Sandwich ELISA, or when quantitative results are needed, using competitive inhibition assays. Of particular interest are type I and II interferons, TNF𝛼, IL-2, IL-4, IL-17 and immune modulators such as TGFβ and IL-10 [17,51,55–57]. Correlational analysis between viral load (identified through antibody or viral PCR) and cytokine profile, can be particularly useful. For example, it was shown that in the context of HIV infection, viremic non-progressors actually shut down type I INF production, with a resultant reduction in immunosuppression [3]. More comprehensive sequence comparative

14

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

studies assessing HLA alleles and genetic polymorphisms in infected individuals, survivors, and viremic non-progressors can also be conducted [2].

Limited almost only by funding constraints, all aspects of humoral and cell-mediated immunity should be tested, as these will prove instrumental in understanding viral biology, the mechanism behind immunosuppression, and why some people survive the infection. A crude in-vivo DTH assay can provide a general measure of cellular immunity, however the specificity and standardization of DTH assays has (despite their ubiquitous use) been called into question [2,56,57]. Many of the assays mentioned above require the knowledge and isolation of specific viral antigens/epitopes that are immunogenic in nature. This is probably a key bottleneck, and may force us to rely on non-specific or polyclonal lymphocyte proliferation and cytokine production assays at the earlier stages of research into the viral epidemic. Once the virus structure is known, it becomes possible to quantify and isolate antigen-specific CTL (and, with technical difficulty, TH) using MHC:peptide tetramers. Purified viral antigens would also allow the assessment of proliferation and cytokine production by antigen-specific using assays like IFN-𝛾 capture ELISPOT assays [17,56,57].

2c. Determination of virus structure and classification

In-depth understanding of viral biology and virulence requires knowledge of viral structure and genomic sequence. There are various systems for classifying viruses, the most established of which are “classical” ICTV classification and the Baltimore classification [58–60]. Viral particles are isolated from a variety of potential sources like the sputum or victims, infected tissue from autopsies, or propagated using mammalian cell culture. Viral genome is then sequenced, facilitating the classification of virus into one of the seven key genomic classes. Techniques like random-primed PCR and comprehensive DNA microarrays can be very helpful in sequencing the new viral agent, given the limited knowledge we have at the start of the outbreak [61]. The Baltimore classification of viruses relies on the fact that viral mRNA synthesis occurs using different intermediary mechanisms depending on the original viral genetic material. This helped provide a general grouping of viruses based on their infection cycle (transcription/replication), and hence their biology. While the genomic classification is very useful, it only provides a crude picture of viral biology, and it is necessary to perform further in-depth analysis of other structural aspects of viral biology. This includes Cryo-EM visualization of 3D assembled viral structure as well as the structure of purified protein components [62]. Any abnormal viral peptides detected in plasma of victims (eg by mass spectrometry) should be purified and characterized thoroughly, as these might prove to be interesting biomarkers or indicators of viral load in a similar fashion to the HBsAg in HBV infection [63]. The amino acid of various viral proteins should be sequenced and efforts to determine the 3D crystal structure should be initiated [64]. In the meantime, while crystallization efforts are underway, computational approaches should be used to predict

15

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

approximate 3D structure, with a special focus on surface glycoproteins or other factors hypothesized to mediate viral attachment [65].

2d. Investigating viral tissue tropism and host binding targets

An in-depth analysis of natural viral tropism can be assessed by performing detailed autopsies of non-survivors, including testing for viral presence (or effects) in all key organ systems (respiratory, digestive, cardiovascular, nervous, musculoskeletal, etc) [66]. Gross examination of all systems can help determine immediate cause of death. Of particular interest is presence of arterial thrombi and venous emboli, especially pulmonary emboli. Histopathologic examination of these tissues is important including: H&E stain, for structural damage, necrosis, emboli, inflammatory microenvironment; special stains, depending on tissue; IHC for in-situ examination of apoptosis (Caspase-3), viral presence using antiviral antibodies, and markers of immune exhaustion (PD-1, CTLA-4); TUNEL assay for detection and quantification of apoptosis in various tissue [66,67].

Bioinformatics approaches should be deployed to identify likely binding partners to viral genetic material and protein [68]. Potential binding targets include host surface receptors, signalling/development proteins and transcription factors, as well as cytokines and chemokines. Homology analysis can help identify evolutionary equivalents of viral genes/proteins, and would thus be potentially useful in understanding viral biology. Moreover, such homologies can help guide our therapeutic and experimental approaches as we draw parallels from well-known viral infections.

2e. Investigating viral antigenic determinants and natural antibody targets

The identification of viral antigenic determinants is important for understanding viral biology, explaining why serum is non-protective, and determining potential targets for vaccine development. A simple approach would be to isolate the virus (eg. from infected tissue), dissociate and run it on an SDS-PAGE gel. These antigens are then transferred to a nitrocellulose and overlaid with serum from infected or convalescent individuals. Enzyme-linked anti-IgG antibody can then be used to detect which viral antigens are bound by naturally-formed antibody [17]. Similarly, antibodies can be attached on beads and immunoprecipitation can be used, followed by use of SDS-PAGE to determine viral fragments which interact with naturally-formed antibodies. Alternatively, various viral components can be purified and attached to an insoluble matrix. Affinity chromatography is then used using the various viral antigens to screen and purify binding antibodies. [69]

Determining the strength of antigen-antibody binding requires more complex procedures. Various viral antigens can be placed in a dialysis chamber, and equilibrium dialysis used to determine the affinity and avidity of naturally-formed antibodies using Scatchard analysis. This would help us determine if serum non-protectiveness stems from poor affinity (eg due to lack of

16

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

affinity maturation). A more advanced approach involves immobilization of the target viral antigens to special gold-plated surfaces and the use of a “biosensor” assay to quantitate ligand association and dissociation dynamics [17,70].

2f. Establishment of in-vitro and in-vivo experimental models

Based on the results of the viral tropism experiments, we should try to establish various human (and other mammalian) cell lines for in-vitro propagation of the virus. Several human cell line registries exist including, for instance, the 844 cell lines at the Japanese Collection of Research Bioresources, which may be systematically screened [71]. In-vitro propagation of the virus is critical not only for detailed studying of viral biology, but also for testing various antiviral assays. We should also try to establish in-vivo models for viral infection, so that we can: a) study natural disease history and biology in a controlled setting, b) develop and test various antiviral therapies and c) develop and test vaccine candidates. A mouse model is ideal in that it is cheap and easy to manipulate, although this may be limiting as it may not show the same clinical and laboratory findings as observed in human infection (this is the case for Ebola for example). If mice do not get infected in the same way, a humanized mouse model can be created by altering the virus in various ways, for example through a random mutagenesis and screening assay, until it infects the mouse. While this is not ideal, it often resembles the human infection, while maintaining the flexibility of mouse models (compared to non-human primates for example) [50,72].

The best confirmatory models are usually those based in non-human primates (NHP). These often resemble viral infection in humans the most, but are also the most expensive and difficult to manipulate [50,72]. We begin by checking if other mammalian species get infected by the virus (or its evolutionary homologous), and which ones show clinical and laboratory findings that resemble the signs and symptoms observed in humans. Of particular interest are “natural hosts” of the virus which despite having high viremia show no clinical signs or symptoms. Sooty mangabeys, the natural hosts for SIV infection, are extensively used in HIV research for this reason, and have been shown to survive long-standing SIV infection due to a combination of factors including: a) reduction of long-term interferon response; b) lack of chronic immune stimulation and hence prevention of immune exhaustion; c) SIV preferentially infecting terminal TH effectors and sparing TH central memory cells [3,73,74].

3. Management and vaccination strategy

3a. Non-vaccine management strategies

While vaccine efforts are underway, we should experiment with short-term solutions to control viral infections in infected individuals and reduce lethality. Depending on the virus classification, it may be possible to use some antiviral therapies (eg. antiretrovirals, acyclovir, etc) that are used in some currently-known viral infections. Since these drugs are already FDA approved, Phase II clinical trials can be immediately begun using various existing antiviral regimens. A more “risky”

17

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

line of investigation would be examining the efficacy and safety of blockers of interferon action as well as immune exhaustion antagonists like PD-1 and CTLA-4 blockers [75,76]. These risky investigations would be warranted if reasonable doubt exists, based on evidence from immune profiling in survivors, that interferon dysregulation and immune exhaustion were partially responsible for the immunosuppression phenotype. Depending on regulatory standards, such studies may be allowed to proceed directly as Phase II clinical trials (using existing PD-1/CTLA-4 blockers used in cancer immunotherapy, for example), or compelling evidence using NHP models may be required first.

While active immunization is the only option that allows for long-lasting immunity, we should also consider (and test the effectiveness of) passive immunization using monoclonal antibody for protection of case contacts or as a form of post-exposure prophylaxis. High purity monoclonal antibody can be obtained using B-cell hybridomas (fusion of myeloma cells with spleen cells from mouse immunized with viral antigens) or using systematic recombinant approaches like phage display libraries. ZMapp, a cocktail or targetes pure EBOV mAbs, for example, showed protective value despite weak or non-protectiveness of sera in EBOV passive immunization [50,77].

3b. Dynamics of infection spread and herd immunity threshold

One of the factors that can inform vaccine researchers and public health regulators when facing a new epidemic is the reproduction number (R0) and the herd immunity threshold, also known as the critical vaccination level (Vc) [78]. R0 is defined as the average number of susceptible individuals (i.e. unvaccinated and non-exposed) that one infected case can transmit the disease to over the course of its infectious period. R0 = 1 is a critical value, since it signifies an equilibrium where disease is stable. When R0 < 1, the disease is shrinking will die out on its own, while R0 > 1 indicates the disease is spreading. R0 varies widely between different infection (measles has R0=12-18, while Ebola has R0=2), but generally airborne or droplet-transmitted infections have a high R0 value. Our target is to bring down R0 below 1, which means that the critical fraction of population that needs to be vaccinated to stop disease spread Vc=1 - (1/R0). The calculation of R0 is empirically driven, and assumes that the population is homogeneous and freely mixing (i.e. does not take into account quarantine measures or small communities). It also assumes that the the vaccine offers 100% protection. Using these and other (more sophisticated) metrics, we can get a better idea of: 1. Whether a widespread vaccination effort is likely to succeed, and 2. How well we are doing, and the point at which we can consider the epidemic to be regressing. If the virus is obligately human (as are smallpox and poliovirus) then an effective vaccine could in fact eradicate the virus if the critical vaccination level is reached [79].

3c. Vaccine development strategy

There are numerous vaccine development strategies that have been tried, with variable success, in past viral infections [54]. “Classic” vaccines rely on inactivated viruses or on live-attenuated

18

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

viruses inducing natural infection in the host, while newer approaches include individual or combined viral subunits/proteins, DNA vaccines, and “vectorized” viral vaccines [17,50,54]. “Killed”/inactivated virus vaccines rely on viral inactivation by formalin, heat or irradiation. The advantage of course is that this technique is safer than live-attenuated vaccines, and usually triggers a good humoral response. Unfortunately, though, this also means that it is less immunogenic and is does not trigger strong cell-mediated response, since there is no presentation of viral antigens on MHC class I or II molecules [17]. Nonetheless, for viral infections where humoral immunity plays the major role, such as poliovirus and Hepatitis A, a killed vaccine provides adequate protection.

Whole-virus attenuation is typically achieved by propagating the wild type virus for many cycles on non-human (eg chicken or monkey) culture cells. The virus is selected for survival in these cell lines and acquires mutations through random antigenic drift that attenuate its virulence in human hosts. This type of vaccine usually triggers a strong humoral and cellular immune response, and is thus a very effective vaccine type that has been critical in preventing measles and VZV infections [54]. This type of vaccine, however, is not suitable for immunocompromised or pregnant individuals, since the prolonged time to viral clearance increases the chance of virus re-acquiring selective mutations that enable it to be virulent and potentially lethal in humans. Genetically-engineered live-attenuated viral vaccines, while more difficult to develop, circumvent this limitation. The virulence gene, determined using any of the approaches mentioned below, is deleted (eg. using restriction enzymes) or mutated (eg. using random or site-directed mutagenesis assays). These mutations are introduced in a controlled fashion, and are engineered to be extensive enough to make it extremely unlikely that the virus can re-acquire virulence.

The most famous viral vaccine consisting of a viral peptide is the HBsAg based Hepatitis B vaccine, produced by recombinant yeast [17,80]. This technique is more safe, due to absence of virulence, and has the added advantage of mass production. Its main disadvantage is the need for multiple doses and that key viral immunogenic peptides need to be known in advance. Although no vaccine for Ebola has been approved by the FDA yet, a Viral-Like Particle (VLP) containing EBOV VP40 and GP particles (optionally with other EBOV proteins) were shown to be highly immunogenic and to successfully trigger innate and adaptive immune responses [50]. Recombinant DNA vaccines are a relatively new technology, and so far there are no FDA-approved DNA vaccines for human use. Nonetheless, they have been shown in some settings to induce a strong humoral and cellular immunity, albeit while requiring multiple doses [54]. “Vectorized” viral vaccines involved the use of a non-virulent viral replicon (eg from VSV or adenovirus) to deliver antigenic peptides [54]. Due to the high flexibility of this technique, it has been the focus of many of the vaccine efforts in response to recent epidemics like SARS and EBOV, often showing very high success rates and induction of humoral and cell-mediated immunity. One factor to consider when choosing vectors is pre-existing immunity of the host; if the individual has strong immunity (eg to adenovirus) they are somewhat likely to reject an adenovirus-type replicon [81]. Another important consideration is the replication competence of

19

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

delivery vectors, as replication-competent vectors may pose safety issues in pregnant or immunocompromised individuals . Various preparations (oral, subcutaneous, intradermal, parenteral/intramuscular, aerosolized/intranasal) of the vaccine should be tested for immunogenicity and strength of resultant humoral and cellular response to a lethal viral challenge in animal models. Since this virus seems to primarily infect the respiratory tract, intranasal administration may prove to be a very effective way of triggering a strong immune response. For peptide vaccines, various adjuvants and doses should also be considered and tested for immunogenicity [17].

3d. Identification of peptide/subunit vaccine candidates

How can we identify virulent viral subunits/peptides so that are virulent (to be deleted/mutated) or immunogenic (to be expressed) when designing a recombinant live-attenuated, DNA, or vectorized vaccine? We may begin to answer this question by identifying viral components that bear some homology to known virulence genes/peptides, or for which there is a potential biological mechanism that mediates virulence. For example, a glycoprotein probably mediates attachment to the host cell. One may also predict vaccine candidates by their chemical properties; good vaccine candidates are typically large, complex peptides that are highly dissimilar to known human antigens [17]. A more direct approach would be to use in-vitro and in-vivo virulence screening assays; delete or mutate various viral components and introduce modified virus to cultured cells and quantify viral infectivity using a plaque assay or flow cytometry [82]. The same principle can be applied in-vivo by introduction to animal models. More complex techniques, called “reverse immunogenetics” allow identification of viral peptides that are displayed by MHC class I and II. After MHC molecules from infected tissue are isolated (eg using native PAGE assays), peptides are eluted from the MHC binding groove and highly-sensitive mass spectrometry is used to characterize the peptide fragments [17,83].

3e. Testing vaccine candidates

Potential vaccines are tested in-vivo by incolulation into a mouse or NHP host (a negative control is injected with an saline). After ~10 days, a challenge test is performed by injection with a lethal dose of the virus, and the animal is observed for clinical symptoms and signs, viral load by RT-PCR, viral shedding in sputum and bronchoalveolar secretions, FCS of lymphocyte subpopulations [17]. A crude in-vivo test for activation humoral and cell-mediated immunity involves transfer of serum or spleen cells to a syngeneic recipient, followed by challenge with a lethal dose of antigen. Quantification of viral level (RT-PCR) and antiviral immunoglobulin levels in serum (IgM, IgG) and in respiratory secretions (IgA) should be done after 10 days and after weeks of primary inoculation [17]. It is also important to monitor cell-mediated immune response by in-vivo testing (DTH) or MHC:peptide tetramer-FCS of antigen-specific T-cells [50,54]. Alternative and more flexible trial designs should be considered, although a classic

20

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

individually randomized design with placebo controls should be the default (to maximize statistical power and scientific value) [84,85]. 4. Acknowledgements

This work originated as an assignment in Emory University’s graduate Concepts of Immunology class, Fall 2018. The authors would thus like to deeply thank the course instructors for the inspiration and guidance. Special thanks to Dr. Haydn T. Kissick and Dr. Steven E. Bosinger for their support.

5. References 1. Riley LK, Rupert J. Evaluation of Patients with Leukocytosis. Am Fam Physician. 2015;92:

1004–1011. 2. Poropatich K, Sullivan DJ Jr. Human immunodeficiency virus type 1 long-term

non-progressors: the viral, genetic and immunological basis for disease non-progression. J Gen Virol. 2011;92: 247–268.

3. Rotger M, Dalmau J, Rauch A, McLaren P, Bosinger SE, Martinez R, et al. Comparative transcriptomics of extreme phenotypes of human HIV-1 infection and SIV infection in sooty mangabey and rhesus macaque. J Clin Invest. 2011;121: 2391–2400.

4. Kuchar E, Miśkiewicz K, Nitsch-Osuch A, Szenborn L. Pathophysiology of Clinical Symptoms in Acute Viral Respiratory Tract Infections. Advances in Experimental Medicine and Biology. 2015. pp. 25–38.

5. McNab F, Mayer-Barber K, Sher A, Wack A, O’Garra A. Type I interferons in infectious disease. Nat Rev Immunol. 2015;15: 87–103.

6. Cheever FS. General Principles of the Laboratory Diagnosis of Viral Infections. N Engl J Med. 1947;237: 584–590.

7. Amini A, Varsaneux O, Kelly H, Tang W, Chen W, Boeras DI, et al. Diagnostic accuracy of tests to detect hepatitis B surface antigen: a systematic review of the literature and meta-analysis. BMC Infect Dis. BioMed Central; 2017;17. doi:10.1186/s12879-017-2772-3

8. Chen J, Subbarao K. The Immunobiology of SARS*. Annu Rev Immunol. 2007;25: 443–472.

9. Rasmussen AL. Host Factors in Ebola Infection. Annu Rev Genomics Hum Genet. 2016;17: 333–351.

10. Leroy EM, Baize S, Volchkov VE, Fisher-Hoch SP, Georges-Courbot MC, Lansoud-Soukate J, et al. Human asymptomatic Ebola infection and strong inflammatory response. Lancet. 2000;355: 2210–2215.

11. Limper AH, Adenis A, Le T, Harrison TS. Fungal infections in HIV/AIDS. Lancet Infect Dis. 2017;17: e334–e343.

12. Marcinkiewicz J, Bryniarski K, Nazimek K. Ebola haemorrhagic fever virus: pathogenesis, immune responses, potential prevention. Folia Med Cracov. 2014;54: 39–48.

13. Baseler L, Chertow DS, Johnson KM, Feldmann H, Morens DM. The Pathogenesis of Ebola Virus Disease. Annu Rev Pathol. 2017;12: 387–418.

14. Zhu M. SARS Immunity and Vaccination. Cell Mol Immunol. 2004;1: 193–198. 15. Openshaw PJM, Chiu C, Culley FJ, Johansson C. Protective and Harmful Immunity to RSV

Infection. Annu Rev Immunol. 2017;35: 501–532.

21

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

16. Boonnak K, Vogel L, Feldmann F, Feldmann H, Legge KL, Subbarao K. Lymphopenia associated with highly virulent H5N1 virus infection due to plasmacytoid dendritic cell-mediated apoptosis of T cells. J Immunol. 2014;192: 5906–5912.

17. Murphy K, editor. Janeway’s immunobiology / Kenneth Murphy ; with acknowledgment to Charles A. Janeway Jr., Paul Travers, Mark Walport ; with contributions by Allan Mowat, Casey T. Weaver. 8th ed. The Quarterly Review of Biology; 2012.

18. Vossen MTM, Westerhout EM, Söderberg-Nauclér C, Wiertz EJHJ. Viral immune evasion: a masterpiece of evolution. Immunogenetics. 2002;54: 527–542.

19. Horner SM, Gale M. Intracellular Innate Immune Cascades and Interferon Defenses That Control Hepatitis C Virus. J Interferon Cytokine Res. 2009;29: 489–498.

20. Giamarellos-Bourboulis EJ, Raftogiannis M, Antonopoulou A, Baziaka F, Koutoukas P, Savva A, et al. Effect of the novel influenza A (H1N1) virus in the human immune system. PLoS One. 2009;4: e8393.

21. Huang Y, Zhu W, Zeng X, Li S, Li X, Lu C. Innate and adaptive immune responses in patients with pandemic influenza A(H1N1)pdm09. Arch Virol. 2013;158: 2267–2272.

22. Bouvier NM, Palese P. The biology of influenza viruses. Vaccine. 2008;26 Suppl 4: D49–53.

23. Cheng VCC, To KKW, Tse H, Hung IFN, Yuen K-Y. Two years after pandemic influenza A/2009/H1N1: what have we learned? Clin Microbiol Rev. 2012;25: 223–263.

24. Schneider WM, Chevillotte MD, Rice CM. Interferon-Stimulated Genes: A Complex Web of Host Defenses. Annu Rev Immunol. 2014;32: 513–545.

25. Beachboard DC, Horner SM. Innate immune evasion strategies of DNA and RNA viruses. Curr Opin Microbiol. 2016;32: 113–119.

26. Ma DY, Suthar MS. Mechanisms of innate immune evasion in re-emerging RNA viruses. Curr Opin Virol. 2015;12: 26–37.

27. Rong L, Perelson AS. Treatment of hepatitis C virus infection with interferon and small molecule direct antivirals: viral kinetics and modeling. Crit Rev Immunol. 2010;30: 131–148.

28. Alcami A. Viral mimicry of cytokines, chemokines and their receptors. Nat Rev Immunol. 2003;3: 36–50.

29. McSharry BP, Avdic S, Slobedman B. Human cytomegalovirus encoded homologs of cytokines, chemokines and their receptors: roles in immunomodulation. Viruses. 2012;4: 2448–2470.

30. Arvin A. Varicella-zoster virus: molecular virology and virus–host interactions. Curr Opin Microbiol. 2001;4: 442–449.

31. McFadden G, Murphy PM. Host-related immunomodulators encoded by poxviruses and herpesviruses. Curr Opin Microbiol. 2000;3: 371–378.

32. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M, et al. HIV evolution: CTL escape mutation and reversion after transmission. Nat Med. 2004;10: 282–289.

33. Persidsky Y, Poluektova L. Immune privilege and HIV-1 persistence in the CNS. Immunol Rev. 2006;213: 180–194.

34. Chan YK, Gack MU. Viral evasion of intracellular DNA and RNA sensing. Nat Rev Microbiol. 2016;14: 360–373.

35. Bowie AG, Unterholzner L. Viral evasion and subversion of pattern-recognition receptor signalling. Nat Rev Immunol. 2008;8: 911–922.

22

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

36. Swanson PA, II, Lukacher AE, Szomolanyi-Tsuda E. Immunity to Polyomavirus Infection: The Polyoma Virus-Mouse Model. Semin Cancer Biol. NIH Public Access; 2009;19: 244.

37. Scheinberg M, Blacklow NR, Goldstein AL, Parrino TA, Rose FB, Cathcart ES. Influenza: response of T-cell lymphopenia to thymosin. N Engl J Med. 1976;294: 1208–1211.

38. Varani S, Landini MP. Cytomegalovirus-induced immunopathology and its clinical consequences. Herpesviridae. 2011;2: 6.

39. Younan P, Iampietro M, Bukreyev A. Disabling of lymphocyte immune response by Ebola virus. PLoS Pathog. 2018;14: e1006932.

40. Baize S, Leroy EM, -C. Georges-Courbot M, Capron M, Lansoud-Soukate J, Debré P, et al. Defective humoral responses and extensive intravascular apoptosis are associated with fatal outcome in Ebola virus-infected patients. Nat Med. 1999;5: 423–426.

41. O’Donnell R, Tasker RC, Roe MFE. SARS: understanding the coronavirus: apoptosis may explain lymphopenia of SARS. BMJ. 2003;327: 620.

42. Kamphuis E, Junt T, Waibler Z, Forster R, Kalinke U. Type I interferons directly regulate lymphocyte recirculation and cause transient blood lymphopenia. Blood. 2006;108: 3253–3261.

43. Panesar NS, Lam CWK, Chan MHM, Wong CK, Sung JJY. Lymphopenia and neutrophilia in SARS are related to the prevailing serum cortisol. Eur J Clin Invest. 2004;34: 382–384.

44. Panesar NS. What caused lymphopenia in SARS and how reliable is the lymphokine status in glucocorticoid-treated patients? Med Hypotheses. 2008;71: 298–301.

45. Mudd JC, Busman-Sahay K, DiNapoli SR, Lai S, Sheik V, Lisco A, et al. Hallmarks of primate lentiviral immunodeficiency infection recapitulate loss of innate lymphoid cells. Nat Commun. 2018;9: 3967.

46. Pinschewer DD, Brinkmann V, Merkler D. Impact of sphingosine 1-phosphate modulation on immune outcomes. Neurology. 2011;76: S15–9.

47. Chahroudi A, Silvestri G, Lichterfeld M. T memory stem cells and HIV: a long-term relationship. Curr HIV/AIDS Rep. 2015;12: 33–40.

48. Pascutti MF, Erkelens MN, Nolte MA. Impact of Viral Infections on Hematopoiesis: From Beneficial to Detrimental Effects on Bone Marrow Output. Front Immunol. 2016;7. doi:10.3389/fimmu.2016.00364

49. Mire CE, Geisbert JB, Agans KN, Thi EP, Lee ACH, Fenton KA, et al. Passive Immunotherapy: Assessment of Convalescent Serum Against Ebola Virus Makona Infection in Nonhuman Primates. J Infect Dis. 2016;214: S367–S374.

50. Feldmann H, Feldmann F, Marzi A. Ebola: Lessons on Vaccine Development. Annu Rev Microbiol. 2018;72: 423–446.

51. Betters PhD RN DM, Betters DM, RN. Use of Flow Cytometry in Clinical Practice. Journal of the Advanced Practitioner in Oncology. 2015;6. doi:10.6004/jadpro.2015.6.5.4

52. Hassett J, Parker J. Laboratory practices in reporting flow cytometry phenotyping results for leukemia/lymphoma specimens: Results of a survey. Cytometry. 1995;22: 264–281.

53. Storch GA. Diagnostic Virology. Clin Infect Dis. 2000;31: 739–751. 54. Gillim-Ross L, Subbarao K. Emerging respiratory viruses: challenges and vaccine strategies.

Clin Microbiol Rev. 2006;19: 614–636. 55. Hatzakis A, Magiorkinis E, Haida C. HBV virological assessment. J Hepatol. 2006;44:

S71–S76. 56. Clay TM, Hobeika AC, Mosca PJ, Lyerly HK, Morse MA. Assays for monitoring cellular

23

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

immune responses to active immunotherapy of cancer. Clin Cancer Res. 2001;7: 1127–1135. 57. Folds JD, Schmitz JL. 24. Clinical and laboratory assessment of immunity. J Allergy Clin

Immunol. 2003;111: S702–S711. 58. Baltimore D. Expression of animal virus genomes. Bacteriol Rev. 1971;35: 235–241. 59. Lefkowitz EJ, Dempsey DM, Hendrickson RC, Orton RJ, Siddell SG, Smith DB. Virus

taxonomy: the database of the International Committee on Taxonomy of Viruses (ICTV). Nucleic Acids Res. 2018;46: D708–D717.

60. Yu C, Hernandez T, Zheng H, Yau S-C, Huang H-H, He RL, et al. Real time classification of viruses in 12 dimensions. PLoS One. 2013;8: e64328.

61. Mileham AJ. Identification of microorganisms using random primed PCR. Mol Biotechnol. 1997;8: 139–145.

62. Sirohi D, Chen Z, Sun L, Klose T, Pierson TC, Rossmann MG, et al. The 3.8 Å resolution cryo-EM structure of Zika virus. Science. 2016;352: 467–470.

63. Majchrzykiewicz-Koehorst JA, Heikens E, Trip H, Hulst AG, de Jong AL, Viveen MC, et al. Rapid and generic identification of influenza A and other respiratory viruses with mass spectrometry. J Virol Methods. 2015;213: 75–83.

64. Wynne SA, Crowther RA, Leslie AG. The crystal structure of the human hepatitis B virus capsid. Mol Cell. 1999;3: 771–780.

65. Freedman H, Logan MR, Hockman D, Koehler Leman J, Law JLM, Houghton M. Computational Prediction of the Heterodimeric and Higher-Order Structure of gpE1/gpE2 Envelope Glycoproteins Encoded by Hepatitis C Virus. J Virol. 2017;91. doi:10.1128/JVI.02309-16

66. Balsitis SJ, Coloma J, Castro G, Alava A, Flores D, McKerrow JH, et al. Tropism of dengue virus in mice and humans defined by viral nonstructural protein 3-specific immunostaining. Am J Trop Med Hyg. 2009;80: 416–424.

67. Kyrylkova K, Kyryachenko S, Leid M, Kioussi C. Detection of apoptosis by TUNEL assay. Methods Mol Biol. 2012;887: 41–47.

68. Halehalli RR, Nagarajaram HA. Molecular principles of human virus protein–protein interactions. Bioinformatics. 2014;31: 1025–1033.

69. Njayou M, Quash G. Purification of measles virus by affinity chromatography and by ultracentrifugation: a comparative study. J Virol Methods. 1991;32: 67–77.

70. Kimuda SG, Biraro IA, Bagaya BS, Raynes JG, Cose S. Characterising antibody avidity in individuals of varied Mycobacterium tuberculosis infection status using surface plasmon resonance. PLoS One. 2018;13: e0205102.

71. Shioda S, Kasai F, Watanabe K, Kawakami K, Ohtani A, Iemura M, et al. Screening for 15 pathogenic viruses in human cell lines registered at the JCRB Cell Bank: characterization of in vitro human cells by viral infection. Royal Society Open Science. 2018;5: 172472.

72. Keshwara R, Johnson RF, Schnell MJ. Toward an Effective Ebola Virus Vaccine. Annu Rev Med. 2017;68: 371–386.

73. Paiardini M, Cervasi B, Reyes-Aviles E, Micci L, Ortiz AM, Chahroudi A, et al. Low levels of SIV infection in sooty mangabey central memory CD4+ T cells are associated with limited CCR5 expression. Nat Med. 2011;17: 830–836.

74. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, Yassine-Diab B, et al. HIV reservoir size and persistence are driven by T cell survival and homeostatic proliferation. Nat Med. 2009;15: 893–900.

24

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019

75. Palmer BE, Neff CP, Lecureux J, Ehler A, Dsouza M, Remling-Mulder L, et al. In vivo blockade of the PD-1 receptor suppresses HIV-1 viral loads and improves CD4+ T cell levels in humanized mice. J Immunol. 2013;190: 211–219.

76. Porichis F, Kwon DS, Zupkosky J, Tighe DP, McMullen A, Brockman MA, et al. Responsiveness of HIV-specific CD4 T cells to PD-1 blockade. Blood. 2011;118: 965–974.

77. Winkler AM, Koepsell SA. The use of convalescent plasma to treat emerging infectious diseases: focus on Ebola virus disease. Curr Opin Hematol. 2015;22: 521–526.

78. Fine P, Eames K, Heymann DL. “Herd Immunity”: A Rough Guide. Clin Infect Dis. 2011;52: 911–916.

79. Russell CD. Eradicating Infectious Disease: Can We and Should We? Front Immunol. 2011;2. doi:10.3389/fimmu.2011.00053

80. Adkins JC, Wagstaff AJ. Recombinant hepatitis B vaccine: a review of its immunogenicity and protective efficacy against hepatitis B. BioDrugs. 1998;10: 137–158.

81. Chao Zhang DZ. Adenoviral vector-based strategies against infectious disease and cancer. Hum Vaccin Immunother. Taylor & Francis; 2016;12: 2064.

82. Gates IV, Zhang Y, Shambaugh C, Bauman MA, Tan C, Bodmer J-L. Quantitative measurement of varicella-zoster virus infection by semiautomated flow cytometry. Appl Environ Microbiol. 2009;75: 2027–2036.

83. Davenport MP, Hill AV. Reverse immunogenetics: from HLA-disease associations to vaccine candidates. Mol Med Today. 1996;2: 38–45.

84. Thielman NM, Cunningham CK, Woods C, Petzold E, Sprenz M, Russell J. Ebola clinical trials: Five lessons learned and a way forward. Clinical Trials: Journal of the Society for Clinical Trials. 2016;13: 83–86.

85. Kahn R, Rid A, Smith PG, Eyal N, Lipsitch M. Choices in vaccine trial design in epidemics of emerging infections. PLoS Med. 2018;15: e1002632.

25

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27518v1 | CC BY 4.0 Open Access | rec: 5 Feb 2019, publ: 5 Feb 2019