Viruses are obligatory intracellular microorganisms that use components of the nucleic acid and protein synthetic machinery of the host to replicate. Viruses typically infect various cell types by receptor-mediated endocytosis after binding to normal cell surface molecules. Viruses can cause tissue injury and dis ease by any of several mechanisms. Viral replication interferes with normal cellular protein synthesis and function and leads to injury and ultimately death of the infected cells. This result is one type of cytopathic effect of viruses, and the infection is said to be lytic because the infected cells are lysed. Synthesis of large amounts of viral proteins in infected cells leads to accumulation of misfolded proteins, which induce apoptotic death of the cells. Viruses can stimulate inflammatory responses that cause damage to tissues. Viruses can also cause latent infections, dis cussed later. Some viruses are oncogenic.
Innate and adaptive immune responses to viruses are aimed at blocking infection and eliminating infected cells (Fig. 1).
Fig1. Innate and adaptive immune responses against viruses. (A) Kinetics of innate and adaptive immune responses to a virus infection. (B) Mechanisms by which innate and adaptive immunity prevent and eradicate virus infections. Innate immunity is mediated by type 1 interferons (IFNs), which prevent infection, and natural killer (NK) cells, which eliminate infected cells. Adaptive immunity is mediated by antibodies and cytotoxic T lymphocytes (CTLs), which block infection and kill infected cells, respectively. DC, Dendritic cell.
Innate Immunity to Viruses
The principal mechanisms of innate immunity against viruses are inhibition of the spread of infection by type I IFNs and NK cell–mediated killing of infected cells. Infection by many viruses is associated with production of type I IFNs by infected cells and by plasmacytoid DCs and macrophages responding to viral products. Several biochemical path ways trigger IFN production. These include recognition of viral RNA and DNA by endosomal TLRs and activation of cytoplasmic RIG-like receptors and the STING (stimulator of interferon gene) pathway by viral RNA and DNA. These pathways converge on the activation of protein kinases that phosphorylate and thus activate IRF transcription factors, which stimulate IFN gene transcription. Type I IFNs inhibit viral replication in both infected and uninfected cells. Type I IFNs also stimulate the production of other host proteins that interfere with viral replication and are called restriction factors.
NK cells kill virus-infected cells and are an important mechanism of immunity against viruses, especially DNA viruses (herpesviruses, human papillomavirus [HPV], and others). NK cells may be activated early in the course of infection before adaptive immune responses have developed in response to upregulation of activating ligands on infected cells. They may also be activated later in the infection because MHC class I expression is often shut off in virus-infected cells as an escape mechanism from CTLs and this enables NK cells to kill the infected cells because the absence of class I releases NK cells from a normal state of inhibition.
Adaptive Immunity to Viruses
Adaptive immunity against viral infections is mediated by antibodies that block virus binding and entry into host cells and by CTLs, which eliminate the infection by killing infected cells (see Fig. 1). The most effective antibodies are high-affinity antibodies produced in T-dependent germinal center reactions. Antibodies are effective against viruses only during the extracellular stage of the lives of these microbes. Viruses will be extracellular when they first enter the body before they infect host cells, or when they are released from infected cells by virus budding or by lysis of infected cells. Antiviral anti bodies bind to viral envelope or capsid antigens and function mainly as neutralizing antibodies to prevent virus attachment and entry into host cells. Thus, antibodies prevent both initial infection and cell-to-cell spread. Secreted antibodies, especially of the IgA isotype, are important for neutralizing viruses within the respiratory and intestinal tracts. Oral immunization against poliovirus works by inducing mucosal immunity. In addition to neutralization, antibodies may opsonize viral particles and promote their clearance by phagocytes. Complement activation may also participate in antibody-mediated viral immunity, mainly by promoting phagocytosis and possibly by direct lysis of viruses with lipid envelopes.
The importance of humoral immunity in defense against viral infections is supported by several observations. First, naturally produced or monoclonal antibodies administered to individuals provide effective protection against infection, especially against acute infections. Such protection has been demonstrated in numerous viral infections, including influenza, polio, and many others; an important example is SARS-CoV-2, dis cussed later. Second, resistance to a particular virus, induced by either infection or vaccination, is often specific for the serologic (antibody-defined) type of the virus. An example is influenza virus, in which exposure to one serologic type does not confer resistance to other serotypes of the virus. Neutralizing antibodies block viral infection of cells and spread of viruses from cell to cell, but after the viruses enter cells and begin to replicate intracellularly, they are inaccessible to antibodies. Therefore, humoral immunity induced by previous infection or vaccination is able to protect individuals from viral infection but cannot by itself eradicate established infection.
Elimination of viruses that reside within cells is mediated by CTLs, which kill the infected cells. As we have mentioned in previous chapters, the principal physiologic function of CTLs is surveillance against viral infection. Most virus-specific CTLs are CD8+ T cells that recognize viral peptides derived from cytosolic, usually endogenously synthesized, viral proteins, presented by MHC class I molecules. If the infected cell is a tissue cell and not an antigen-presenting cell (APC), such as a DC, the infected cell or viral proteins released from the cell may be phagocytosed by DCs, which process the viral antigens and present them to naive CD8+ T cells to initiate the antiviral T-cell response. We described this process of cross-presentation, or cross-priming. Full differentiation of CD8+ CTLs often requires cytokines produced by CD4+ helper cells or costimulators expressed on APCs. CD8+ T cells undergo massive proliferation during viral infection, and most of the proliferating cells are specific for only a few viral pep tides. The activated T cells differentiate into effector CTLs, which can kill any infected nucleated cell that is producing viral antigens in the cytosol and presenting peptides from these anti gens on cell surface class I molecules. The antiviral effects of CTLs are mainly due to killing of infected cells, but other mechanisms include activation of nucleases within infected cells that degrade viral genomes and secretion of cytokines, such as IFN γ, which activates phagocytes.
Many lines of experimental and clinical evidence support the importance of CTLs in the defense against viral infection. Susceptibility to such infections is increased in patients and animals deficient in T lymphocytes. Humans and experimental animals can be protected against some virus infections by adoptive transfer of virus-specific, MHC class I–restricted CTLs. Viruses have developed numerous strategies to escape attack by CD8+ CTLs, discussed later in the chapter, reflecting the importance of CTL defense against these pathogens.
In latent infections, viral DNA persists in host cells, but the virus does not replicate or kill infected cells. Latency is a feature of infections by several viruses, especially DNA viruses. In latent viral infections, the viral DNA is maintained as nonintegrated episomes in host cell nuclei (e.g., herpeseviruses, pox viruses, and human papillomavirus), but no infectious virus is produced. The host immune response controls but does not eliminate the infection. If the host’s immune system becomes defective for any reason, the virus may be reactivated, causing significant clinical disease. Examples include shingles due to reactivation of latent varicella infection and cold sores due to reactivation of latent herpes simplex virus. HIV, a retrovirus, may become latent as an integrated but transcriptionally silent DNA provirus.
In some viral infections, CTLs may cause tissue injury. Some degree of immunopathology accompanies host responses to many, perhaps most, virus infections. An experimental model of a dis ease in which the pathology is primarily due to the host immune response is lymphocytic choriomeningitis virus (LCMV) infection in mice, which induces inflammation of the spinal cord meninges. LCMV infects meningeal cells, but it is noncytopathic and does not injure the infected cells directly. The virus stimulates the development of virus-specific CTLs that kill infected meningeal cells during a physiologic attempt to eradicate the infection. Therefore, meningitis develops in normal mice with intact immune systems, but T cell–deficient mice do not develop the disease and instead become carriers of the virus. This observation appears to contradict the usual situation, in which immunodeficient individuals are more susceptible to infectious diseases than are normal individuals. The CTL response is also the main cause of tissue injury in viral hepatitis in humans. Hepatitis B virus infection has the unusual feature that immunodeficient persons who become infected often do not develop the disease but become carriers who can transmit the infection to otherwise healthy persons. The livers of patients with acute and chronic active hepatitis contain large numbers of CD8+ T cells, and hepatitis virus–specific, MHC class I–restricted CTLs can be isolated from liver biopsy specimens and propagated in vitro.
Immune responses to viral infections may be involved in producing disease in other ways. Antiviral antibodies may promote activation of macrophages and other myeloid cells through binding to Fc receptors and even enhance the entry of viruses into host cells. This reaction, called antibody-mediated enhancement, has been postulated to increase lung inflammation associated with some virus infections, such as coronaviruses, dengue, and respiratory syncytial virus. A consequence of persistent infection with some viruses, such as hepatitis B, is the formation of circulating immune complexes composed of viral antigens and specific antibodies. These complexes are deposited in blood vessels and lead to systemic vasculitis. Some viral proteins contain amino acid sequences that are also present in self antigens. It has been postulated that because of this molecular mimicry, antiviral immunity can lead to immune responses against self antigens. An example is the association between Epstein-Barr virus and the autoimmune disease multiple sclerosis.
Immune Evasion by Viruses
Viruses have evolved numerous mechanisms for evading host immunity (Table 1).
• Viruses can alter their antigens and are thus no longer targets of immune responses. The antigens affected are most commonly surface glycoproteins that are recognized by antibodies, but T-cell epitopes may also undergo variation. The principal mechanisms of antigenic variation are point mutations and reassortment of RNA genomes (in RNA viruses), leading to antigenic drift and antigenic shift, respectively. These processes are of great importance in the spread of influenza virus. The two major antigens of influenza virus are the trimeric viral hemagglutinin (the viral spike protein) and neuraminidase. Viral genomes undergo mutations in the genes that encode these surface proteins, and the variation that occurs as a result is called antigenic drift. Alternatively, the segmented RNA genomes of various strains of influenza viruses that normally inhabit different host species can recombine in host cells that are simultaneously infected with two different influenza viruses, and these reassorted viruses can differ quite dramatically from prevalent strains (Fig.2). Reassortment of viral genes results in major changes in an antigenic structure called antigenic shift, which creates distinct viruses such as the avian flu or the swine flu viruses. Because of antigenic variation, a virus may become resistant to immunity generated in the population by previous infections. The influenza pandemics that occurred in 1918, 1957, and 1968 were due to different strains of the virus, and the H1N1 pandemic of 2009 was due to a strain in which the strands of the RNA genome were reassorted among strains endemic in pigs, fowl, and humans. Subtler viral variants arise more frequently. There are so many serotypes of rhinovirus that vaccination against the common cold may not be a feasible preventive strategy. HIV-1, which causes AIDS, is also capable of tremendous antigenic variation because of a high error rate in reverse transcription of its RNA genome during viral reproduction. In these situations, prophylactic vaccination may have to be directed against invariant viral proteins.
• Some viruses inhibit the ability of the host to induce the antiviral state. Many viruses that cause severe acute dis eases, including the Ebola virus, Marburg virus, and the Betacoronaviruses SARS-CoV-1, MERS-CoV, and SARS CoV-2, encode proteins that shut off the host type I IFN response. The significance of this in SARS-CoV-2 infection is discussed later.
• Some viruses inhibit MHC class I–associated presentation of cytosolic protein antigens. Viruses make a variety of proteins that block different steps in antigen processing, trans port, and presentation (Fig. 3). Inhibition of antigen presentation blocks the assembly and expression of stable MHC class I molecules and the display of viral peptides. As a result, cells infected by such viruses cannot be recognized or killed by CD8+ CTLs. As discussed earlier, NK cells are activated by infected cells, especially in the absence of MHC class I molecules, and this may compensate to some degree for the defective recognition of infected cells by CD8+ CTLs. Some viruses may produce proteins that act as ligands for NK cell inhibitory receptors and thus inhibit NK cell activation.
• Some viruses produce molecules that inhibit the effector phase of immune responses. Poxviruses encode molecules that are secreted by infected cells and bind to several cytokines, including IFN-γ, TNF, IL-1, IL-18, and chemokines. The secreted cytokine-binding proteins may function as competitive antagonists of the cytokines. Epstein-Barr virus produces a protein that is homologous to the cytokine IL-10, which inhibits activation of macrophages and DCs and may thus suppress cell-mediated immunity. These examples probably represent a small fraction of immunosuppressive viral molecules. Identification of these molecules raises the intriguing possibility that viruses have acquired genes encoding endogenous inhibitors of immune responses during their passage through human hosts and have thus evolved to infect and colonize humans.
• Some chronic viral infections are associated with failure of CTL responses, called exhaustion, which allows viral persistence. Studies of a chronic infection with LCMV in mice have shown that this type of immune deficit may result from persistent antigen stimulation in the setting of inflammation leading to upregulation of T-cell inhibitory receptors, such as PD-1 (programmed cell death protein 1). There is evidence for CD8+ T-cell exhaustion in chronic human viral infections, including HIV and hepatitis virus infection. The physiologic importance of PD-1-mediated T-cell inhibition may be that it limits pathology caused by strong immune responses to viruses.
• Viruses may infect and either kill or inactivate immunocompetent cells. The clearest example is HIV, which survives by infecting and eliminating CD4+ T cells, the key inducers of immune responses to protein antigens.
Table1. Mechanisms of Immune Evasion by Viruses
Fig2. Generation of new influenza virus strains by genetic recombination (antigenic shift). The genome of the influenza virus is composed of eight separate RNA strands, which allows genetic recombination by reassortment of the segments in various hosts, such as pigs (not shown), birds, or humans, that are simultaneously infected with two different strains. These genetic reassortments create new viruses that are antigenically distinct from their precursors and thus are able to evade immune detection in large numbers of newly infected hosts. The H1N1 influenza virus, which was responsible for the pandemic of 2009, was generated by reassortment of swine, avian, and human viruses in pigs and then passed back to humans. HA, Hemagglutinin; NA, neuraminidase.
Fig3. Mechanisms by which viruses inhibit recognition by CD8+ T cells and natural killer (NK) cells. The pathway of major histocompatibility complex (MHC) class I–associated antigen presentation is shown, with examples of viruses that block different steps in this pathway. In addition to interfering with recognition by CD8+ T cells, some viruses produce “decoy” MHC molecules that engage inhibitory receptors of NK cells. CMV, Cytomegalovirus; CTL, cytotoxic T lymphocyte; EBV, Epstein-Barr virus; ER, endoplasmic reticulum; HSV, herpes simplex virus; TAP, transporter associated with antigen processing.
SARS-CoV-2 and COVID-19
The pandemic caused by the SARS-CoV-2 Betacoronavirus started in December 2019 and has had an enormous impact on societies worldwide. It is useful to summarize the salient features of this viral infection and the diseases it causes, COVID-19, because of its health implications and because it illustrates many features of the pathogenesis of, and defense against, an acute cytopathic viral infection.
Coronaviruses are enveloped, positive-sense, single-stranded RNA viruses. SARS-CoV-2 belongs to a subclass of coronaviruses (with SARS-CoV-1 and MERS) that cause severe acute respiratory diseases; these viruses are sometimes referred to as Sarbecoviruses (severe acute respiratory infection linked Betacoronaviruses). SARS-CoV-2 is particularly easily trans mitted from person to person since it is carried by droplets that rapidly gain entry into the upper respiratory tract (Fig. 4). The RNA of the virus is coated by the nucleocapsid (N) protein, and ribonucleoprotein capsids emerge from infected cells within a lipid envelope derived from host cell membranes. From a clinical standpoint, the most frequently used tests for infection rely on identification of the SARS-CoV-2 N protein in nasal swabs by antibody binding assays, though PCR-based diagnostic approaches are also used. The spike (S) protein of SARS-CoV-2 is embedded in the envelope and binds to ACE2 (angiotensin converting enzyme 2) on epithelial cells, most frequently in the upper and lower respiratory tracts. All approved vaccines against SARS-CoV-2 target the S protein. Currently, tests for antibodies to the S protein cannot distinguish between previous infection and vaccination. However, the presence of antibodies to the N protein are diagnostic of previous or recent infection.
Fig4. Pathogenesis and outcome of COVID-19. SARS-CoV-2 enters through the airways and establishes initial infection in the respiratory tract. Subsequent disease severity is determined in large part by the host innate and adaptive immune response and subsequent spread and persistence of the virus. ACE2, Angiotensin converting enzyme 2; IFN, interferon.
The other proteins produced by SARS-CoV-2 include polymerases and proteases that are required for viral replication. Some of these enzymes have been targets for antiviral drugs.
Remarkably, 10 of the 29 proteins encoded by the viral genome are dedicated to dampening innate immunity by attenuating host type I IFN production. They do so by various means. Some viral proteins ensure that the host endoplasmic reticulum extrudes double-membrane vesicles (DMVs) within which viral RNA can replicate largely hidden from cytosolic RNA sensors. Another viral protein forms a pore in the DMVs to allow replicated RNAs to escape into the cytosol and be rapidly pack aged into capsids. Some viral proteins modify viral RNA so it can more closely resemble host mRNA (by adding a version of a 7-methyl guanosine cap and by adding a 2′O-methy group to the ribose in the next nucleotide). These modifications prevent recognition by the cytosolic RNA sensors RIG-I and MDA-5. Other viral proteins attenuate the translation of host type I IFNs. The importance of type I IFNs in antiviral immunity is further highlighted by the observation that individuals who make autoantibodies against their own type I IFN or have genetic defects affecting IFN production or signaling are at increased risk for developing severe COVID-19 following infection with SARS-CoV-2.
SARS-CoV-2 infections can be asymptomatic or result in mild to moderate infections with flulike symptoms. However, this virus can also be the cause of severe respiratory infections that compromise lung function and can be life threatening. Severe infections occur more often in the elderly, and more frequently in males. The frequency of severe life-threatening infections that required hospitalization in intensive care units was far greater before widespread herd immunity was achieved in populations by vaccination and/or infection. Initially, in the early pandemic period, protection from vaccination resulted from neutralizing antibodies, nonneutralizing antibodies, and T-cell immunity. The widespread induction of adaptive immunity however proved to be a driving force for viral evolution resulting in alterations, largely in the Spike gene, that have led to the emergence of waves of viral variants with newer Spike proteins that evade preexisting neutralizing antibodies generated by previous SARS-CoV-2 infection or vaccination. The vaccines used now protect largely through nonneutralizing antibodies and T-cell immunity.
Disease manifestations, whether mild, moderate, or severe, result from a localized or systemic inflammatory reaction to the virus. The failure of innate immunity, especially type I IFN dependent antiviral mechanisms, to contain the infection contributes to the pathology and clinical manifestations of COVID-19. The inflammation is associated with increased production of numerous cytokines. Severe disease is characterized by largely myeloid cell–driven inflammation, with exuberant monocyte and neutrophil activation and complement activation. Inflammatory cytokine production is likely induced by viral RNAs that target endosomal TLRs and possibly by damage-associated molecular patterns released by dying and dead cells. Often, there is severe lymphopenia with decreased T-cell numbers. The contribution of adaptive immune cells to inflammation in this disease is not well established. While preexisting antibodies to type I IFNs and genetic variants causing reduced production or function of type I IFN may contribute to disease susceptibility in some people, it is still not well understood why severe disease is more frequent in older individuals, especially in males with preexisting conditions such as type 2 diabetes.
Although severe SARS-CoV-2 infection is less frequent in children, postinfectious sequelae are seen in a small fraction of children and these represent a life-threatening illness referred to as multisystem inflammatory syndrome in children (MIS-C). MIS-C can present with fever and cardiac inflammation and a Kawasaki disease–like syndrome, or some combination of fever with inflammation of the heart, skin rashes, shock, gastrointestinal symptoms, a low platelet count, and sometimes lymphopenia. The disease is accompanied by elevated acute-phase proteins and is clearly immunologically driven. Intravenous immunoglobulin injections and steroids are the most commonly used therapeutic interventions. The underlying immunological mechanisms that drive MIS-C remain poorly understood.
In adults, postinfectious sequelae of COVID-19 are observed in many individuals, and can affect different organs in different individuals. These sequelae are collectively referred to as PASC (postacute sequelae of COVID-19) in the scientific literature, but this syndrome is widely called “long COVID.” The underlying basis of long COVID is thought to be some combination of persistent infection and a dysregulated immune response, but the pathogenesis of this syndrome is poorly understood.
