Emerging Antiviral Therapeutics for Human Adenovirus Infection: Recent Developments and Novel Strategies

Abstract

Human adenoviruses (HAdV) are ubiquitous human pathogens that cause a significant burden of respiratory, ocular, and gastrointestinal illnesses. Although HAdV infections are generally self-limiting, pediatric and immunocompromised individuals are at particular risk for developing severe disease. Currently, no approved antiviral therapies specific to HAdV exist. Recent outbreaks underscore the need for effective antiviral agents to treat life-threatening infections. In this review we will focus on recent developments in search of potential therapeutic agents for controlling HAdV infections, with a focus on those targeting post-entry stages of the virus replicative cycle.

1. Adenovirus-associated Illnesses

Human adenoviruses (HAdV) are prevalent human pathogens. Acute infections pose a serious threat, particularly in children and immunocompromised individuals. Additionally, HAdV has been periodically associated with outbreaks causing significant illness and mortality (Louie et al., 2008; Meehan, 2018; Metzgar et al., 2007; Nedelman, 2018; Paquet, 2014). Transmission of HAdV is typically person-to-person via aerosols or droplets; however, fecal-oral and fomite transmission has also been documented (Khanal et al., 2018). HAdV infections were historically diagnosed by directly culturing samples on A549 lung adenocarcinoma cells or MRC5 lung fibroblasts, often in combination with immunofluorescent detection of viral antigens (Hematian et al., 2016). However, culture-based techniques have largely been supplanted by molecular assays, including real-time PCR-based detection and quantification of virus in blood and other sample types. Additionally, multiplexed PCR-based assays allowing detection of HAdV and other pathogens in respiratory and stool specimens are routinely used (Cybulski Jr et al., 2018; Echavarría et al., 2018; Khanal et al., 2018; Lee et al., 2019; Lu et al., 2013).

HAdV infections present with a wide array of symptoms and disease manifestations, which largely depend on the type of adenovirus, the immune status of the host, and the site of infection (Khanal et al., 2018; Lion, 2014; Lynch and Kajon, 2016). There are over 60 different types of genetically distinct HAdVs (divided into species A-G), each of which preferentially infect certain tissues and cause a spectrum of different illnesses (Lion, 2014). The more common sites of infection for HAdV include the respiratory tract, corneal epithelia, and the intestinal tract. HAdV is responsible for 90% of viral conjunctivitis cases, typically related to infection by species B, D, or E. This can be in the form of pharyngoconjunctivitis or epidemic keratoconjunctivitis (EKC), often seen in military recruits (Khanal et al., 2018). Species B, C, and E are associated with adenovirus-related respiratory infections, with HAdV-B7 being of particular concern (Scott et al., 2016). Two more serious outcomes of respiratory adenovirus infections are pneumonia, which can be lethal in children, and acute respiratory distress syndrome, again more common in military recruits (Khanal et al., 2018). Finally, species A, F, and G are associated with gastrointestinal infections (Garcia-Zalisnak et al., 2018; Lion, 2014). HAdVs are a major cause of childhood gastroenteritis, second only to norovirus and rotavirus (Khanal et al., 2018). Urinary tract infections are also possible, particularly in transplant recipients, often presenting as hemorrhagic cystitis or, in renal transplant recipients, allograft dysfunction (Lynch and Kajon, 2016).

Overall, adenovirus infections are typically self-limiting; however, immunocompromised individuals can experience far more severe illness and additional complications. Specifically, hematopoietic stem cell and solid organ transplant recipients, individuals with congenital immunodeficiencies, or those with unmanaged human immunodeficiency virus (HIV) infections, may experience colitis, hepatitis, encephalitis, or disseminated disease, and those patient populations are at substantially increased risk for mortality due to HAdV infection (Khanal et al., 2018; Lynch and Kajon, 2016). It is important to note that HAdV disease in immunocompromised individuals can be related to de novo infection, wherein the individual undergoes a primary viral infection. Alternatively, disease can also be caused by reactivation of a latent pool of virus that was already present in the individual or an engrafted organ (Lion, 2014).

2. Adenovirus Biology

Adenoviruses are non-enveloped, with a 26–45 kbp linear dsDNA genome. Their host tropism ranges from birds and reptiles to mammals (Davison et al., 2003). For the purpose of this review, we will focus on HAdVs. The virion structure is icosahedral and contains hexon, penton, and fiber proteins. Hexon proteins comprise the majority of the virion, with pentons at every vertex, and a fiber protein projecting from each penton base. Additional viral proteins support the structure of the virion from inside and outside, acting as a cement to increase virion stability and contribute to virion maturation (Nemerow et al., 2012).

The HAdV replicative cycle is complex and dynamic. There are three broadly defined phases of its replicative cycle: entry, early phase, and late phase. Each phase represents a distinct cellular milieu, with accompanying changes in cellular networks which could unveil novel susceptibilities for therapeutic intervention. The entry phase of the HAdV replicative cycle includes all processes from virion attachment to viral genome delivery to the nucleus. We will be focusing on inhibitors that are effective after the viral entry phase. However, to give context, we will briefly describe these three phases of infection.

2.1. Entry Phase

The HAdV replicative cycle, depicted in Figure 1, begins with attachment of the virion to the cell, which is mediated by the viral fiber protein interacting with a cellular receptor. The target cell receptor varies depending on HAdV type. Specifically, the coxsackievirus and adenovirus receptor (CAR) is used by most HAdV-C species. Other HAdVs use heparan sulfate, sialic acid, CD46, CD80, CD86, GD1a, polysialic acid, or desmoglein-2 for attachment (Stasiak and Stehle, 2020; Zhang and Bergelson, 2005). Once fiber is bound to its receptor, additional interactions occur to promote virion internalization in a clathrin-dependent process. Typically, this is mediated by an RGD amino acid motif in the penton base interacting with αvβ3 and αvβ5 integrins (Stasiak and Stehle, 2020; Zhang and Bergelson, 2005). After virion internalization, fiber proteins are forcibly torn off by disruptions caused by receptor interactions. These molecular tears expose pVI, which damages the vesicle membrane, leading to virion release (Figure 1: step 2)(Flatt and Butcher, 2019). Upon entering the cytoplasm, viral structural proteins guide the now partially disassembled HAdV particle to the nuclear pore complex (NPC). Specifically, the interaction of hexon with dynein motors traffics virions to the nucleus via microtubules (Bremner et al., 2009; Leopold et al., 2000). Once the virion reaches the perinuclear microtubule organizing center, it detaches and binds to the NPC, triggering capsid disassembly. The HAdV genome is organized into a nucleosome-like structure by its associated pVII core proteins. Once the viral genome is free from the capsid, it is directed into the nucleus via pVII, which contains nuclear localization signals (NLSs) that orchestrate nuclear translocation of the genome (Lynch et al., 2019; Mirza and Weber, 1982; Strunze et al., 2011; Vayda et al., 1983; Wodrich et al., 2006).

Figure 1: Potential antiviral therapeutics targeting the HAdV replicative cycle.

Candidate antivirals are shown in blue boxes with inhibitor arrows indicating where they function to disrupt the human adenovirus (HAdV) replicative cycle. 1) HAdV virions bind to cell surface receptors to mediate internalization into endosomes. 2) virions escape the endosome, tearing off fiber proteins in the process, and use microtubules to traffic to the 3) nuclear pore complexes (NPC) to release their genome. 4) HAdV gene expression occurs via transcription of the genome into mRNA. Gene expression inhibitors, epigenetic-based inhibitors, and cyclin-dependent kinase (CDK) inhibitors target the transcriptional process. mRNA are exported into the cytoplasm for translation into both early and late viral proteins, which are imported back into the nucleus. 5) HAdV protein-primed genome replication occurs, a process targeted by protease inhibitors. DNA synthesis of new viral genomes is disrupted by cidofovir-based derivatives and other DNA synthesis inhibitors. 6) virion assembly occurs in the nucleus, with the newly produced viral genomes assembled with late HAdV structural proteins. Protease inhibitors target the protease-driven maturation of these structural proteins. 7) Viral egress is triggered by cell lysis, induced by adenovirus death protein (ADP) in some HAdV species, and could be targeted by egress inhibitors. Created with BioRender.com

2.2. Early Phase

Phase two of the HAdV replicative cycle is the early phase, which commences almost immediately after viral DNA enters the nucleus (Figure 1: steps 3 and 4). During this phase, the early viral gene products begin to be transcribed and translated, inducing large changes in the cellular transcriptome and protein interactome. This process begins with pVII core proteins, which interact with histone chaperones to orchestrate their removal from the HAdV genome. These interactions spark the initial burst of viral gene expression, including that of the early region 1 A protein product, E1A. As the first viral protein produced de novo during the HAdV replicative cycle, the expression of E1A is highly dependent on host factors. E1A is essential for viral replication, as it is responsible for transactivation of the other early viral genes (Jones and Shenk, 1979). Once E1A expression has begun, pVII directs E1A to the HAdV genome, where it recruits host factors to induce subsequent rounds of transcription (Haruki et al., 2006; Johnson et al., 2004; Komatsu et al., 2011; Lynch et al., 2019; Ostapchuk et al., 2017).

The primary role of E1A is to dramatically alter the intracellular environment, making it more amenable to HAdV replication. One of the best known paths E1A takes to achieve this is by forcing the cell to divide. HAdV typically infects quiescent cells, terminally differentiated cells. In these non-dividing cells, the retinoblastoma tumor suppressor gene product (pRb) is in the active hypophosphorylated state. In this state, pRb is bound to members of the E2 factor (E2F) family of transcription factors, inhibiting their transcriptional activation function. Normally, cyclin-dependent kinases (CDKs) are required to phosphorylate and inactivate pRb at the appropriate time to allow entry into the cell division cycle. HAdV E1A bypasses this regulation by sequestering pRb, inappropriately freeing E2F to induce cell cycle gene expression (Berk, 2005). Alternatively, E1A has been shown to bind E2F directly to promote cell cycle entry (Pelka et al., 2011).

Primary transcripts from most adenovirus genes undergo alternative splicing, and E1A is no exception. For HAdV-C, five E1A isoforms are produced, and the relative ratios of these products change during the infectious cycle depending on splice site selection (Perricaudet et al., 1979; Radko et al., 2015; Stephens and Harlow, 1987). The two largest E1A isoforms are responsible for the majority of the well-known functions of E1A. Although they share many overlapping functions, they also possess some unique roles, particularly with respect to transcriptional control (Berk, 2005; King et al., 2018b; Pelka et al., 2009; Prusinkiewicz et al., 2020).

The HAdV early region 1 B (E1B) encodes two early proteins, which also arise through differential splicing. E1B-55K is best known for its ability to inhibit apoptosis through promoting proteasomal degradation of p53 (Blackford and Grand, 2009; Steegenga et al., 1998). Similarly, E1B-19K acts as a BCL-2 homologue to block the pro-apoptotic functions of other BCL-2 family members. Thus, both of the E1B gene products act to prevent apoptosis in infected cells (Han et al., 1996). The combination of E1A and E1B expression causes a dramatic reprograming of cellular behaviour, forcing quiescent, terminally differentiated infected cells to divide, while rendering them unable to respond to apoptotic cues (Gallimore and Turnell, 2001).

Following E1A and E1B expression, early region 2 A and B (E2A and E2B) are the next viral early genes expressed. E2A encodes the single-stranded DNA binding protein (DBP) and E2B encodes the terminal protein (pTP) and viral DNA polymerase. DBP is used to bind single stranded viral DNA as it is synthesized during genome replication and to stimulate polymerase initiation and elongation. TP is essential as a protein primer for HAdV DNA replication and polymerase is responsible for conducting the actual process of DNA synthesis (de Jong et al., 2003; de Jong and van der Vliet, 1999; van Breukelen et al., 2003). The conditions created by HAdV infection to force the infected cell into the cell cycle are thought to increase the availability of resources, specifically nucleotides and metabolic substrates, for viral DNA synthesis and progeny production (King et al., 2018b; Pelka et al., 2008; Prusinkiewicz and Mymryk, 2019).

The HAdV early region 3 (E3) gene encode a plethora of immunomodulatory proteins that primarily function to maintain cell viability during lytic infection and may also contribute to establishing and/or maintaining persistent HAdV infection. In HAdV-C, E3-gp19K protects infected cells from recognition by MHC-restricted cytotoxic T cells by binding MHC I antigens and preventing their intracellular transport to the cell surface (Andersson et al., 1985). The E3-encoded RID complex prevents the action of proapoptotic cytokines like FAS and TRAIL, and blocks TNF-mediated inflammation (Lichtenstein et al., 2004). Interestingly, the E3 region of some HAdV encode an adenovirus death protein (ADP) that is expressed at later stages of infection and appears critical for efficient release of progeny virions (Georgi and Greber, 2020).

The early region 4 (E4) viral gene encodes multiple products that modulate transcription, cell cycle, cell signaling, and DNA repair (Weitzman, Matthew, 2005). E4orf3 and E4orf6 are particularly important as they promote the transition from the early phase of infection to the late phase. These proteins activate the switch for the virus to produce nascent viral genomes and begin late gene expression. Later in the replicative cycle, they help to disrupt host protein synthesis, which sets the stage for cell death, lysis, and virion release (Huang and Hearing, 1989). Like E1A, the multifunctional E4 gene products induce large-scale changes to the cellular environment. In particular, E4orf3 forms polymers, creating unusual tracks of promyelocytic leukemia (PML) protein in infected cell nuclei. This viral compartmentalization of host chromatin induces broad changes to cellular gene expression, including suppression of the p53-dependent DNA damage response pathway (Soria et al., 2010).

Finally, HAdV generally expresses two functional RNAs, VA RNA I and II. These inactivate PKR and also disrupt the function of cellular microRNA processing machinery, using it to target cellular factors instead (Coventry and Conn, 2008; O’Malley et al., 1986; Piedade and Azevedo-Pereira, 2017).

2.3. Late Phase

The third, and final phase of HAdV replication is the late phase, which involves nascent virion assembly, genome packaging, viral egress, and cell death (Figure 1: steps 6 and 7). The viral late genes 1–5 (L1-L5) are transcribed from a common major late promoter (MLP) and are produced through differential splicing (Young, 2003). Since virion assembly occurs in the nucleus, viral mRNAs encoding structural proteins must be exported to the cytoplasm for translation, and the late proteins imported back into the nucleus. The general process of virion assembly occurs in a stepwise fashion; first the structural proteins assemble, then the non-structural proteins, followed by genome packaging and virion maturation (Ostapchuk and Hearing, 2005). Specifically, pVII is involved in genome packaging by directly binding the HAdV genome. Additional interactions between pVII and pV allow stabilization of the genome with the capsid (Lynch et al., 2019). An important component of HAdV virion assembly is the adenovirus protease (AVP), encoded by the late cassette L3. AVP is absolutely essential for infection, as it proteolytically cleaves multiple viral proteins into their final functional forms, including proteins IIIa, VI, and VIII, core proteins VII, X, and pTP (Gupta et al., 2017).

The HAdV replicative cycle culminates in massive levels of virus production, which coincides with cell death (Figure 1: step 7). This is triggered by the E3-encoded ADP, which by nomenclature is an early gene product, but has higher expression when the MLP is active during the late phase of infection (Tollefson et al., 1992). ADP may promote virus escape and cell lysis by nuclear envelope destabilization (Georgi and Greber, 2020; Tollefson et al., 1996). Aside from ADP-mediated cell lysis, the molecular details surrounding virion assembly, genome encapsidation, and viral egress remain largely unknown (Pied and Wodrich, 2019).

3. Current Therapies: Cidofovir, Ganciclovir, and Ribavirin

In many situations, supportive care and, when relevant, reduction of immunosuppression are primary management strategies for patients with HAdV infection. There are currently no specific antivirals approved for the treatment of HAdV infections. Instead, treatment of severe HAdV infection revolves around the off-label use of broad-spectrum antivirals like cidofovir, ganciclovir, and ribavirin (Kuwatsuka et al., 2020; Lenaerts et al., 2008; Lopez et al., 2018). Ribavirin has shown little, if any, evidence at improving the outcomes of HAdV infections (Gavin and Katz, 2002; Lankester et al., 2004). Ganciclovir was originally developed for treatment of herpesvirus infections (Matthews and Boehme, 1988). However, much like ribavirin, most evidence suggests little value for ganciclovir as a HAdV therapeutic. This is unsurprising as ganciclovir requires activation by a viral thymidine kinase to be most effective, which HAdV does not encode (Chen et al., 1997; Duggan et al., 1997; Matthes-Martin et al., 2012; Matthews and Boehme, 1988; Naesens et al., 2005; Ying et al., 2014). Interestingly, in vitro studies of HAdV co-infection with human cytomegalovirus (CMV) suggests improved anti-HAdV activity of ganciclovir, which is proposed to be due to CMV viral kinase-mediated activation of ganciclovir (Aguilar-Guisado et al., 2020). Currently, cidofovir is used as standard of care, albeit as an off-label treatment. Originally, cidofovir was developed for the treatment of CMV infections, but significant evidence supporting its use as an anti-HAdV agent has been established through in vitro experiments and some animal models (Gordon et al., 1992b; Kaneko et al., 2004; Mul et al., 1989; Romanowski and Gordon, 2000).

Cidofovir is a cytosine nucleoside analog that acts as an inhibitor of viral DNA synthesis (Figure 1: step 5). The anti-HAdV activity of cidofovir likely occurs through two mechanisms (Chamberlain et al., 2019). Firstly, cidofovir acts as a chain terminator of viral DNA synthesis. Incorporation of cidofovir into nascent viral DNA strands by the virally-encoded E2B polymerase creates a structure which is incompatible for further synthesis to occur. The second mechanism involves cidofovir directly binding and inhibiting the viral DNA polymerase (Chamberlain et al., 2019). Off-label use of cidofovir as a therapeutic agent for serious HAdV disease has been associated with clinical benefits (Alcamo et al., 2020; Doan et al., 2007; Fanourgiakis et al., 2005; Ganapathi et al., 2016; Ljungman et al., 2003; Neofytos et al., 2007). However, there are undeniable problems with this drug that severely limit its usefulness. Cidofovir has low bioavailability, meaning more drug must be administered to reach adequate serum concentrations to provide a clinical benefit. Additionally, the quick uptake, but slow release of cidofovir from tubular kidney cells leads to significant nephrotoxicity (Caruso Brown et al., 2015). In addition, there are conflicting reports on the efficacy of cidofovir, with some studies suggesting little or no effect on the clinical outcome of HAdV infections (Ronchi et al., 2014; Thomas et al., 2020; Zając-Spychała et al., 2020). Finally, single amino acid changes in the viral polymerase are sufficient to confer cidofovir resistance in HAdV, suggesting that virus escape mutants could become problematic (Kinchington et al., 2002). The bioavailability and nephrotoxicity remain problems of paramount importance. HAdV disease manifestations are particularly severe and burdensome in children and immunocompromised individuals. These populations may also be more susceptible or predisposed to nephrotoxicity (Caruso Brown et al., 2015; Hanna et al., 2016). Therefore, novel therapeutic avenues must be pursued, with drug concentration and toxicity as important considerations.

4. New Antivirals

4.1. Cidofovir-based Derivatives

Initial attempts to develop novel antivirals that overcome the problems of cidofovir focused on derivatives of the nucleoside analog (Dropulic and Cohen, 2010). Specifically, brincidofovir (formerly CMX-001) was developed as an oral, bioavailable, lipid-conjugated derivative of cidofovir. Similar to ganciclovir, this derivative also requires intracellular modifications to become active. Brincidofovir is cleavage-activated by cellular phospholipases, followed by phosphorylation by cellular kinases into its active, cidofovir diphosphate form (Beadle et al., 2002). Brincidofovir was shown to be more effective against HAdV in vitro at lower concentrations than cidofovir (Hartline et al., 2005). Brincidofovir was also effective at reducing HAdV replication and associated mortality in an in vivo model for HAdV replication (Toth et al., 2008). Phase I trials on the safety of brincidofovir for HAdV treatment supported further testing, with most treated cases exhibiting a quick decline in HAdV viremia (V. K. Prasad et al., 2017). Unfortunately, phase II trials testing the efficacy of brincidofovir for pediatric and adult HAdV infection revealed no statistically significant difference from the placebo treatment. (Grimley et al., 2017). Despite early promising results, brincidofovir was not approved for treatment of patients with HAdV infections, and is no longer available from the manufacturer. Similarly, brincidofovir showed exciting preclinical effects on CMV infection, but was also deemed unfit for patient use after clinical evaluation (Marty et al., 2019). The disconnect between in vitro results and efficacy in patients highlights the importance of clinical trial evaluation for emerging therapies.

4.2. Post-Entry Inhibitors

Viral inhibition can be achieved at two broad timepoints: entry and replication. There is some evidence supporting the development and use of entry inhibitors to block HAdV infection, most of which can be attributed to blocking viral attachment, including some interesting findings on the anti-HAdV capacity of defensins (Aplander et al., 2011; Caraballo et al., 2015; Colpitts and Schang, 2014; Nguyen et al., 2010; Spjut et al., 2011). As mentioned in Section 2.1, the cellular receptors used for HAdV entry are diverse, and antiviral development against these targets would likely be specific to a narrow subset of HAdVs. As such, for the purposes of this review, we will be focusing on post-entry inhibitors of the HAdV replicative cycle (Figure 1).

Post-entry inhibitors of HAdV infections can be tackled from two directions. The obvious approach is to pursue direct acting antivirals (DAA), which are compounds that directly interfere with the function of a viral gene product. Early attempts at developing antivirals typically focused on this method, which has the disadvantage that viral escape mutants can quickly arise that are no longer sensitive to treatment. More recently, however, host-directed therapies (HDT) have become attractive alternatives for antiviral therapy. This approach takes advantage of the reality that viruses are obligate intracellular parasites that absolutely rely on the host cell to complete their replicative cycles. As one example, viruses rely on the host for protein synthesis and metabolic energy. Indeed, HAdV infection causes an altered cellular metabolic profile through the E1A and E4 encoded proteins (Mayer et al., 2019; Prusinkiewicz et al., 2020; Prusinkiewicz and Mymryk, 2019). Because HAdV is reliant on host metabolic processes and remodels many of these pathways to promote viral replication, these processes represent a unique opportunity for HDT development. An obvious limitation of HDT is the possible toxicities associated with targeting host factors and processes. For these approaches to be successful, the target must be critical to the viral replicative cycle. If the target is also essential for host cell survival, this could produce toxicities that surpass the therapeutic benefits. Fortunately, viral infection creates a unique cellular milieu compared to that of an uninfected cell, which potentially creates virus-induced sensitivities that are essential for survival of infected cells but not uninfected cells (Mast et al., 2020). Ideally, HDTs aim to cripple these unique host-related sensitivities that are themselves created by the process of viral infection, reducing the likelihood of HDT-associated toxicities in uninfected cells. This approach has the additional advantage that virus escape mutants are far less likely to develop because the targets are of host, rather than viral origin.

4.3. DNA Replication Inhibitors

One approach to antiviral therapy for HAdV is to interfere with viral DNA replication (Figure 1: step 5). HAdV replicates its genome within the nucleus using a virally-encoded DNA polymerase whose enzymatic activity could be differentially sensitive to an inhibitory substrate compared to host cell DNA polymerases. This difference is what enables the current generation of antivirals, like cidofovir and ganciclovir, to be effective, as they selectively interfere with the viral polymerase. Importantly, the adenovirus polymerase has been shown to be susceptible to inhibition by a broad range of inhibitors (Naesens et al., 2005). This unique susceptibility of HAdV is worth considering for future development of therapies beyond cidofovir derivatives. A newly developed nucleoside analog, filociclovir, has been shown to be promising against HAdV in vitro and in vivo in the Syrian Hamster model for HAdV infection, discussed in Section 6.1. Prophylactic treatment of filociclovir one day prior to infection with HAdV and daily for two weeks was able to significantly reduce the late stages of viral replication. Currently, phase I clinical trials are underway for the use of filociclovir in herpesvirus treatment, giving insight into the drug’s maximally tolerated dose and adverse effects that could help inform future investigations for HAdV treatment (Toth et al., 2020).

4.4. Disruptors of HAdV Gene Expression

Given that the HAdV replicative cycle is contingent on a carefully orchestrated program of viral gene transcription, drugs that interfere with viral gene expression should cripple the replicative cycle. A high-throughput screening assay identified a set of compounds that disrupted HAdV-C5 gene expression, and these were experimentally validated to possess antiviral activity in vitro (Saha et al., 2019). Many, but not all of these compounds were also active against a range of other HAdV species. The inhibitors identified included digoxin, digitoxigenin, and lanatoside C, which are all cardiac glycosides that inhibit cellular Na⁺/K⁺ ATPases. These compounds disrupt HAdV early gene expression, DNA replication, and late gene expression. Interestingly, digoxin and digitoxigenin were previously shown to have anti-HAdV effects, but it was believed to be through disrupting splicing of E1A isoforms (Grosso et al., 2017). It is possible that these drugs exhibit a multi-facetted assault on HAdV infection. Both dexamethasone acetate and flunisolide also reduced early and late HAdV gene expression, possibly through alterations in cell signaling pathways (Saha et al., 2019).

This study also identified cytarabine as a drug that specifically reduced HAdV late gene expression in a dose-dependent fashion (Saha et al., 2019). Cytarabine, also known as arabinosyl cytosine or ara-C, is a nucleoside analog that has been used classically to block HAdV DNA replication (Feldman and Rapp, 1966; Harter et al., 1976). Since HAdV late gene expression is intimately tied to the number of viral genome templates available, it is difficult to discern whether a compound affects late gene expression itself or simply disrupts genome replication. However, it is theoretically possible to identify drugs that inhibit late gene expression without blocking viral genome replication. Currently, we are not aware of any identified drugs that function in this fashion.

The salicylanilide drugs niclosamide, oxyclozanide, and rafoxanide were recently shown to exhibit anti-HAdV effects in vitro (Marrugal-Lorenzo et al., 2019). While niclosamide and rafoxanide seem to inhibit endosomal escape of the virus, leading to reduced nuclear HAdV genome levels, oxyclozanide seems to disrupt E1A expression. Furthermore, there exists some synergistic activity between these drugs, with the strongest effect being the combination of niclosamide and oxyclozanide (Marrugal-Lorenzo et al., 2019). Interestingly, niclosamide appears to have broad spectrum antiviral activity against a variety of DNA and RNA viruses (Xu et al., 2020).

4.5. Epigenetic Regulator-Based Inhibition

The replicative cycle of HAdV is intertwined with the epigenetic landscape of the host cell (Lynch et al., 2019). Upon delivery to the nucleus, the viral genome becomes assembled with host nucleosomes, which can undergo histone modifications. Furthermore, multiple HAdV proteins interact with histone modification machinery to reprogram cellular gene expression to support optimal viral replication (Horwitz et al., 2008; Hsu et al., 2018; King et al., 2016; Mymryk and Smith, 1997).

Disrupting these epigenetic regulators has been proposed as a means of interfering with HAdV replication (Figure 1: step 4). A screen for molecules that disrupt epigenetic regulation and have anti-HAdV effects identified several drugs as potential antivirals. These include gemcitabine, a nucleoside analog used for cancer therapy that has multiple functions, including inhibiting DNA methyltransferase activity (Gray G. et al., 2012). Other inhibitors include chaetocin, which directly inhibits the SUV39H1 histone methyltransferase, and lestaurtinib, which inhibits JAK and PRK1 phosphorylation of histone methyltransferases (Greiner et al., 2005; Hexner et al., 2008). While all these compounds could disrupt HAdV replication by other mechanisms, their potential to function as antiviral agents by targeting epigenetic regulation is particularly intriguing (Saha and Parks, 2020).

Several histone deacetylase (HDAC) inhibitors, including valproic acid and SAHA (suberoylanilide hydroxamic acid, also known as vorinostat) have also been shown to disrupt HAdV replication in tissue culture models (Höti et al., 2006; Saha and Parks, 2019). Oddly, HDAC inhibitors promote latent HAdV reactivation from tonsil tissue, suggesting that this class of drugs may behave differently in vivo than expected from in vitro studies (Wang et al., 2020).

Additional small molecules that alter epigenetic modification and possess anti-HAdV activity include GSK126 and GSK343. These are inhibitors of the EZH1/2 histone methyltransferase machinery, which methylate histone 3 on lysine 27. These small molecules were able to inhibit E1A expression and subsequently supress viral infection (Arbuckle et al., 2017).

4.6. Nuclear Transport Inhibitors

The majority of the HAdV replicative cycle takes place within the nucleus (Figure 1). However, protein synthesis occurs in the cytoplasm and viral proteins must be localized to the nucleus for replication and virion assembly. In addition, HAdV has been shown to redirect key cellular regulatory proteins from the cytoplasm to the nucleus to promote viral gene expression and replication (King et al., 2018a, 2018b). Indeed, many other viruses are under the same constraints, requiring them to manipulate cellular nuclear transport machinery to complete their replicative cycles (Tessier et al., 2019). Therefore, compounds that could disrupt the flow of essential material in and out of the nucleus are strong candidates for investigation as HAdV antivirals.

Ivermectin is an anti-parasitic agent that has recently been shown to have broad spectrum antiviral activity (Tessier et al., 2019). Specifically, ivermectin appears to prevent the interaction between importin-α and the NLS of its cargo protein but does not disrupt the importin-α/β interaction (King et al., 2020). During HAdV infection, ivermectin prevents the nuclear localization of E1A and reduces DBP nuclear localization, as visualized by smaller and fewer viral DNA replication centers in infected cell nuclei. Interestingly, E1A protein levels are unchanged at early timepoints, but become completely ablated over the course of infection. Ultimately, ivermectin results in reduced genome replication and a reduction in viral gene expression, likely due to hindered E1A function (King et al., 2020).

Verdinexor, a selective inhibitor of nuclear export (SINE), also displays anti-HAdV capacity (Widman et al., 2018). Verdinexor selectively binds and inhibits the function of XPO1 (exportin 1/crm1), causing nuclear retention of many proteins. Initial investigation of the anti-HAdV capacity of verdinexor shows good efficacy but some cytotoxicity in vitro. Verdinexor should be considered for further investigation along with other SINEs (Widman et al., 2018).

4.7. Protease/Egress Inhibitors

The HAdV-encoded AVP is absolutely essential for the completion of its replicative cycle, as described in Section 2.3. Indeed, without the AVP-dependent cleavage of pTP there would be no genome replication, and without cleavage of the various structural proteins there would be no capsid maturation and virion assembly (Gupta et al., 2017). Viral proteases have classically been thought of as important targets for the development of DAA, due to their essential role across diverse viral replicative cycles and potentially druggable enzymatic activity (Patick and Potts, 1998). HAdV is no exception, and thus AVP represents a potentially important target for DAA development.

Multiple reports indicate that piperazin derivatives inhibit AVP, reduce viral transcript levels, and DNA replication (McGrath et al., 2013; Sánchez-Céspedes et al., 2016)(Sanchez-Cespedes et al., 2014). Tetrapeptide nitrile-based derivatives have also been investigated, with a novel group of compounds showing efficacy against AVP, though cytotoxicity may be a problem with these compounds (Grosche et al., 2015). Notably, little in vitro testing has been done with these novel agents in cell culture and even less is known about their pharmacological properties in vivo.

Viral egress is the final stage of the HAdV replicative cycle, in which cell lysis occurs, releasing a wave of progeny virions to trigger subsequent rounds of infection. As mentioned in section 2.3, HAdV egress and genome encapsidation are underexplored fields of research that could represent opportunities for antiviral intervention. Alternatively, the E3-encoded ADP is well-studied for its role in inducing cell death at the culmination of the late stages of infection (Figure 1: step 7). Targeting ADP could significantly handicap the virus’s ability to spread and induce viremia, potentially slowing or curbing infection. Recently, nelfinavir was identified as a potent inhibitor of HAdV in a novel microscopy-based antiviral screen (Georgi et al., 2020b). Nelfinavir was developed as an anti-HIV therapy that specifically targets the retroviral protease. In vitro studies of the effects of nelfinavir on HAdV infection showed a reduction in viral progeny by several orders of magnitude. Nelfinavir did not affect genome replication, early gene expression, or the first round of infection in general, but it reduced subsequent rounds of infection. Nelfinavir likely disrupts post-translational modifications of ADP, hindering its ability to rupture membranes to induce cell lysis (Georgi et al., 2020a).

4.8. CDK Inhibitors

Protein kinases are attractive drug targets, and many kinase inhibitors are already approved by the Food and Drug Administration (FDA), with many more in clinical trials. Although many kinase inhibitors were developed or are being developed as potential cancer therapies, they represent attractive targets for repurposing to other diseases, including viral infection (Knapp, 2018). One class of kinases that may be an attractive target with respect to HAdV antivirals are the CDKs. These are a family of highly conserved serine/threonine kinases that play critical roles in cell cycle regulation (CDK1, CDK2, CDK3, CDK4, CDK6) and transcription regulation (CDK5, CDK7, CDK8, CDK9, CDK11, CDK12, CDK13) (Asghar et al., 2015; Bregman, David, 2000; Drapkin et al., 1996; Kohoutek and Blazek, 2012; Malumbres and Barbacid, 2009; Meyerson et al., 1992). CDKs also regulate key aspects of the replicative cycles of a number of DNA and RNA viruses, such as herpes simplex virus types 1 and 2, human papillomavirus (HPV), HIV, CMV, and HAdV (Schang, 2002; Schang et al., 2006; Yamamoto et al., 2014). Specifically with respect to HAdV, CDK2 was shown to promote the replication of HAdV, while CDK9 has been suggested to enhance transcription of the HAdV E1A gene through recruitment of the transcriptional mediator complex to E1A (Alevizopoulos et al., 1998; Prasad et al., 2017; Schang, 2002; Vijayalingam and Chinnadurai, 2013).

The critical role of CDKs in cell cycle progression has made them attractive targets for the development of anti-cancer drugs, with a number of existing active compounds available that have been tested in various clinical trials. These CDK inhibitors are a family of small purines, pyrimidines, flavonoids, or bis-indoles that bind to the ATP binding pocket of their target CDK to compete with CDK’s interaction with ATP (De Azevedo et al., 1997). A number of these have been tested in a variety of ways as candidate HAdV antivirals and are summarized below.

The earliest successful CDK inhibitors were the pan-CDK inhibitors flavopiridol and roscovitine (Asghar et al., 2015; Veselý et al., 1994). Flavopiridol inhibits CDK1, CDK2, CDK4, CDK5, CDK6, CDK7, and CDK9, suppressing HAdV-C5 and HAdV-C2 replication. However, flavopiridol also had substantial effects on cellular transcription and induced cell cytotoxicity under certain circumstances (Asghar et al., 2015; V. Prasad et al., 2017; Schang et al., 2006; Thomas et al., 2002; Veselý et al., 1994). Flavopiridol has had a recent resurgence as a potential topically applicable antiviral treatment for EKC-causing HAdV-D8 and HAdV-D37 (V. Prasad et al., 2017). Roscovitine targets CDK1, CDK2, CDK5, CDK7, CDK9, and CDK12 and suppresses HAdV-E4 replication, but it halted cell cycle progression and exhibited adverse cellular effects that suggest substantial toxicity (Asghar et al., 2015; Diwan et al., 2004; Holcakova et al., 2010; Schang, 2002; Yamamoto et al., 2014).

Following the initial wave of general CDK inhibitors, other CDK inhibitors were developed with the aim of high selectivity. FIT-039 was developed as a selective CDK9 inhibitor for antiviral use and is currently in a phase I/II clinical trial for treating HPV-induced warts. FIT-039 potently inhibited DNA replication of HAdV-C5, HAdV-D53, HAdV-D19, and HAdV-C57, showing a dose-dependent effect for the first two. It also reduced E1A expression at 10 μM, while cidofovir could not (Yamamoto et al., 2014). FIT-039 had minimal effects on cellular transcription when compared to roscovitine at 10 μM and did not affect cell cycle progression or cell viability (Yamamoto et al., 2014). The authors suggested that FIT-039 suppresses viral transcription by inhibiting phosphorylation of the C-terminal domain of RNA polymerase II by CDK9 (Yamamoto et al., 2014).

CDK7 is another important target involved in cell cycle regulation, transcription, and potentially the phosphorylation of viral proteins (Hutterer et al., 2015). LDC4297 is a selective CDK7 inhibitor that inhibits HAdV-C2 replication and exerts intermediate antiviral activity in cultured cells (Hutterer et al., 2015). However, the exact mechanism underlying LDC4297-mediated CDK7 inhibition and its effects on HAdV-C2 replication have yet to be unraveled. Overall, CDK inhibitors have seen significant advancements over the past 20 years and appear to have significant potential as HAdV antivirals (Figure 1:step 4).

The field of HAdV antiviral research has exponentially expanded in recent years, as highlighted in this section. However, despite the large number of exploratory studies into single agent anti-HAdV molecules, alternative therapeutic approaches are worth considering to improve clinical benefits.

5. Emerging and Alternative Therapeutic Approaches

5.1. T-cell Immunotherapy

There is growing interest in the use of adoptive T-cell therapy to combat severe HAdV infection in hematopoietic stem transplant recipients and other immunocompromised patients. A clinical trial on the safety and feasibility of multivirus-specific T-cells from donor cord blood suggests this therapy to be safe, effective, and provide long-lasting protection against HAdV and other viruses (Abraham et al., 2019). Other studies have shown T-cells targeting E1A or hexon to be effective at treating HAdV infection or preventing HAdV reactivation in hematopoietic stem cell transplant recipients (Di Ciaccio et al., 2020; Roex et al., 2020; Tzannou et al., 2017).

5.2. Combination Therapy

The traditional approach for treating viral infections is to target a single, indispensable component of the virus, in the hopes of abrogating its ability to replicate or cause disease. Unfortunately, antiviral resistance may inevitably lead to the failure of any monotherapy (Irwin et al., 2016). Viruses are under strong selective pressure due to the evolutionary arms race between them and their hosts and can quickly adapt to new environments given their rapid reproduction time. For this reason, monotherapies are often ultimately ineffective for antiviral treatment, as the virus will be quick to evade the effects of a single therapy if their replicative capacity is not entirely suppressed (Irwin et al., 2016). Although viruses with DNA genomes have less potential to rapidly evolve resistance to antivirals, likely due to their higher fidelity polymerases compared to that of RNA viruses, they are still more than capable of evolving resistance (Irwin et al., 2016; Shafer and Chou, 2015). HAdV is no exception to this fact, as there are accounts of HAdV polymerase evolving resistance to cidofovir and brincidofovir (Chamberlain et al., 2019; Kinchington et al., 2002).

An additional problem with antiviral monotherapies is that many candidate drugs are only effective at high serum concentrations which are accompanied with significant adverse events that may result from off-target effects. Indeed, the significant nephrotoxicity of cidofovir complicates its use substantially (Caruso Brown et al., 2015; Christensen and Hermann, 2012). In addition, many of the candidate anti-HAdV drugs described in the literature and discussed in this review require effective concentrations that are unachievable in the human body without adverse consequences. Overall, the issues of antiviral resistance and effective drug serum concentration may represent unavoidable barriers to the success of monotherapy-based treatments for HAdV infection.

Importantly, an increasing body of evidence suggests that combination therapy may represent the most viable route for treating complex diseases, including viral infections. For instance, use of multi-drug antiretroviral therapy for individuals with HIV infection is standard practice (Panel on Antiretroviral Guidelines for Adults and Adolescents, 2019). There are a number of benefits to using a combination of treatments, such as synergistic effects, reduced potential for resistance to each individual therapeutic agent, and increased drug tolerance (Bayat Mokhtari et al., 2017; Hayden, 1996; Hofmann et al., 2009; Pirrone et al., 2011). The ideal situation for using combination therapy occurs when the effects of two or more therapies become synergistic, i.e. greater than additive, in their therapeutic effects. Synergistic drug combinations are more desirable because they yield a greater therapeutic effect at a given dose, compared to those combinations that are merely additive. With this increase in potency, reduced concentrations of each drug are required to achieve clinical efficacy. This leads to two beneficial outcomes of drug synergy: improved desired effects of the drugs and diminished adverse effects (Tallarida, 2011). Notably, monotherapeutic approaches that are ineffective because clinically achievable concentrations are insufficient for function can theoretically become useful at achievable concentrations through synergistic activity with other drugs. Although the idea of drug synergism is exciting, the possibility of drug antagonism must also be considered. While antagonism can sometimes be predicted through mechanistic insight, preclinical evaluations are essential to confirm a lack of apparent antagonism prior to clinical trial testing.

While it is imperative to identify drug combinations that improve the antiviral capacity of the individual agents, it is also important to identify drug combinations that mitigate the possible toxicities associated with treatment. This strategy would reduce concerns over individual drug toxicity, and still have clinically relevant efficacy. The question remains of how to identify candidates for combination antiviral therapy, and this is an exciting area for future investigation.

5.3. Drug Repositioning

Another problem with the field of antiviral therapy is that the cost and incentive for developing, testing, and approving novel antivirals is often prohibitive. Simply put, the process of new drug development is financially unattractive, burdensome, and long. This can be addressed through drug repositioning, the process of repurposing existing drugs as therapeutics for different conditions than their original purpose. Aside from saving cost and time for development, drug repositioning alleviates some risk of failure, as the drug has already undergone safety evaluation for its initial approval process (Pushpakom et al., 2019).

Experimental and computational approaches are two broad methods to identify drug candidates for repurposing. Experimental approaches consist of assays that identify drug interacting partners through techniques such as tandem affinity chromatography and mass spectrometry, and high throughput screening techniques to test drug efficacy on various model systems. Computational approaches involve comparing large datasets to identify novel opportunities for repositioning. These include comparing transcriptomic or proteomic signatures between drug-induced and diseased states, in silico molecular docking techniques to map drug structures onto 3D protein structures, and genome-wide association studies to compare diseased states to identify potentially relevant targets or pathways (Pushpakom et al., 2019). Drug repositioning is becoming an increasingly relevant strategy to circumvent the struggles of novel drug discovery and is highly applicable to the development of new antiviral therapies (Bekerman and Einav, 2015; Pushpakom et al., 2019).

5.4. Synthetic Lethality

Synthetic lethality is a genetic concept used to exploit gene or pathway dependencies and redundancies. Genes are considered essential if their removal causes an organism to become inviable. Synthetic lethality refers to a set of gene pairs which are individually non-essential, but are detrimental to survival when both are lost. Loss of one gene creates a biological condition that is uniquely sensitive to perturbation of the other gene. Synthetic lethality has been a rising concept for cancer and infectious disease therapy, providing a new approach to treatment and drug development (Huang et al., 2020; Lord and Ashworth, 2017; Tyers and Wright, 2019).

While antiviral resistance is a significant problem with monotherapy, it may be advantageous for the development of synthetic lethal combination therapies. As one example, evolution of a viral protein to escape an inhibitor can come at a fitness cost for the virus. This cost is ultimately beneficial to the virus, as it ensures survival during drug-induced evolutionary pressure, and the initial mutation can be reverted or compensated for when the pressure is removed (Irwin et al., 2016). However, the pressure of a given drug on the fitness of a virus might produce a unique environment in which the virus is particularly sensitive to a different class of drugs, leading to a synthetic lethality scenario. This presents a unique opportunity to develop combination therapy approaches to HAdV infection. Many of the drugs discussed in this review may be insufficient to cause full inhibition of HAdV replication, but they may synergize with other classes of inhibitors, and these drugs should be evaluated for synthetic lethal combination opportunities. Although monotherapeutic agents are the current focus in developing and investigating anti-HAdV agents, concepts like synthetic lethality may guide combination studies, leading to improvements in clinical utility of compounds discussed in Section 4. Furthermore, a focus on repositioning drugs could ensure a more rapid and seamless transition to clinical care.

6. Future Directions in Assisting Drug Development

The past decade has seen exponential growth in the field of HAdV antivirals. However, there is still ground to be covered before an effective therapy is clinically available. Progress in this field is intimately related to the strength and power of the tools available. Specifically, drug screening platforms and model organism systems for preclinical testing are critical tools to advance research on HAdV antivirals.

6.1. Animal Models

Various animal models have been investigated to find a suitable method of testing HAdV replication and pathology. This has been difficult because HAdV replication is largely restricted to human cells. However, HAdV infection models in pigs and New Zealand White rabbits have been developed that mirror interstitial pneumonia and keratoconjunctivitis in humans, respectively (Clement et al., 2011; Gordon et al., 1992a; Jogler et al., 2006). Cotton rats have also been used as models for HAdV-C5 lung and ocular infections, supporting evidence of antiviral efficacy of cidofovir and ganciclovir (Kaneko et al., 2004; Prince et al., 1993; Trousdale et al., 1994). Interestingly, Yorkshire pigs have recently been identified as a promising large animal model for HAdV infection (Koodie et al., 2019). Many aspects of acute respiratory disease seen with HAdV can be replicated in mouse models using murine adenovirus in immunocompetent mice and in a mouse model of bone marrow transplantation (McCarthy et al., 2015; Weinberg et al., 2005).

Although these models each have their own strengths and weaknesses, their utility is eclipsed by that of the Syrian hamster, which has been a particularly insightful model for testing HAdV pathogenesis and antiviral treatment. Hamsters are treated with cyclophosphamide to induce an immunosuppressed state where infection with HAdV closely mimics the pathogenesis of HAdV in immunocompromised humans (Toth et al., 2008). This model has been used in preclinical testing of many different putative anti-HAdV compounds and even anti-HAdV microRNA (Naesens et al., 2005; Schaar et al., 2017; Toth et al., 2020, 2018, 2015, 2008; Wold et al., 2019; Ying et al., 2014).

Despite these existing models, the identification, characterization, and development of better immunocompetent animal models for testing anti-HAdV agents remains a key unmet need for future progress.

6.2. Cell-based Screening Approaches

In addition to individually testing candidate antiviral compounds with suspected activity, multiple screening approaches have been developed to identify compounds with anti-HAdV activity in cell-based assays. These methods make use of libraries of either pre-existing clinically approved drugs, or small molecules to be used as lead compounds to identify candidates for HAdV treatment.

In 2010, a fluorescence-based whole-cell screen was developed using a replication competent HAdV-B11 vector with the green fluorescent protein (GFP) reporter gene driven by a CMV promoter (CMV-GFP) inserted in the E1 region (Andersson et al., 2010). This system was used to screen a library of 10 000 compounds for hits that showed an 80% reduction in GFP expression and less than 50% cell death. This screen would identify both inhibitors of entry and the early phase of HAdV infection. Of the 400 hits initially identified, dose-response testing narrowed the candidates to 24 dose-dependent inhibitors. One compound was especially potent, referred to as A02, and was validated to prevent infection of A549 cells by representative HAdV types from species A-F at μM concentrations. A02 did not block virus attachment, but did block viral DNA replication and late gene expression via an unspecified mechanism (Andersson et al., 2010).

Similarly, Saha et al. applied another fluorescence-based screening technique in 2019, this time using red fluorescent protein (RFP) under the expression of the MLP in a replication competent HAdV-C5 platform (Saha et al., 2019). The authors used RFP expression as a surrogate for late gene expression, and therefore viral replication. A library of 1200 FDA-approved compounds were tested using a constant drug concentration of 250 nM for the initial screen. Interestingly, eleven compounds which decreased RFP expression in a dose-dependent fashion were identified, as well as multiple compounds that increased RFP expression. Those compounds which inhibited RFP expression included cardiotonic steroids and several cancer chemotherapy agents as mentioned in Section 4.4 (Grosso et al., 2017; Saha et al., 2019).

In 2020, a new image-based drug screening technique for compounds that block HAdV cell transmission was reported (Georgi et al., 2020b). Specifically, the screen involved the use of automated fluorescence imaging for multiple rounds of infection to identify compounds with anti-HAdV activity at any stage of the viral replicative cycle. A HAdV-C2 vector with CMV-GFP in place of the E3B gene was used to infect A549 cells. The authors measured infected cell nuclei, number of plaques, and levels of GFP signal. Of the 1280 clinical or preclinical compounds screened, they identified nelfinavir as the best inhibitor of HAdV replication. This drug specifically functioned to block subsequent rounds of infection, not the initial round (Georgi et al., 2020b).

Based on these existing results, cell-based high-throughput studies have clear potential to identify small molecules that exhibit antiviral activity, including currently approved drugs that could be repositioned. The relative ease of construction of recombinant HAdVs creates a multitude of different possible screens that could be developed to test libraries in addition to the early, late, or transmission screens described above.

6.3. Omics-based Approaches

An alternative approach to screening for drugs with anti-HAdV activity is to identify cellular targets that represent virally-induced weaknesses of the host cell (Mast et al., 2020). The workflow utilized by Wang et al. in 2018 exemplifies how to identify these potential weaknesses for exploitation as novel HAdV therapies. The authors identified a set of genes which had differential expression due to HAdV infection at 6 and 12 hours post-infection (hpi). Further filtering of these genes for biological relevance allowed the identification of upstream regulatory factors representing potential targets for therapeutic intervention. Specifically, retinoic acid receptor β (RARβ) expression was consistently repressed during infection. Knockdown and overexpression of RARβ resulted in increased and decreased HAdV progeny production respectively, confirming its importance in the HAdV replicative cycle. Tazarotene, a specific RARβ agonist, was used to further validate their model (Wang et al., 2018). Taken together, this study provides proof-of-concept that “omics” technologies combined with bioinformatics can identify cell targets or pathways that can be modulated pharmacologically to control HAdV infection. These approaches could be particularly effective in combination with drug repurposing or for identifying new targets for novel drug development.

With relevant animal models and powerful screening techniques as tools, the field of anti-HAdV therapy has the potential to evolve rapidly in the coming years, hopefully translating to a new line of therapeutic options for individuals who suffer from severe HAdV infections.

7. Conclusions

HAdVs are ubiquitous infectious agents that cause significant morbidity and even loss of life, particularly in select groups of at-risk individuals. Troubling outbreaks of HAdV infection occur far too frequently, and these can also affect otherwise healthy individuals not normally at risk for serious infection. The realization that we need more effective and safer antiviral drugs for HAdV infection seems to have passed a critical threshold in recent years, and the surge of studies in the field of anti-HAdV development is highly encouraging.

Highlights.

• Adenovirus infections are associated with significant morbidity and mortality, especially for immunocompromised individuals.

• Existing antiviral therapy options for adenovirus infections are limited.

• Many new inhibitors of adenovirus infection have been recently identified.

• Rewiring cellular pathways by virus–host, protein–protein interactions may create chemically addressable vulnerabilities otherwise absent in healthy cells and tissues.

• Repurposing existing approved drugs to treat adenovirus infections is a promising approach.