2. Adenoviruses
2.1. General Properties
Adenoviruses are double-stranded DNA viruses with a non-enveloped, icosahedral capsid of approximately 90 nm in diameter (
Figure 1). Their prevalence in the population is high, and they are responsible for common diseases such as rhinitis, gastroenteritis, and conjunctivitis. The family of human adenoviruses is divided into species A–G, which are further classified into many different serotypes
[17][18]. The majority of knowledge on adenovirus biology is based on the extensive research of HAdV-2 and HAdV-5, which both belong to subgroup C. In addition to humans, different adenoviruses have been isolated from various other vertebrates, such as reptiles, birds, and mammals. No adenovirus identified in other species has been shown to cause clinically significant disease in humans
[18].
Figure 1. Adenovirus and its structural components. The double-stranded adenovirus DNA is encapsulated by an icosahedral capsid, formed by 252 major capsomers named hexons and pentons. The faces are formed by the hexons, whereas the pentons serve as the bases for fiber shafts that protrude from each vertex and culminate into tip structures called fiber knobs. The fiber proteins are responsible for the interactions between the adenovirus and the host cell receptors. In addition to the major proteins, four minor capsid-associated proteins stabilize the capsid structure. The five core proteins are associated with the adenoviral DNA. Adenovirus protease catalyzes the maturation of the precursor forms of the capsid proteins IIIa, VI, and VIII, as well as the DNA-associated proteins VII, μ, and terminal protein.
Adenovirus infection is typically self-resolving and usually causes only mild upper respiratory tract symptoms and fever in immunocompetent individuals. Up to half of all adenovirus infections are asymptomatic or cause only very mild symptoms
[17][19]. Adenovirus infections account for approximately 5 to 10% of febrile infections in children and neonates, but their frequency is significantly lower in adults
[19]. The incubation time for adenovirus infection is approximately six days
[20]. Antibodies against HAdV-5 can be measured in nearly 85% of the population
[21].
The most critical risk groups for adenovirus infections are neonates and immunocompromised individuals, in which the adenoviruses may cause severe pneumonia and diseases affecting organ systems other than the respiratory tract, depending on the level of immunocompetency
[9][17][18]. In individuals with normal immune function, adenoviruses cause restricted focal infections, the location depending on the transmission route. There is variation in the tissue tropism between different adenovirus serotypes, and adenoviruses can theoretically replicate locally in several different organs
[18]. Disseminated adenovirus infections are mainly seen in immunocompromised patients and are usually connected with other severe systemic symptoms
[22].
2.2. Adenovirus Infection
The adenovirus infection can be mediated through a variety of receptors. Some adenovirus serotypes infect host cells by attaching to the coxsackie-adenovirus receptors (CARs) with their fiber knobs. The CARs are present in the tight junctions between epithelial cells in many organs. This primary interaction is followed by the binding of the pentons to the α
vβ
3 and α
vβ
5 integrins on a cell surface, which initiates the internalization of the virus via clathrin-dependent endocytosis. There are also serotypes that bind to the membrane proteins belonging to the cluster of the differentiation system and some other surface structures
[18][22].
Cytolysis, which takes place at the end of the viral replication cycle, and the subsequent virus release from the host cell are thought to be the result of the viral hijacking of the host translational machinery and accumulation of mature viruses inside the host cell. The proteins encoded by the regions E1, E2, and E4 enable adenovirus replication in the host cell (
Figure 2A). In addition, adenovirus death protein (ADP) encoded by the E3 region plays a crucial role in cytolysis, and its overexpression increases viral spread into adjacent cells
[23].
Figure 2. Genomic organization of the adenovirus and its vector derivatives. A schematic representation of the wild-type adenovirus genome and its transcription units is shown on top (A). The double-stranded adenoviral DNA (purple arrows) is flanked by inverted terminal repeats (vertical blue lines), and the packaging signal (Ψ) is at the left end of the genome. Early genes (E1–E4) are presented as blue arrows, and late genes (L1–L5) as red arrows with their respective transcription directions. Major late promoter (MLP) directs the transcription of the late genes. The picture below (B) presents the adenoviral vector generations with their respective genomic deletions (Δ), highlighted by the blue color over the adenovirus DNA (purple line). Under the deletion map, insertion spaces for the gene of interest (GOI) are shown. Inverted terminal repeats and packaging signals are present in each generation. Third-generation vectors are devoid of all coding adenoviral DNA (AdDNA). Additional stuffer DNA is inserted to reach the genomic size required for the encapsidation of the vector. The range for the maximal transgene packaging capacity (kb) is presented on the right.
The tropism of adenoviruses varies greatly depending on the adenovirus species. Viruses from species A, F, and G infect gastrointestinal epithelial cells. Species C, E, and part of species B have tropism to the respiratory tract, and the rest of species B to the urinary tract. In contrast, adenovirus species D infects the conjunctival surface of the eye. HAdV-5 belongs to the species C and has thus natural tropism to the respiratory epithelial cells, but also to the hepatocytes
[18].
Intravenous administration of wild-type HAdV-5 at a dose of 2 × 10
12 vp results in biodistribution mainly in the lungs and liver in a porcine model. In addition, minor biodistribution occurs in the kidney, heart, skeletal muscle, brain, and gonads, but the concentration of the viral DNA is many thousand-fold lower than in the lungs
[24]. Using immunohistochemical and hybridization-based assays, it has been shown that infection by certain adenovirus serotypes can remain latent in the tonsils, adenoids, and lungs
[25].
Wild-type adenovirus does not integrate into the host genome in replication-permissive cells, and the infection of the host cell leads to its inevitable death by cytolysis at the end of the adenoviral replication cycle
[26]. Some adenovirus serotypes have been found to be oncogenic in rodent cells
[27]. Adenovirus serotypes coexisting in the same individual can recombine and form hybrids causing new infections, especially in immunocompromised but also in healthy individuals
[28][29].
2.3. Environmental Aspects
Adenoviral shedding from an infected person takes place via respiratory aerosols and feces. Horizontal transmission can occur, for example, if viruses enter the respiratory system or the conjunctiva via droplets. Transmission by the fecal-oral route is also possible, especially in children
[18].
Adenoviruses of serotypes 5, 8, and 19 can survive for up to 49 days on environmental plastic and metal surfaces
[30]. A study comparing twenty-one disinfectants concluded that mixtures containing at least 1900 ppm of chlorine, 65% ethanol, and 0.63% quaternary ammonium compounds or 79% ethanol and 0.1% quaternary ammonium compounds should be used to disinfect surfaces contaminated with serotype 8 adenoviruses
[31].
Although the efficacy of some antiviral drugs that alter the function of DNA polymerase has already been tested in clinical trials, no antiviral drugs for adenoviral infections have yet been approved for clinical use
[22]. On the other hand, enteric vaccines based on replication-competent wild-type adenoviruses have been used to prevent acute respiratory distress syndrome in some risk groups, for example, in military personnel. These vaccines have been proven to be safe and well-tolerated
[32]. After administering an adenovirus vaccine for serotype 4, fecal shedding of adenovirus takes place for 2 to 3 weeks, but the transmission to unvaccinated individuals using the same facilities, such as toilets, has not been shown to take place
[25][33].
3. Adenovirus Vectors
Based on the extent of their genetic modifications, the replication-deficient AdVs are categorized as first-, second-, or third-generation constructs. Additionally, conditionally replicating or oncolytic AdVs form another distinct and important AdV group.
3.1. First-Generation Adenovirus Vectors
First-generation AdVs are the simplest of all adenoviral constructs. They contain a genomic deletion of the E1 region and an expression cassette of the preferred transgene (Figure 2B). The E1 region consists of two separate E1A and E1B transcription units. These units encode proteins to bring the infected cell into the state that allows viral particles to be produced and evasion of the virus from the host immune response activated by the synthesis of adenovirus DNA.
Two of the most abundant proteins encoded by the E1A unit are E1A 12S and 13S. The proteins are identical in their amino acid sequences apart from a 46 amino acids long segment in the middle of the polypeptide 13S. 12S and 13S regulate viral replication by binding to multiple cellular proteins, thereby shifting the cell cycle from the G
0 phase to the S phase
[17]. The E1B unit mainly produces two proteins called E1B 19K and E1B 55K. E1B 55K inhibits the tumor protein p53-induced apoptosis and increases the degradation of p53 as a complex with the E4 open reading frame (ORF) 6 protein
[34][35]. E1B 19K also contributes to the inhibition of apoptosis by preventing the induction of tumor necrosis factor (TNF) α and Fas-mediated cell death pathways
[17]. E1-deleted AdVs cannot replicate in human cells because the genomic region encodes transcription factors essential for adenovirus replication. If some vectors regain the deleted E1 region by recombination, they can replicate nearly as efficiently as the wild-type adenoviruses
[12].
In addition to the replication enabling E1 region, the E3 region can also be deleted to produce more space for genomic insertions. Although not necessary for viral replication, the seven proteins encoded by the E3 region contribute to the protection of the virus from the host immune responses, apart from E3 12.5K, whose function is still unknown, and ADP, which serves as a factor to aid the cytolysis and release of the mature virions from the cell during the late infection
[36][37]. From the rest of the proteins, E3 gp19K prevents the introduction of antigens by a major histocompatibility complex 1 allowing the infected cell to escape from the cytotoxic T cells
[38], E3 receptor internalization and degradation (RID) α and β proteins form a complex on the membrane of the infected cell that inhibits TNFα and Fas-mediated apoptosis
[39], E3 6.7K forms a complex with RID β, inhibiting the apoptotic cascade induced by TNF-related apoptosis-inducing ligand
[40], and E3 14.7K participates in the inhibition of apoptosis caused by TNFα and Fas
[41][42].
3.2. Second-Generation Adenovirus Vectors
Second-generation AdVs have the genomic deletions of the first-generation AdVs and additional genomic deletions in the regions E2 or E4, intending to reduce immunogenicity and increase the transgene packaging capacity even further (
Figure 2B). The three translation products of the E2 region, divided into the promoter-proximal region E2A and the distal E2B, participate in DNA replication and are denoted as DNA-binding protein, precursor terminal protein, and adenoviral DNA polymerase
[43]. The E4 region encodes at least seven different ORFs. These proteins have diverse functions, ranging from transcriptional regulation to the augmentation of late protein synthesis
[44][45].
Despite the elegant rationale behind the additional modifications of second-generation AdVs, it seems that in terms of immunogenicity and toxicity, the deletion of the regions E2 and E4 offers only a limited benefit compared to omitting only the E1 and E3 regions
[46][47][48]. Moreover, the transgene expression seems to be unstable
[48]. In fact, at least one study reported that E4 deletion in the first-generation AdV enhances endothelial apoptosis compared to the control vectors
[49]. On top of having no apparent benefit over the first-generation AdVs concerning safety and efficacy, the production yields of the second-generation vectors remain lower
[50].
3.3. Third-Generation Adenovirus Vectors
Finally, the third-generation AdVs, also called gutless, gutted, helper-dependent, or high-capacity AdVs, have been modified to be devoid of all adenovirus DNA, aside from the sequences mandatory for the packaging of the expression cassette (
Figure 2B). This also includes the late-phase transcription regions L1–L5, which are all controlled by the same major late promoter and encode proteins forming the capsid and participating in the assembly, genomic packaging, and regulation of the maturing virions
[51].
Third-generation AdVs are attractive vectors for gene therapy. Due to up to 35–37 kB of removed adenoviral DNA, they have room for a larger transgene cassette and, at least in theory, are also less immunogenic and capable of more extended transgene expression
[52]. Compared to the first- and second-generation, the production line of the third-generation AdVs is more complex and, as the name “helper-dependent” implies, requires a separate helper virus to provide all the necessary proteins for vector propagation
[50]. However, the additional effort might be worthwhile, as the current evidence supports the capability of the third-generation AdVs to produce substantially longer-lasting transgene expression than the first-generation AdVs with less inflammation
[53][54][55].