MicrobiologyBytes: Virology: Adenoviruses Updated: August 31, 2010 Search


MicrobiologyBytes: Latest Updates


Adenoviruses are a frequent cause of acute upper respiratory tract (URT) infections, i.e. "colds". In addition, they also cause a number of other types of infection. They were first isolated in 1953 by investigators trying to establish cell-lines from adenoidal tissue of children removed during tonsillectomy and from military recruits with febrile illness.
In 1962, some Adenoviruses were shown to cause tumours in rodents - this caused a considerable panic! (N.B. Adenovirus oncogenesis appears to be associated with abortive infections and has never been observed in humans.)
During investigation of the Adenovirus genome and gene expression, many techniques were developed which were subsequently used to examine other viruses/cellular genes - these viruses are an important model system for the understanding of many other viruses. Some characteristic features of Adenoviruses are:

There are at least 51 human adenovirus serotypes (genus Mastadenovirus) which have been divided into subgroups – (A to F) based on their capacity to agglutinate erythrocytes of human, rat and monkey as well as on their oncogenicity in rodents. The serotypic origin of the E1A gene determines the oncogenic phenotype of adenovirus-transformed cells. Viruses belonging to subgroup A (such as adenovirus 12, Ad12) induce tumours with high frequency and short latency, while viruses from subgroup B (such as Ad3 and Ad7) are weakly oncogenic. Adenoviruses from subgroup C (which includes the well studied serotypes Ad2 and Ad5), D, E and F are non-oncogenic. All human adenoviruses studied so far can transform primary rodent cells in culture, however, only cells transformed by viruses of subgenus A and B are oncogenic in newborn rodents, paralleling the oncogenic properties of the parental viruses.




Type Species




Ovine adenovirus D



Fowl adenovirus A



Human adenovirus C



Turkey adenovirus B



Adenovirus particle

Detailed three dimensional structural models of adenovirus particles have been constructed based on a combination of cryoelectron microscopy and X-ray crystallography. There are at least 13 proteins in the Adenovirus capsid:

Name: Location: Known Functions:
IIHexon monomerStructural
IIIPenton basePenetration
IIIaAssociated with penton basePenetration
IVFibreReceptor binding; haemagglutination
VCore: associated with DNA & penton baseHistone-like; packaging?
VIHexon minor polypeptideStabilization/assembly of particle?
VIIIHexon minor polypeptideStabilization/assembly of particle?
IXHexon minor polypeptideStabilization/assembly of particle?
TPGenome - Terminal ProteinGenome replication
Mu Nucleoprotein Genome replication?
IV2a Nucleoprotein Genome packaging
Protease Associated with pentons? Maturation

All Adenovirus particles are similar: non-enveloped, 60-90nm diameter. They have icosahedral symmetry easily visible in the electron microscope by negative staining and are composed of 252 capsomers: 240 "hexons" + 12 "pentons" at vertices of icosahedron (2-3-5 symmetry).
Individual protomers can be isolated by progressive chemical disruption of purified virus particles. The hexons consist of a trimer of polypeptide II with a central pore; VI, VIII and IX are minor polypeptides also associated with the hexon, thought to be involved in stabilization and/or assembly of the particle. The pentons are more complex; the base consists of a pentamer of peptide III, 5 molecules of IIIa are also associated with the penton base. The pentons have a toxin-like activity, purified pentons causing c.p.e. in the absence of any other virus components (a unique property). A trimeric fibre protein extends from each of the 12 vertices (attached to the penton base proteins) and is responsible for recognition and binding to the cellular receptor. A globular domain at the end of the adenovirus fiber is responsible for recognition of the cellular receptor.

To view a negatively-stained electron micrograph of adenovirus particles, click here. The thin fibres protruding from each vertex of the icosahedral particle are just visible (look closely!) and the triangular faces of the icosahedral particle can be made out. Image reconstruction of a type 2 adenovirus particle. (Crystal Structure of Human Adenovirus at 3.5 Resolution. (2010) Science 329(5995): 107 -1075 doi: 10.1126/science.1187292)


Linear, non-segmented, d/s DNA, 30-38kbp (size varies from group to group) which has the theoretical capacity to encode 30-40 genes. Genome structure (cross-hybridization, restriction map) is one of the characters used to assign viruses to groups (70-95% homology within groups, 5-20% homology between groups).

Adenovirus genome


Replication of all Adenoviruses is similar and occurs in the NUCLEUS:
Adenovirus infection

Replication is divided into EARLY and LATE phases, the latter defined as beginning with the onset of DNA replication (N.B. this division is characteristic of the replication of DNA viruses!). Attachment to cells is rather slow, taking several hours to reach a maximum.

Uptake of the adenovirus particle is a two stage process involving an initial interaction of the fibre protein with a range of cellular receptors, which include the MHC class I molecule and the coxsackievirus-adenovirus receptor. The penton base protein then binds to the integrin family of cell surface heterodimers allowing internalization via receptor-mediated endocytosis. Most cells express primary receptors for the adenovirus fibre coat protein, however internalisation is more selective. (Virus yoga: the role of flexibility in virus host cell recognition. Trends Microbiol. 2004 12:162-169).

PENETRATION involves phagocytosis into phagocytic vacuoles, after which the toxic activity of the pentons is responsible for rupture of the phagocytic membrane and release of the particle into the cytoplasm.

Adenovirus replication

Uncoating follows an ordered sequence, first the pentons, releasing a spherical, partially uncoated particle into the cytoplasm. The core migrates to the nucleus where the DNA enters through nuclear pores, whereupon it is converted into a virus DNA-cell histone complex.

Bestsellers - Music - DVDs - Videos - Electronics
Search for ... (keywords):
Search for ... (keywords):

Bestsellers - Music - DVDs - Videos - Electronics

Gene Expression:

Adenovirus gene expression

Before and independently of genome replication, immediate early and early mRNAs are transcribed from the input DNA. Transcription of the Adenovirus genome is regulated by virus-encoded trans-acting regulatory factors. Products of the immediate early genes regulate expression of the early genes. Early genes are encoded at various locations on both strands of the DNA (l = "leftward strand" and r = "rightward strand"). Multiple protein products are made from each gene by alternative splicing of mRNA transcripts - splicing was first discovered in Adenoviruses (Sharp, 1977).
Flint, J. and Shenk, T. (1997) Viral transactivating proteins. Ann. Rev. Genet. 31: 177-212.

Phase: Genes Transcribed:
Immediate earlyE1A
EarlyE1B, E2A, E2B, E3, E4, some virion proteins
LateLate genes, mostly virion proteins

The first mRNA/protein to be made (~1h after infection) is E1A. This protein is a trans-acting transcriptional regulatory factor whose precise mode of action is not known (not a DNA-binding "transcription factor") but is necessary for transcriptional activation of early genes. The protein is also capable of activating transcription from a variety of other viral and cellular promoters and shows no sequence-specificity, rather a modification of the cellular environment.
The second protein made is E1B. E1A + E1B together (and independently of other virus proteins) are capable of transforming primary cells in vitro (especially Ad5, Ad12).

Transformation is: "A CHANGE IN THE MORPHOLOGICAL, BIOCHEMICAL OR GROWTH PARAMETERS OF THE CELL", which may or may not result in cells which are able to produce tumours in experimental animals ( = NEOPLASTIC transformation).

E1A and E1B

The activities of the two proteins can be dissected out by molecular techniques:

E1A: Can immortalize primary cells in vitro.
E1B: Does not transform cells on its own, but "co-operates" with E1A to stably transform cells.
E1A + E1B: Necessary for full transformation and tumour formation in animals.
Three highly conserved domains or regions, CR1, CR2 and CR3 have been identified in all the human Ad E1A genes and their functions defined by mutational analysis. Mutants in CR1 complement mutations in CR2, demonstrating discrete functional domains that act in trans to alter the host cell. CR1 is involved in binding of a large, cellular phosphoprotein, p300/CPB, which is a transcriptional co-activator with intrinsic histone acetyltransferase activity. E1A mutants that fail to bind p300 also fail to stimulate cellular DNA synthesis in resting cells and do not immortalise cells in culture. The CR1 segment between residues 40-60 forms part of the binding site for the retinoblastoma susceptibility protein pRb and related proteins (p107 and p130). Another region necessary for binding and functional inactivation of these proteins is CR2, which explains why CR2 is absolutely required for the mitogenic activity of E1A.

E1A binds to a cellular protein, p105-RB, the product of the retinoblastoma gene (retinoblastomas result when this gene is deleted or damaged, hence it is an "anti-oncogene" or "tumour suppressor"). The retinoblastoma susceptibility gene product, pRb, and related proteins (p107 and p130) form complexes with a transcription factor, E2F, that are dissociated by E1A to release transcriptionally active E2F. Expression of E2F can stimulate DNA synthesis in quiescent cells. It has also been established that other host cell cycle control proteins, e.g. cyclin A, bind to CR2. E1A proteins can bind p300 and the pRb related proteins simultaneously, thus potentially facilitating their interaction. Loss of Rb function also results in the induction of p53, by increasing the expression of the tumour suppressor protein p14ARF, which prevents MDM2 controlled degradation of p53. This is one of the mechanisms by which E1A causes apoptosis in primary cells.

E1B binds to p53, another tumour suppressor involved in the control of the cell cycle. N.B. Binding of DNA virus nuclear proteins to cellular tumour suppressors is a shared mechanism of cell transformation found in several virus families. The E1B region encodes two proteins of 19 kDa and 58 kDa that are unrelated in amino-acid sequence, but both have anti-apoptotic function. These two proteins act independently and by different mechanisms. The E1B 58 kDa protein directly abolishes the activity of the tumour suppressor, p53, the most commonly mutated protein in human cancer. The p53 protein is a transcription factor that negatively regulates cell proliferation. It is thought that p53 is not constitutively involved in the cell cycle, but is activated in response to DNA damage. When p53 is in a complex with E1B-58K protein, it is unable to transactivate transcription. Mutants of E1B-58K that are unable to form a complex with p53 are also unable to transform cells in culture in cooperation with E1A. The E1B 19-kDa protein inhibits apoptosis by a mechanism similar to that of human bcl-2 protein.

Adenoviruses also encode several other proteins that modulate immune-mediated apoptotic mechanisms. Most of these are derived from the E3 transcription unit. There are seven E3 proteins, none of which is required for replication in cultured cells, implying anti-immune functions. E3-gp19K (a 19 kDa glycoprotein coded by the E3 transcription unit) is the first line of defense against CTL, binds to all haplotypes of human class I antigens and is conserved in all respiratory adenoviruses. Adenoviruses which infect the gastrointestinal tract lack this protein, but they downregulate class I molecules by repressing at the transcriptional level, repression is meditated by the Ad-coded E1A proteins. Other E3-coded proteins named RID and E3-14.7K inhibit some of the killing pathways induced in infected cells by CTL, those that involve apoptosis rather than the perforin-granzyme pathway. E3 has therefore been called the "stealth" gene, allowing adenoviruses to evade the host immune response.

Together, these observations indicate that Adenoviruses, in the course of sequestering cellular machinery and altering the intracellular environment to favour viral replication, have profound effects on cellular functions. Viewed in this light, transformation is just an accidental (and rare) outcome of infection. The basis for oncogenesis (c.f. immortalization of cells in vitro - above) is not clear, but it is known that Ad12 E1A turns off class I MHC expression, possibly allowing tumours to escape destruction by CTLs.

DNA Replication:

Adenovirus DNA replication

Adenovirus DNA replication has been studied extensively both in vivo (t.s. mutants in infected cells) and in vitro (nuclear extracts). At least 3 virus-encoded proteins are known to be involved in DNA replication:

In addition, many cellular proteins in the nucleus also participate in replication of the genome (e.g. NFI, NFII, topoisomerase I).
The adenovirus genome has inverted terminal repeats (ITRs) of about 100 bp. Located within the ITRs are the cis-acting DNA sequences which define ori, the origin of DNA replication. Covalently attached to each 5' end is a terminal protein (TP) which is likely to be an additional cis-acting component of ori. Within the terminal 51 bp of the adenovirus 2 genome, four regions have been defined that are involved in initiation of replication. The terminal 18 bp are regarded as the minimal replication origin and these sequences can direct limited initiation with just the three viral proteins involved in replication: preterminal protein (pTP), DNA polymerase (pol) and DNA binding protein (DBP). However, two cellular transcription factors, nuclear factor I (NFI) and nuclear factor III (NFIII) are required for efficient levels of replication. In contrast adenovirus 4 replicates efficiently without NFI and NFIII. A further cellular factor, a topoisomerase, is required for complete elongation.

The steps involved in adenovirus DNA replication can be summarized as follows:

Although 2 components of the adenovirus replication system have been crystallized individually, namely DBP and POU (NFIIIDBD), we are still a long way from having a structural model for the whole preinitiation complex.

Late Transcription:

At the onset of DNA replication, the pattern of transcription changes radically from the early to the late genes. There is cis-acting control of this switch, i.e. only newly replicated DNA is used for late gene transcription, but the mechanism controlling this is not understood. Late phase transcription is driven primarily by the major late promoter. Although transcription from this promoter is complex involving multiple polyadenylation signals and an elaborate usage of RNA splicing, five gene clusters can be defined (L1-L5). Late phase gene expression is primarily concerned with the synthesis of virion proteins.
Assembly occurs in the nucleus, but begins in the cytoplasm when individual monomers form into hexon and penton capsomers. Empty, immature capsids are assembled from these protomers in the nucleus, where the core is formed from genomic DNA + associated core proteins.
Although host cell macromolecular synthesis ceases earlier in the infection, infected cells remain intact and do not lyse (disruption of cytoskeleton changes shapes of infected cells, which round up). Unlike other viruses, such as HHV-1, there is no specific pathway for virus release, although prolonged infection will eventually lead to cell lysis which may be facilitated by the adenovirus death protein (E3-11.6K), which also induces apoptosis. Virus particles tend to accumulate in the nucleus and are visible in the microscope as eosinophilic crystals - INCLUSION BODIES. These are thought to be the basis of latent infections - reactivation is caused by accidental lysis of infected cells, releasing virus particles from the nuclei - effectively a re-infection. More properly, this type of mechanism of persistence is known as "occult" (hidden) infection, rather than "latent" (c.f. Herpesviruses).
Adenoviruses are known to interact with other viruses, notably defective Parvoviruses (Adeno Associated Viruses). SV40 (Polyomavirus) has also been shown to act as a helper virus for Adenoviruses - co-infection overcomes a late block to Adenovirus replication in certain cell types which are normally non-permissive. This may involve functional substitution of SV40 T-antigen for Ad DBP (?). Infectious Adeno-SV40 hybrids have been isolated by rescuing Adenovirus deletion mutants by super-infecting with SV40, indicating functional overlap between the two families.


Certain types of Adenovirus are commonly associated with particular clinical syndromes:
Disease: At Risk:
Acute Respiratory Illness Military recruits, boarding schools, etc.
Pharyngitis Infants
Gastroenteritis Infants
Conjunctivitis All
Pneumonia Infants, military recruits
Keratoconjunctivitis All
Acute Haemorrhagic Cystitis Infants
Hepatitis Infants, liver transplant patients

Note that this list includes relatively common infections (at the top) and some rare infections (bottom). Most Adenovirus infections involve either the respiratory or gastrointestinal tracts or the eye.

Adenovirus infections are very common, most are asymptomatic. Most people have been infected with at least 1 type at age 15. Virus can be isolated from the majority of tonsils/adenoids surgically removed, indicating latent infections. It is not known how long the virus can persist in the body, or whether it is capable of reactivation after long periods, causing disease (it is hard to isolate this occult virus as it may be present in only a few cells). It is known that virus is reactivated during immunosuppression, e.g. in AIDS and transplant patients. Pre-existing adenovirus infection is a major problem in the rejection of transplanted hearts and lungs: Shirali GS, et al. N Engl J Med 2001;344:1498-1503,1545-1547.

A characteristic feature of adenovirus infection is their ability to subvert the host immune response by using the early E3 cassette of genes (Immune evasion by adenovirus E3 proteins: exploitation of intracellular trafficking pathways. 2004 Curr Top Microbiol Immunol 273, 29–85). The E3 membrane proteins are able to downregulate critical recognition structures for the host immune system from the cell surface. The molecular mechanism for removal depends on the E3 protein involved: E3/19K prevents expression of newly synthesized MHC molecules by inhibition of ER export, whereas E3/10.4-14.5K down-regulate apoptosis receptors by rerouting them into lysosomes.


None. Inactivated vaccines have been developed and are routinely used for military recruits in some countries (notably the USA). This is because adolescents and others in close daily contact are at risk for epidemic spread of respiratory infections - risk to general population is so low that vaccination is not a viable proposition.
Beginning in 1971, all US military recruits were vaccinated against adenovirus, but the sole manufacturer of the vaccine halted production in 1996. Since 1999, approximately 10% to 12% of all recruits have become ill with adenovirus infection in basic training and several death have been attributed to adenovirus infection. New vaccine lots are in preparation but will not be available for several years.

In recent years, there has been considerable interest in developing Adenoviruses as defective vectors to carry and express foreign genes for therapeutic purposes. One reason for this is that the Adenovirus genome is relatively easily manipulated in vitro (c.f. Retroviruses) and the genes coupled to the late promoter are efficiently expressed in large amounts.

MicrobiologyBytes: Adenovirus vectors - new genes, new vaccines

Adenovirus vectors may be either replication-competent or replication-deficient. In replication-competent adenovirus recombinants, transgenes are usually inserted in the E3 region and very high levels of expression can be achieved with transcription driven by the late promoter. Replication-deficient adenovirus vectors are usually based on adenovirus E1-E3 deletion mutants. Deletion of the E1 gene region (both E1A and E1B) permits vectors to accommodate significantly larger inserts, removes a region of the genome associated with cellular transformation and generates a vector that is not only replication-deficient but also cannot activate early phase gene expression. This defect in early phase gene expression generates a vector which can promote efficient transgene delivery and expression in the absence of significant vector gene expression. However, breakthrough to early and late phase gene expression can occur, particularly following infection at high m.o.i. or if the target cell expresses an endogenous "E1A-like" function. Breakthrough expression from the vector genome has been associated with the induction of an immune response following in vivo gene delivery. Replication-deficient adenovirus vectors incorporating additional mutations or deletions have been generated to address this problem. The deletion of additional sequences from the vectors also permits the insertion larger transgenes.

Vector systems have been developed in which most or all adenovirus protein coding sequences are removed. From the viewpoint of safety, size of transgene insertion and absence of vector gene expression, these so-called 'gutless' adenovirus vectors are attractive. They contain the inverted terminal repeats (ITRs) and adjacent sequences essential for replicating the adenovirus genome, including the cis-acting signal sequence (psi) which directs packaging of adenovirus DNA into virus particles. The gutless vectors require virtually all adenovirus gene functions to be provided for vector propagation and, since this cannot yet be achieved using a packaging cell line, they must be provided using a helper cell. Although their potential is enormous, current gutless vectors are complex and technically difficult to use. However, there are dangers associated with the use of adenovirus vectors, and a few deaths have occurred in clinical trials.

Adenoviruses: update on structure and function. J Gen Virol. 2009 90(1): 1-20

Update on adenovirus and its vectors. J. Gen. Virol. (2000), 81: 2573-2604.

Roulston, A. et al. (1999) Viruses and apoptosis. Ann. Rev. Microbiol. 53: 577-628

Yewdell, J.W. and Bennink, J.R. (1999) Mechanisms of Viral Interference With MHC Class I Antigen Processing and Presentation. Ann. Rev. Cell. Dev. Biol. 15: 579-606.

Search the WWW for more information on this topic

Search MEDLINE for the latest publications on this topic

© MicrobiologyBytes 2010.