The term Adenovirus refers to a family of viruses encompassing five genera including Mastadenovirus capable of infecting mammals. Originally found to present as a mild respiratory infection in humans, there are now seven known species capable of infection, associated with a wide range of clinical manifestations and the potentiality to be fatal in the immunocompromised. The lack of specific treatment targeting the virus teamed with the introduction of its use as a vector in gene therapy means research and interest into Adenovirus, its associations with the human immune system and the subsequent induction of the inflammatory response has gained momentum as of recent.
This report looks at the protein coded in the third open reading frame of the fourth early transcriptional unit (E4 Orf3) and its role in disrupting the innate transcriptional response the host initiates in the face of a potentially pathogenic infection.
Through transfection of Human Embryonic Kidney cells type 293 and HeLa cells, the association of NF-kB (a transcription factor) with Promyelocytic Leukaemia protein isoform II (PML-II), and the subsequent transcription of genes, such as the Interferon b promoter, were assessed, both in the presence and absence of the viral protein and under conditions mimicking an immune response.
The provisional findings were that E4 Orf3 does indeed disrupt the association of NF-kB with PML-II, at least in the absence of mock stimulation, and there is a subsequent downstream decrease on NF-kB transcribed genes both in the absence and presence of stimulation. The transcriptional decrease was found to higher in those cells with an activated inflammatory cascade. It was then postulated that the binding of NF-kB and E4 Orf3 to PML-II was competitive, giving interesting scope to future research and genetic manipulation, particularly in the direction of gene therapy where a subdued host response would be advantageous but also to therapy for the immunocompromised, where the converse effect would be desirable.
Contents page
1. Introduction
1.1 Invading Pathogens and Host Defences ………………………….………………………………… 3
1.2 Signal Transduction and the Transcriptional Response …………………………………..… 4
1.3 Adenoviridae …………………………………………………………………………………………..……..… 6
1.4 The Genome and E4 Orf3 …………………………………………………………………………………. 9
1.5 The Aims …………………………………………………………………………………………………….…… 12
2 Materials and Methodology ……………………………………………………………………………………..… 12
3 Results
Co-Immunoprecipitation ……………………………………………………………………………….……… 14
Luciferase Reporter Assay ……………………………………………………………………………..……… 16
4 Discussion
4.1 Co-Immunoprecipitation ………………………………………………………………………………… 18
4.2 Reporter assay …………………………………………………………………………………………..…… 20
4.3 Overarching statements …………………………………………………………………………….…… 22
5 Conclusive Statements ………………………………………………………………………………………….…… 23
6 Acknowledgements ……………………………………………………………………………………………..……. 24
7 References ………………………………………………………………………………………………………..………. 25
8 Appendices ………………………………………………………………………………………………………………… 27
1. INTRODUCTION
1.1 Invading Pathogens and Host Defences
In order to maintain overall organism health humans have evolved a complex immune system capable of recognising a variety of pathogens, ranging from parasitic worms to viruses. The immune system can be subdivided into two arms; innate and adaptive. The innate immune system consists of inherent hurdles an invading microbe must overcome in order to gain entry, survive and begin replication. It is specific only to the level of ‘non-self’ and consists of specialised barriers such as the epithelial surface, and mucin layer of the respiratory tract, occupational cells such as macrophages and dendritic cells detect and phagocytose foreign molecules in order to prevent establishment of infection alongside priming further immune defences, and basic antimicrobials such as defensins are produced by epithelial cells in response to a potentially pathogenic invasion. It also encompasses various intracellular changes (see section 1.2), requiring a conversion in cellular gene activation and expression. Conversely, the adaptive immune system is highly specific at the epitopic level but takes time to initiate, and requires induction via the innate immune response in order to proceed. It is composed of the humoral and cellular arms, with the first, humoral, mediated primarily through B cell activation and the secretion of antibodies. The cellular arm is conducted by T cell activation and the eradication of reservoirs of infection. Together the immune system provides protection against potential invasions, defence against current infection, and immunological memory for a vast range of intracellular and extracellular pathogens, key to host survival.
An important feature of the immune system is the ability of an immune cell to directly interpret whether the invading pathogen is virus, bacteria or fungi; or indeed intracellular or extracellular, due to the inherent, germline encoded, detection mechanism – pattern recognition receptors (PRRs). The detection is largely of pathogen associated molecular patterns (PAMPS) found on the extracellular surface of microbes, or as vital internal components, and their damage associated molecular patterns (DAMPS) making the likelihood of detection as great as possible. The detection is through specific PRRs in a specific location, for instance Rig-1 detects cytosolic DNA, thus implying an intracellular pathogen (likely a virus).
1.2 Signal Transduction and the Transcriptional Response
In order for a cell to alter its environment in defence against a potentially pathogenic invasion, gene expression needs to change from proteins required by the cell under normal conditions, with a high potentiality to be advantageous to an invading pathogen, to those which will help the cell defend itself, modulating the immune system in order to extend protection to a whole organism level. By definition this requires the transient shut down and activation of alternate genes under strict control, as to provide the most efficient and effective of responses, whilst mitigating the damage extended onto host tissue itself. An individual cell also has to overcome the apparent challenge of a limited number of transcription factors capable of potentiating highly differential responses, activating different genes at different times [3]. One such transcriptional response is the production of cytokines such as interferon. The interferon response can be initiated through the release of inflammatory stimuli such as Tumour Necrosis Factor , secreted by professional immune cells (such as natural killer cells) [4] in response to a potential pathogen detection. TNF binds it’s cognate receptor on the surface of cells and through the signal cascade simplified in figure 1, initiates interferon production that induces autocrine and paracrine effects.
Interferons are secreted proteins that bind to heterodimer receptors which triggers an intracellular signal transduction cascade. The result is the activation of the Janus Kinase family and signal transducers and activators of transcription, the JAK/STAT pathway, which elicits the antiviral state through oligomerisation of the interferon receptor heterodimer. This allows a phosphorylation cascade ultimately leading to the nuclear localisation of the STAT1 and STAT2 transcription factors and the subsequent binding of the interferon response element, granting the transcription of interferon stimulated genes [2].
There are several hundred upregulated genes specific to viral infection and the establishment of the antiviral response [2]. Of these many are thought to to act on viral processes directly such as viral genome transcription or translation (Protein Kinase R and 2’ 5’ OAS), restriction of viral replication (Mx) but some act in the prevention of the cell itself supporting the viral lifecycle (P21 – an inhibitor of the G1-S phase specific cyclin-dependant kinases).
A key part of the interferon response in regards to this context is the action of Promyelocytic Leukaemia Proteins (PML), which although have functional associations in a wide range of cellular activities, for the purpose of this report I will discuss isoform II which is implicated in the transcription of both IFN-b and the subsequent stimulated genes. PML-II has been shown to promote recruitment of transcription factors, such as NF-kB increasing efficiency of gene induction. It is therefore a logical phenomenon that a virus evolves measures to counteract such, and the fact that they do so shows the inherent importance of individual host mechanisms in order to defend and inhibit pathogenic disease [5].
1.3 Adenoviridae
The Adenoviridae family encompasses five genera, Mastadenovirus (containing the mammalian adenoviruses), Aviadenovirus, Atadenovirus, Siadenovirus, and Ichtadenovirus found across 7 host species with 57 serotypes, in humans adenoviruses (HAdV) are classified into species A-G based on phylogenetic distance (primarily DNA polymerase amino acid sequence), genome organisation (particularly within the early three transcription unit and the number of VA RNA genes), nucleotide composition (guanine and cytosine percentage), oncogenic ability in rodents, host range, cross-neutralization and their ability to recombine [6]. The viral structure is shown in figure 2.
Initially isolated in humans from patients suffering acute febrile illness, the presence of Adenovirus has since been associated with several other common disease states, including gastroenteritis (HAdV-F types 40, 41, HAdV-G type 52), hepatitis (HAdV-C), keratoconjunctivitis (HAdV-B and -D), meningoencephalitis (HAdV-A, -B, and -D), cystitis (HAdV-B), upper and lower respiratory tract infections (most commonly species HAdV-B and C and E presenting as pneumonia), and myocarditis. Interestingly there are additional associations with non-inflammatory conditions, for example obesity (HAdV-A type 31, HAdV-C type 5, HAdV-D types 9, 36, 37) [7]. Although there are virus species associated with specific infection sites in the host, (indicated) this is in no way definitive.
Adenoviruses possess a double stranded linear DNA genome contained in a non enveloped icosahedral shell embedded with fibre proteins, responsible for binding primarily to host Coxsackie Adenovirus Receptor (CAR) but occasionally CD46 and desmoglein 2 (DSG2), all expressed on multiple tissue including endothelial and epithelial cells. The use of common host receptors is partially responsible for the wider tissue tropism of these viruses seen in the diverse human pathogenesis [7]. After the initial association between the carboxyl terminus knob domain of the adenovirus fiber protein and the CAR, lower affinity interactions occur between integrins present on the host cell, and an arginine–glycine–aspartic acid (RGD) motif on the adenovirus penton base capsid. This initiates receptor mediated endocytosis in clatherin coated vesicles and subsequent host cell entry. Adenoviruses are believed to start uncoating before entering the cytoplasm by penetration from the vesicle in the early endosomal stage, mediated by the pH change and the penton base protein (thought to be around 10 minutes’ post internalization [8]). Partially uncoated virions are then trafficked to the nuclear pore via microtubular transport [7].
Adenovirus is a highly immunogenic virus and the induction of the immune response by adenoviruses is a phenomenon well studied in vitro and in vivo and the ability of immuno-competent individuals to be able to clear the infection and recover signifies its effectiveness. The fact that its infection usually only requires treatment in those that are immunocompromised, shows the importance of a functional immune system in the clearance of the pathogen. Therefore, any interactions the virus has in suppressing the immune system are interesting areas of therapeutics aimed research.
Adenovirus is also under study as a vector for gene therapy. This is an interesting alternative take on virus interaction with the cellular immune reactions as in this case it would be of interest for the virus to suppress the immune response, however full cellular countermeasures to such a phenomenon would need to be considered, but this adds an interesting take on such scientific endeavors. Figure 3 gives a simplistic summary of the viral lifecycle with a few key areas of viral detection by host innate receptors.
1.4 The Genome and E4 Orf3
Although Adenovirus genome size varies between 30–40 kb, its organisation remains consistent throughout the species. The genome is separated into three distinct regions defined by their transcription patterns in relation to the infectious cycle; early, intermediate and late [6]. There are five early open reading frames, E1a, E1b, E2, E3 and E4. Proteins in the E1 region are responsible for transcriptional regulation of both viral and host genome, including driving the host cell into the S phase of the cell cycle, advantageous as the cell can now replicate the viral DNA. E1a genes have also been demonstrated to inhibit the induction of genes usually activated by interferon. E2 proteins are responsible for allowing the genome to replicate via a protein priming mechanism involving a preterminal protein covalent attachment to the 5’ end of the genome, the DNA polymerase and a DNA binding protein required for the replication of the genome itself as well as the overall regulation. The E3 region has its role in subverting the host immune response, allowing viral survival through altering the function of host cell surface receptors, associated intracellular signalling events and the secretion of proinflammatory molecules such as cytokines and chemokines. The E4 transcriptional unit is activated by E1a and also involved in avoidance of the host immune response, therefore is of key interest in this report. Viruses mutated in the E4 region show little phenotypic change when mutations are made in individual open reading frames, indicating functional redundancy and thus the evolutionary importance of each of these proteins cohesively. In fully disrupted E4 virus, viral DNA replication, accumulation of late viral messages and proteins, virus particle assembly and shut-off of host protein synthesis are all disrupted causing severe attenuation [9].
For clarity it is important to mention that there are two delayed early (intermediate) open reading frames encoding protein IX, a minor capsid protein essential for packaging, and protein Iva2, an enhancer of the major late promoter. The major late unit is initiated after viral genome replication has begun, post transcriptional modification allows five proteins to be produced mainly for structural roles or involved in the morphogenesis of the virion [10]. Figure 4 shows the transcriptional map of the genome based on Adenovirus 5.
For the continuation of this report particular attention will be payed to the third open reading frame of the forth early region (E4 Orf3) and its subsequent protein. Many viral proteins are multifunctional due to the evolutionary drive to have streamlined genome for transcriptional efficiency and although E4 shows the most diversity between Adenovirus species, the E4 Orf3 protein has surprising conservation [9]. Its roles have been demonstrated to be necessary due to the inherent changes the virus causes to the cell environment in order to survive, replicate and attempt to inhibit the immune response, all of which can be detected by the cell, initiating retaliation [11]. Orf3 has functions that are complementary to Orf6 of the same transcriptional unit, such as inhibition of the Mre11-Rad50-Nbs1 DNA repair complex. However, many functions can be ascribed to Orf3 singlehandedly including differential splicing of the major late transcriptional unit and, in the context of this report, most importantly, the association with the Promyelocytic leukaemia protein nuclear bodies (specifically PML-II) causing rearrangement of the nuclear domains into track like structures [12] disrupting their aforementioned functions. Orf3 has been demonstrated to oligomerize and localise to the nucleus where it can associate with PML-II and its functions have claimed to be necessary for efficient viral DNA replication in the face of the antiviral response, initiated by interferon treatment [13]. Due to the observed effect of significantly inhibiting the transcription of IFN-b and its associated genes [14], it has been postulated that there is a competition for binding of PML-II by NF-kB and E4 Orf3.
1.5 The Aims
This research compliments existing research, investigating further whether E4 Orf3 disrupts the association of PML-II with NF-kB and whether it subsequently inhibits the transcriptional activation of genes requiring NF-kB as a transcription factor. This will be carried out, firstly though Co-Immunoprecipitation experiments looking at the presence of NF-kB through pull down via PML-II (via a FLAG tag and anti-FLAG elution) in the presence and absence of Orf3 and TNFa (simulating a mock immune stimulation) in order to determine their association, and disruption of. The second part of the research will consist of using an NF-kB linked luciferase reporter assay to determine the effects of these conditions on the transcription of NF-kB requiring genes. The expected result is that in the presence of stimulation, mimicking the inflammatory response, the amount of PML-II associated NF-kB would increase, which is hypothesized to subsequently decrease in the presence of E4 Orf3. This is expected to cause a subsequent downstream decrease in activity of the genes requiring NF-kB in the presence of the viral protein. In the presence of stimulation the genes associated are expected to increase, with the effect of addition of Orf3 reversing this increase through binding PML-II, rearrangement if its structure into track-like features, and prevention of the NF-kB associated gene transcription.
2. MATERIALS AND METHODOLOGY
Cells. Escherichia coli type DH5a used for transformation via addition of 1ml of plasmid DNA (Orf3 or PCI-neo empty vector with DRBCC FLAG-PML-II for the Co-Immunoprecipitation, or pNiFty-Luc and b-galactosidase for the reporter assay). Bacteria had competence induced through ice and heat treatment and were plated on Ampicillin+ lysogeny broth medium. Selected colonies were then grown overnight in liquid lysogeny broth (with 1:1000 ampicillin) and plasmid extraction carried out according to Quigen DNA Mini Kit protocol using the full overnight culture. Human Embryonic Kidney cells type 293 and HeLa cells were maintained in 10% Foetal Bovine Serum supplemented Dulbecco’s Modified Eagle Medium, cultured at 37oc in 5% CO2 and passaged every 4 days.
Co Immunoprecipitation; transfection, stimulation, harvest and assay. HEK 293 cells were grown to 90% confluence (approximately 2.5×106 cells in a 60mm culture dish) before transfection with 375ng FLAG-PML-II DRBCC and 1000ng either PCI-neo (control) or Orf3 in the presence of LT1 transfection reagent (1ml/mg DNA). 24 hours later, appropriate cells were stimulated either with 10ng/ml TNFa or by transfection with poly(I:C) (1ml/3.2×106 assumed cells at confluency, in the presence of 20ml Lipofectamine 2000) and incubated either 1 hour or 16 hours respectively. Cell harvest carried out using a high salt lysis buffer containing 0.1% protease and 0.1% phosphatase inhibitors, made using the protocol described by Hoppe et al 2006 [12] before sonication and addition of equal volume 50mM salt buffer. Samples of extract to represent the input to the Co-Immunoprecipitation were removed at this point, before the addition of agarose anti-FLAG beads (20ml/sample). Immunoprecipitation and elution were carried out according to the protocol, again described by Hoppe et al 2006 [12]. 2xSB buffer and dithiothreitol (DTT) were added to samples before protein electrophoresis in a 7.5% resolving gel during the detection of PML-II and NF-kB pull down and 15% resolving gel in the detection of E4Orf3 all at 100V for 1.5 hours. Membrane transfer occurred at 350 milliamps for 1.5 hours before soaking in 2.5% blocking buffer consisting of milk powder diluted in 0.05% PBS-tween. Antibodies used to detect FLAG and Orf3 were monoclonal mouse (M2) antibodies and NF-kB, rabbit monoclonal antibodies. Therefore, the secondary antibody was, for FLAG and Orf3, anti-mouse (rabbit origin) conjugated horseradish peroxidase (HRP) polyclonal antibody or, for NF-kB, goat anti-rabbit conjugated HRP. The addition of substrate was according to Thermo-Fischer SuperSignal Chemiluminescent protocol, before exposure to film.
Reporter assay; transfection, stimulation, harvest and assay. HeLa and HEK 293 cells were plated in a 96 well plate at a concentration of approximately 105 cells per well. 36 hours later they were transfected with 225ng pNiFty-Luc and 25ng b-galactosidase with 250ng of either PCI-neo or Orf3 in HEK 293, HeLa cells were found to require transfection of 100ng of b-galactosidase. Stimulation of the appropriate cells took place 24 hours later using 10ng/ml TNFa and a 1 hour incubation. Cells were harvested using 1x passive lysis buffer, orbital shaking for 1 hour at room temperature and centrifugation to remove cell debris. The b-galactosidase assay involved assay reagent Z buffer (1.16% Na2HPO4 (60mM); 0.55% Na2H2PO4 (40mM); 0.075% KCl2 (1mM); 0.024% MgSo4 (1mM)), O-nitrophenyl-b-D-galactopyranoside (ONPG) 4mg/ml diluted in Z buffer and b-mercaptoethanol before incubation at 37oc (~12 minutes, time varies – indicated by colour change) and addition of 80ml 1M Na2CO3 and reading at 405nm by a spectrophotometer. The luciferase assay included the addition of BrightGlo reagent before reading using luminometer and Ascent software.
3. RESULTS
The presence of Orf3 reduces the association of NF-𝜅B with PML-II, at least in the absence of stimulation. In order to observe the intracellular association of NF-kB with PML-II under normal and immuno-stimulatory conditions, and in the absence and presence of adenovirus protein E4 Orf3, a Co-Immunoprecipitation experiment was carried out. HEK 293 cells were transfected with plasmids containing PML-II that had been FLAG tagged along with the deletion of the RBCC motif (preventing association of this PML isoform with others present within the cell, potentially causing unspecific pull down and subsequent detection [15, 16]) alongside either E4 Orf3 or an empty vector (PCI-neo) – to account for changes in the cell in response to transfection of foreign DNA. This was carried out in duplicate so that one set of each transfection type could be stimulated by TNFa, a mimic of the inflammatory response that would occur in vivo, or poly(I:C) a synthetic, double stranded RNA analogue mimicking pathogen detection through Toll Like Receptor 3 [17]. The results of these Co-Immunoprecipitation experiments are displayed in figure 5.
As the data suggests, the presence of NF-kB through association with PML-II increases upon stimulation, a result that is consistent with previous literature [5] and reasoning. In the presence of E4 Orf3, without stimulation, there is visibly less NF-kB present in association with PML-II. However, in what may be a slightly different to the proposed model, in the presence of stimulation, the impact of Orf3 is a slight increase of PML-II associated NF-kB. Within this however bears the consideration that the amount of NF-kB within the input for the cells that contain E4 Orf3 and have been stimulated (lane 4, figure 5A ) is much higher than that in the comparative lane containing PCI-neo (lane 2, figure 5A), therefore preliminarily suggesting there is no increase nor decrease in PML-II:NF-kB in the presence of E4 Orf3 stimulation. This would still contradict expectations however, due to the idea that E4 Orf3 plays a role in viral lifecycle when in the face of the immune system of the host. For this reason, a tentative explanation may lie with the idea of competitive binding of PML-II by NF-kB and E4 Orf3, allowing the hypothesis that when there is already a high concentration of NF-kB in the cell, associated with PML-II, Orf3 may fail to compete and elicit any consequential impact. Possible support for this theory comes from the fact that it has previously been reported that PML-II:NF-kB association only happens in cells that have established a stress response [5], suggesting the cells used in this experiment were already stressed prior to stimulation, thus upon the addition of TNFa the cell had potentially established too high a response for the amount of E4 Orf3 we transfected to overwhelm and control. It is unknown whether this a phenomenon known to occur upon transfection of the whole virus in cell culture or in vivo, presumably not.
Figures 4C and 4D show an apparent lack of FLAG-PML-II in the samples, however due to the apparent pull down of NF-kB and the fact there was a timing delay between the Co-Immunoprecipitation of NF-kB and FLAG-PML-II it is thought to be due to denaturation and therefore experimental error. Although, this would need to be corrected in order to allow these results to stand. The apparent lack of E4 Orf3 (figure 4E) is postulated to be due to the small size and lack of ability to easily be recoverable from cells subjected to the standard protocol lysis.
The collaboration of these results suggests that E4 Orf3 does indeed have the potential to be able to disrupt PML-II:NF-kB association, however due to the observation it is an action that can apparently be overwhelmed, it may be suggested that this binding is competitive, therefore obliges to associated kinetics.
E4 Orf3 can reverse the increase of transcription on NF-kB requiring genes associated with immunostimulation. HeLa and HEK 293 cells were transfected with b-galactosidase, to account for transformation efficiency of the plasmid into the cells (of which was normalised against cells that had no DNA transfection), and p-NiFty-Luc plasmid of which contains five NF-κB repeated transcription factor binding sites (TFBS) and a Luc (Luciferase) reporter gene of which expression will be representative of any potential activity of other NF-kB stimulated genes (such as IFN-b). This allows the transcriptional response to TNFa and the presence of E4 Orf3 (comparative to PCI-neo) to be compared. The results are visualised in figure 6.
In both cell lines the presence of stimulation, (both E4 Orf3 and PCI-neo transfected) increases the activity of luciferase conjugated, therefore NF-kB instigated, gene activity. This is to be expected, and acts as a control to show correct treatment of the cells. It must however be considered that error bars are absent from PCI-neo data in the HeLa cell line. This is due to lack of useable data, allowing only corrected b-galactosidase, and therefore luciferase values, for one of the three replicates in each of these experimental parts. The reliability of these results therefore is questionable, and hence the experiment was repeated using the HEK 293 cell line due to its inherent higher transfection efficiency.
The defining result of this experiment is the decrease of the activity in the presence of E4 Orf3, occurs in both the stimulated and non stimulated samples of both cell lines. However, although marginal, the impact is slightly larger in stimulated cells, this would support the idea that E4 Orf3 disrupts the association of NF-kB with PML-II in the face of an established antiviral response, thus decreasing the transcriptional activation of subsequent genes.
4. DISCUSSION
4.1 Co-Immunoprecipitation
Interpretation
When analysing the results, it is important to retain that the pull down of NF-kB must be associated with PML-II due to the basis of the experiment using Anti-FLAG elution. Thus the results must be viewed in caution on the basis that the presence of PML-II can neither confirm nor dispute the patterns of NF-kB seen due to its own absence. However, for the purpose of this discussion I will assume that it is correct, and the reason for the lack of FLAG-PML-II is due to the time delay between testing the samples and the subsequent denaturation of the samples leading to an inability to detect the PML-II (assumed) present in the samples.
The absence of E4 Orf3 is provisionally due to the fact the protein is small in size and potentially also in volume, as well as difficult to extract from the cell under standard lysis procedure, such as the one undertaken for this report. Therefore, it would have been idyllic to try to detect the presence of E4 Orf3 to further validate the results in order to confirm its presence to be able to apply the transcriptional changes to its function and to confirm that transfection was indeed successful. This could have been carried out through duplication of the transfection of cells carried out, with cell lysis using SDS, known to be able to cause cell lysis to such a degree that would allow detection.
Another observation in the data is that even in the unstimulated cells, of which would otherwise be assumed to have only background NF-kB, of which should not be in association with PML-II [5] therefore due to the presence of this, it can be concluded with some certainty that the cell stress response must have been activated prior to additional stimulation, perhaps due to over-confluence prior to transfection, the amount of DNA added, or other external factors such as remnant plasmid DNA. If this were the case NF-kB is a known regulator of such a response (18), therefore it would be logical that the increase in this transcription factor, and the fact it was present (potentially associated with PML-II) prior even to the addition of E4 Orf3 DNA could add biological complexity to the E4 Orf3 behavioral observations.
Conclusions
From the distinguished Co-Immunoprecipitation results, it can be suggested that E4 Orf3 has the potential to disrupt association of PML-II with NF-kB, which would provide evolutionary advantages to the virus in order to prevent amplification of the preexistent inflammatory response. This is following the hypothetical arms race of the pathogen vs host immune system in that the host evolves new means of detecting the pathogen in order to initiate a response, thus this protein would allow the virus a fall back in case this, the detection of pathogen and elicitation of response, does occur. It is shown through phylogenetic studies it is advantageous for pathogens to evolve multiple means of preventing a process that would have such a detriment to them carrying out their lifecycle such as the initiation of the inflammatory response.
The pattern observed, in that stimulation of the cell, thus an inherent increase in intranuclear NF-kB by definition, seems to reduce the efficiency of Orf3 to bind PML-II, suggestive of dynamics associated with direct competitive inhibition. It can therefore be postulated that the binding site of PML-II is the same, or at least close enough for direct interaction, for NF-kB as it is for E4 Orf3.
Future
In order to determine the true interplay between NF-kB and E4 Orf3 further investigation into potential NF-kB binding sites could give insight into the characteristics of the association in physiological conditions. It is known that E4 Orf3 binds PML-II at residues 645–684 at the unique C terminus [15], it can therefore be idealized that NF-kB binds at a site within at least close proximity to this, allowing interference of binding. Further research into the NF-κB and PML-II associations may help in determining the full dynamics that occur in this phenomenon. For example, the transiency of the association between PML-II and NF-κB, or the potential modification and release of NF-κB by PML-II.
Further work could include a dose response curve, determining the amount of Orf3 required for full inhibition of the PML-II associated pull down of NF-kB. Upon collecting this data, it would important to place it in context of levels of E4 Orf3 achievable in vitro (and indeed in vivo) under viral replicative control, comparative to levels of PML-II and those of NF-kB initiated by the cell. Perhaps most importantly, the levels of these cellular proteins achievable in the time from initial adenovirus detection to the the point in which the virus could transcribe, translate and create a functional E4 Orf3 of sufficient volume, in order to fully appreciate the dynamics that may occur under viral infection of these cells in vivo. The importance of this would lie in the impact of prior NF-kB and PML-II association before functional E4 Orf3 can carry out its disruption. It was documented that the viral protein has action in the face of an already established inflammatory response [13] thus it wouldn’t be unreasonable to assume that E4 Orf3 may have a higher PML-II affinity than NF-kB, capable of displacement of the previous, interestingly this appears not to have been witnessed in this assay.
4.2 Luciferase Reporter Assay
Interpretation
Similarly, to the previous assay, it remains to be demonstrated that E4 Orf3 was present in a adequate cellular volumes in order to determine it is responsible for the reduction in NF-kB associated gene activity as demonstrated by the results. There is also visible activation of these genes in the absence of external stimulation (particularly within the HeLa cell line), this can be taken in cohesion with the previous results, perhaps also for the reasons previously listed.
Values for the luciferase reporter assay were corrected using a transformation of the lac Z gene of the lac operon, coding for the b-galactosidase enzyme, therefore transformation efficiency of the luciferase reporter itself is accounted for. However, whilst using HeLa cells experimentally some issues arose in calculating a corrected value for the b-galactosidase using the control cells (no transfection) with a putative explanation of low transcription efficiency. This however meant that no standard deviation could be calculated and the experiment would need to be repeated in order to be able to compare the two cell lines properly.
Considering the raw data (appendix III) the reproducibility across the replicates in the HEK 293 cell line is consistent, therefore this stream of data can be taken with some integrity.
Conclusions
Using average relative light units (RLU) emitted by unstimulated cell lines, it can be deduced that in HeLa cells, the presence of E4 Orf3 causes a 20.58% reduction of gene activity, with an 8.99% reduction in HEK 293 cells. In stimulated cells E4 Orf3 reduces gene activity by 49.13% in HeLa and in 18.48% HEK 293 (with a minimum reduction of 138.68RLU calculated using the standard deviation in HEK 293 cells). Obviously these reductions are marginal, thus it would be of interest to repeat the dose response curve suggested as a Co-Immunoprecipitation assay in this context too, in order to calculate the amount of E4 Orf3 necessary to reduce the gene activity significantly. Again, it would be interesting tot compare this to in the amount of E4 Orf3 achievable in an infective dose of Adenovirus. It is however promising that the decrease produced by E4 Orf3 is greater in stimulated than unstimulated cells, as this goes someway to prove the existing model that E4 Orf3 prevents transcriptional responses in light of an existing inflammatory response.
Future
In order to fully consider the results of this experiment it would need to be repeated, particularly within the HeLa cell line, in three replicates that can be used to calculate the average and standard deviation, making the assessment of how much impact E4 Orf3 portrays more accurate (again, the presence of which needs to be confirmed through SDS lysis), whilst also showing reproducibility of data. Without full corrected b-galactosidase results for the HeLa cell line a simple explanation could just be that the transfection off E4 Orf3 has just increased overall transfection efficiency in comparison to PCI-neo, although unlikely due to the similar patterns seen in HEK 293 cells. It is important to consider that since most viral proteins are multifunctional and E4 Orf3 can act in cohesion with E4 Orf6, the implications of additional, alternative viral proteins to E4 ORf3 transfected cells. It was recently suggested that Adenovirus 5 E1A–13S also interacts with PML-II, helping initiate viral genome transcription [19], thus not only adding a further consideration regarding PML-II associations with NF-kB and E4 Orf3 but also in the overall manipulation of the cell by the virus in order to achieve its life cycle, here reverting the normal function of PML-II totally. It would also be of overall interest to attempt a repeat of the experiment in both cell lines with a lower background NF-kB activation, perhaps even looking into the impact made of stimulating the cell lines prior to transfection with E4 Orf3, although no doubt transfection efficiency would get called into doubt here.
4.3 Overarching statements
The impact of research surrounding the action of E4 Orf3 on the transcriptional response to inflammatory stimuli, such as those initiated post detection of the virus itself, will further understanding of the evolution viruses undergo in order to subvert the immune response, revealing not only the arms race between viral pathogen and its ideal host, but potential lacking aspects of human host immunity of which may provide potential targets for therapy. The concept of improving therapies that aid in encouraging the human host response against a pathogen rather than direct antimicrobial effects is becoming ever more important as various resistances are increasing in prevalence and rate of appearance. Of particular interest is the resistance to cidofovir by Adenovirus 5 seen in both laboratory animals and humans [20, 21]. In addition is the concept that Adenovirus treatments are often mostly (or only) required in immunocompromised patients who can suffer from highly disseminated disease complicated by the lack of licensed systemic or topical treatment specific to the virus due largely to the large number of serotypes the ideal drug would be effective against (even just in one specific disease), the lack of distinct viral targets that previous drugs have been based on (such as the DNA polymerase, which although different uses the same nucleotide pool as the host) and other general drug problems such as their clinical trial efficacy [21]. It is for these reasons that research into treatment against adenovirus has begun to move into a more advanced territory, looking at specific viral and host interactions in the form of miRNA. The use of miRNA allows the specific targeting and knockdown of viral RNAs and the subsequent proteins, such as E4 Orf3, that potentially allow the virus to be so successful in the face of a weakened immune system [21].
The immune response to Adenovirus also takes an interesting alternate view in the exploration and use of adenovirus as a vector for gene therapy, in which over 400 trials are currently in place [6]. Adenovirus is an ideal viral vector due to its potentiality to carry a large amount of genetic material and the wide tissue tropism. Such a use requires extensive knowledge of the ‘wild type’ interactions within the host which thus trigger an immune response, in order to minimise such reaction in its therapeutic use [22]. Most adenovirus vectors are genetically modified versions of type 5, in which the E4 Orf3 protein discussed in this report originates. It has been idealized for some time that not all viral proteins can be removed in order for the vector to be efficient, potentially including ones that subvert the host response such as E4 Orf3 due to the fact this virus is highly immunogenic [23]. Conversely once more, Adenovirus can be used as a vector requiring a good host immune response, such as its use in vaccines, one particular use of the strong immunogenic response is when the vector is carrying tumour-associated antigens [24].
The fact that interest in the host response to adenovirus takes two such polar angles makes the outcome of this and surrounding research into Adenovirus and its intercommunication with the host, and the subsequent retaliation particularly exciting.
5. CONCLUSIVE STATEMENT
The inflammatory response is a vital part of the innate immunity, driving the chemotaxis of professional immune cells to the site of infection where they can eliminate reservoirs of infection and prime the adaptive immune response of which can provide immunological memory. It is therefore idyllic of a virus to be able to modulate and subvert this host response in order to keep its host cell alive so that it can carry out its own lifecycle. Adenovirus has evolved a number of proteins for this role, of particular interest in this report was the third open reading frame of the fourth early transcriptional unit, coding the protein E4 Orf3. Alongside other functions, E4 Orf3 was suspected to have roles in the interference of the transcriptional response to inflammatory stimuli, such as released upon detection the presence of the very Adenovirus encapsulating the protein. Prior to this report it was known that E4 Orf3 bound PML-II and reorganised the nuclear bodies into track like structures. It was also shown to significantly inhibit IFN- expression. In this report, putative results are suggestive that E4 Orf3 disrupts association of PML-II with NF-kB (at least in the absence of mock immune stimulation but expected in cells eliciting an inflammatory response) and subsequently reduces the activity of genes requiring NF-kB as a transcription factor. This was shown through a series of Co-Immunoprecipitation and Western blot experiments alongside Luciferase Reporter Assays. The implication of such work lies in the expanding field of knowledge of the interaction between Adenovirus species and the host with polar interests lying in its encouragement of immune subversion for Adenovirus vector progression, and the converse for treatment of disease within immunocompromised individuals.