Cardiotonic Steroids Suppress Adenovirus Replication
by
Filomena S Grosso
A thesis submitted in conformity with the requirements for the degree of Master of Science
Department of Laboratory Medicine and Pathobiology University of Toronto
© Copyright by Filomena S Grosso 2018
ii
Cardiotonic steroids suppress adenovirus replication
Filomena S Grosso
Master of Science
Laboratory Medicine and Pathobiology
University of Toronto
2018
Abstract
Human adenoviruses are common pathogens that can cause life-threatening illness, yet no
approved therapies are available for treatment. This work shows that two cardiotonic steroids,
digoxin and digitoxin, inhibit adenovirus replication in A549 lung carcinoma cells beyond
immediate early E1A expression, with reduced expression of subsequent early regions E1B, E2B
and E4. This alteration in viral gene expression leads to a block in genome replication that in turn
prevents late gene expression. Nuclear changes characteristic of early infection were observed in
treated, infected cells. E1A expression can be reduced if cells are pre-treated with drug provided
that the drug is maintained on the cells after infection. The antiviral effect is abrogated by
increased extracellular concentration of potassium. Digoxin and digitoxin also inhibit adenovirus
replication in primary human nasal epithelial cells. This work supports the idea that cardiotonic
steroids could be developed as antiviral agents for adenovirus infections.
iii
Acknowledgments
First and foremost, I would like to thank my graduate supervisors, Drs. Martha Brown and Alan
Cochrane for their continued guidance, support and teaching throughout my graduate studies.
Thank you for encouraging me to pursue a graduate degree and helping me to realize my full
potential as a scientist. Similarly, I would like to thank past and present members of both
laboratories for their assistance, helpful conversations and words of encouragement, especially
Casandra Mangroo, who trained me as an undergrad and nurtured my interest in research. I
would like to also thank the Laboratory Medicine and Pathobiology Graduate Department,
namely Dr. Harry Elsholtz, who encouraged me to pursue this degree and believed in my
abilities. Lastly, I would like to thank my committee members, Dr. Lori Frappier and Dr. Theo
Moraes for their assistance and advice throughout my Master’s degree.
I would like to acknowledge the individuals who graciously gifted reagents to make this work
possible: Dr. Gary Ketner at the Johns Hopkins Bloomberg School of Public Health, Baltimore
for his gift of E4orf6 antibody, Dr. Arnie Levine, PMV Pharma for his gifts of the E1B-55K, and
E2A-72K antibodies, Dr. Thomas Dobner at the Heinrich-Pette-Institut for the E4orf3 antibody
and Dr. Lucy Osbourne at the University of Toronto for TBP primer sets. A big thank you goes
to Dr. Theo Moraes and his technician Hong Ouyang for growing and graciously providing us
with human nasal epithelial cells. A sincere thank you goes to the staff at the University of
Toronto Microscopy Imaging Lab for teaching me how to process and image samples using
transmission electron microscopy.
I would like to acknowledge CIHR for funding the works described in this thesis.
Lastly, none of this would be possible without my family, significant other and friends who have
been nothing but patient, supportive, encouraging and calming. To my parents, I dedicate this
work to you. Although you may not understand its contents, do know it is a representation of the
work ethic, determination and passion you have instilled in me. Everything I do is a result of
your sacrifice and hard work and this thesis is a part of returning the favour.
iv
Table of Contents
Acknowledgments.......................................................................................................................... iii
Table of Contents ........................................................................................................................... iv
List of Tables ................................................................................................................................ vii
List of Figures .............................................................................................................................. viii
List of Appendices ...........................................................................................................................x
Abbreviations ................................................................................................................................. xi
Introduction .................................................................................................................................1
1.1 Human adenoviruses ............................................................................................................1
1.1.1 Clinical implications of adenovirus infections and current therapies ......................1
1.1.2 Virion structure, entry and genome delivery ...........................................................2
1.1.3 Regulation of adenovirus gene expression ..............................................................5
1.1.3.1 Control of early promoters during the early phase of replication ............................5
1.1.4 Adenovirus early gene expression .........................................................................10
1.1.5 Adenovirus genome replication .............................................................................16
1.1.6 Adenovirus late gene expression ...........................................................................16
1.1.7 Assembly, packaging and release ..........................................................................17
1.2 Cardiotonic steroids ...........................................................................................................18
1.2.1 New potential uses for cardiotonic steroids ...........................................................20
1.2.2 Effects of cardiotonic steroids on the cell ..............................................................21
1.3 Research objective and rationale .......................................................................................22
Materials and Methods ..............................................................................................................24
2.1 Drugs ..................................................................................................................................24
2.2 Viruses and cells ................................................................................................................24
2.2.1 Viruses ...................................................................................................................24
2.2.2 Cells .......................................................................................................................24
v
2.3 Cell viability assays ...........................................................................................................25
2.4 Virus propagation and assay ..............................................................................................25
2.5 Effects of drugs on virus yield ...........................................................................................26
2.6 Infection of primary human nasal epithelial cells ..............................................................26
2.7 Collection of RNA and protein samples ............................................................................26
2.8 Immunofluorescence staining ............................................................................................27
2.8.1 Staining of A549 cells ............................................................................................27
2.8.2 Staining of primary human nasal epithelial cells ...................................................27
2.9 Western blot analysis .........................................................................................................28
2.10 Adenoviral DNA and RNA analysis .................................................................................29
2.10.1 RNA analysis .........................................................................................................29
2.10.2 DNA analysis .........................................................................................................31
2.11 Transmission electron microscopy ....................................................................................31
2.12 Statistical Analysis ............................................................................................................32
Results .......................................................................................................................................33
3.1 Treatment with digoxin and digitoxin reduces viral yield .................................................33
3.2 Digoxin and digitoxin inhibit adenovirus replication prior to viral genome replication ...33
3.3 Digoxin and digitoxin affect early gene expression after E1A expression ........................39
3.4 Digoxin and digitoxin induce nuclear changes in treated cells..........................................48
3.5 Time of addition of digoxin and digitoxin affects their efficacy .......................................48
3.6 Potassium ions can counter the antiviral effects of digoxin and digitoxin ........................54
3.7 Human nasal epithelial cells can be used as a model for adenovirus replication and
assaying drug effects ..........................................................................................................54
Discussion .................................................................................................................................61
4.1 Evaluating the effect of digoxin and digitoxin on adenovirus replication .........................61
4.2 Determining the importance of K+ in the antiviral effects of digoxin and digitoxin .........65
vi
4.3 Assessing nuclear changes in response to drug treatment .................................................66
4.4 Using hNEC as a model for adenovirus infection ............................................................67
4.5 Cardiotonic steroids as a pan-antiviral ...............................................................................69
Future directions........................................................................................................................71
Conclusions ...............................................................................................................................72
References ......................................................................................................................................73
Appendices……………………………………………………………………………...........90
vii
List of Tables
Table 2.1 Primers used for RT-qPCR experiments……………………………………………...30
Table 3.1 Assessing nuclear changes with transmission electron microscopy…………………..50
viii
List of Figures
Figure 1.1 Model of adenovirus structure …………..………………………………………...3
Figure 1.2 The adenovirus replication cycle………………………………………………......4
Figure 1.3 Adenovirus transcription map ………………..…………………………………...9
Figure 1.4 E1A proteins and their interactions with host proteins ……………...………......12
Figure 1.5 Cardiotonic steroids and their effects on the cell………………………………...19
Figure 3.1 Digoxin and digitoxin suppress replication of multiple adenovirus
species………………………………………………………………………………………..34
Figure 3.2 Digoxin and digitoxin have minimal cytotoxic effects on A549 cells…………...35
Figure 3.3 Effect of digoxin and digitoxin on expression of hexon protein…………………36
Figure 3.4 Effect of drug treatment on expression of E1A protein……...…………………..37
Figure 3.5 Digoxin and digitoxin block adenovirus genome replication…………………….38
Figure 3.6 Digoxin and digitoxin reduce E1B-55K protein expression……………………..41
Figure 3.7 The replication protein E2A-72K is decreased and not localized to replication
centers after drug treatment………………………………………………………………….42
Figure 3.8 Effect of digoxin and digitoxin on adenoviral E4orf6 protein
expression…………………………………………………………………………………....43
Figure 3.9 Effect of digoxin and digitoxin on adenoviral E4orf3 protein expression……….44
Figure 3.10 Positioning of primer sets used for RT-qPCR………………………………......45
Figure 3.11 E1B and E4 mRNA levels are decreased after digoxin and digitoxin
treatment……………………………………………………………………………………..46
Figure 3.12 Digoxin and digitoxin reduce E2B mRNA levels……………..………………..47
ix
Figure 3.13 Infected cells display abnormal nuclear appearance after digoxin treatment…...50
Figure 3.14 Digoxin and digitoxin change the localization of splicing factor Tra2β………..51
Figure 3.15 Effect of pre-treatment of digoxin and digitoxin on adenovirus E1A protein
expression……………………………………………………………………………………52
Figure 3.16 Effect of time of addition of digoxin and digitoxin on hexon protein
expression……………………………………………………………………………………53
Figure 3.17 KCl addition to media containing digoxin, but not digitoxin, rescues hexon
protein expression in a dose-dependent manner……………………………………………..56
Figure 3.18 KCl addition to media containing digoxin rescues viral yields, depending on drug
concentration…………………………………………………………………………………57
Figure 3.19 Primary human nasal epithelial cells are susceptible to human adenovirus
infection……………………………………………………………………………………...58
Figure 3.20 Digoxin is effective in primary human nasal epithelial cells…………………...59
Figure 3.21 Adenovirus infection kinetics in human nasal epithelial cells………………….60
x
List of Appendices
Appendix 1.1 Initial screening identifies digoxin and digitoxin as potential adenovirus
inhibitors…………………………………………………………………………………………91
Appendix 1.2 Time course experiments for early mRNA expression…………………………...92
Appendix 1.3 Representation of uninfected and adenovirus-infected nuclei visualized with
transmission electron microscopy………………………………………………………………..93
xi
Abbreviations
2HX-2 hydridoma cells producing anti-hexon antibody
A549 human lung carcinoma cell line
Adpol adenovirus polymerase
ALI air-liquid interface
ATCC American Type Culture Collection
ATF activating transcription factor
ATP adenosine triphosphate
BAK BCL-2 antagonist/killer
BAX BCL-2-like protein 4
BCL-2 B-cell lymphoma 2
BSA bovine serum albumin
CAR Coxsackie B adenovirus receptor
CBP CREB-binding protein
CHIV chikungunya virus
co-IP co-immunoprecipitation assay
CPE cytopathic effect
CR E1A conserved region
DAPI 4',6-diamidino-2-phenylindole
DBP DNA-binding protein
xii
dCMP deoxycytidine monophosphate
ddH2O deionized, distilled water
DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
dNTP deoxyribonucleotide triphosphate
dpi days post-infection
dsDNA double-stranded DNA
ECL enhanced chemiluminescence
EDTA ethylenediaminetetraaceticacid
EGTA ethylene-bis(oxyethylenenitrilo)tetraacetic acid
ER estrogen receptor
FCS fetal calf serum
GAPDH glyceraldehyde 3-phosphate dehydrogenase
HAdV human adenovirus
hCMV human cytomegalovirus
HEK 293 human epithelial kidney cell line
HIF-1α hypoxia inducible factor 1-alpha
HIV human immunodeficiency virus
hNEC human nasal epithelial cells
hpi hours post-infection
xiii
HRP horseradish peroxidase
HSV herpes simplex virus
IF immunofluorescence
IU infectious units
KCl potassium chloride
kD kilodalton
MAdV-1 mouse adenovirus type 1
MCL-1 myeloid leukemia cell differentiation protein
MEM minimal essential medium
MHC major histocompatibility complex
MLP major late promoter
MLTU major late transcriptional unit
mM millimolar
MMLV Moloney murine leukemia virus
MOI multiplicity of infection
MTOC microtubule organizing center
NCX Na+/Ca2+ exchanger
NF1 nuclear factor 1
NFκB nuclear factor κB
NKA Na+/K+ ATPase
xiv
nM nanomolar
OD optical density
PBS phosphate-buffered saline
PCR polymerase chain reaction
PFA paraformaldehyde
pi post-infection
PML progressive multifocal leukoencephalopathy
PP2A protein phosphatase 2A
pRB retinoblastoma protein
pTP preterminal protein
PVDF polyvinylidene difluoride
qPCR quantitative polymerase chain reaction
RIPA radioimmunoprecipitation assay
RNA ribonucleic acid
RSV respiratory syncytial virus
RT-PCR reverse transcription polymerase chain reaction
SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis
SD standard deviation
SEM standard error of the mean
SR serine-arginine rich
xv
TBP TATA-binding protein
TBS Tris-buffered saline
TEM transmission electron microscopy
TNFα tumour necrosis factor alpha
TP terminal protein
VA RNAs virus-associated RNAs
VEGF vascular endothelial growth factor
1
Introduction
1.1 Human adenoviruses
Human adenoviruses (HAdV) are non-enveloped, double-stranded DNA (dsDNA) viruses
belonging to the Mastadenovirus genus. This genus is organized into seven species (A-G) to
which each human adenovirus type is assigned based on sequence similarity, serology,
transformation of cultured primary cells and oncogenicity in rodents (Wold and Ison, 2013). To
date, there are close to 80 different types of adenovirus identified (Yoshitomi et al, 2017).
1.1.1 Clinical implications of adenovirus infections and current therapies
Adenovirus infections are common throughout the human population, as the seroprevalence in
North America can be as high as 70% and up to 90% in sub-Saharan Africa (Barouch et al,
2011). Respiratory infections with adenovirus are usually sub-clinical or show with mild
symptoms and usually clear without complications. Even so, according to the Respiratory Virus
Report obtained from the Public Health Agency of Canada, there can be as many as 100 cases of
adenovirus respiratory disease weekly. Serious infections can occur, often in outbreaks and
especially among military recruits (Potter, 2012). In the immunocompromised population,
infection can become disseminated, and fatal, in 50% of disseminated cases (Lion, 2014;
Echavarrìa, 2008). The risk of complicated adenovirus infections in the immunocompromised
population is highest with those receiving stem cell transplantation, children, patients receiving
grafts depleted of T-cells and patients experiencing graft-vs-host disease (Wold and Ison, 2013).
Limited treatment options are available for these patients, as there is no specific treatment for
adenovirus infections.
To date, several drugs have been used in severe adenovirus infection cases. Studies addressing
the efficacy of these drugs are largely inconclusive, as there is variability on a case-to-case basis.
Some of these drugs also have been shown to be toxic, sometimes causing more harm than good.
One such drug that has been used in severe cases of adenovirus infection is cidofovir. Cidofovir
is a nucleotide analogue which inhibits the adenovirus polymerase during replication and has
been seen to reduce adenovirus replication in lung carcinoma A549 cells and a rabbit model of
2
ocular disease (de Oliveira, 1996). However, the use of cidofovir in patients has been associated
with nephrotoxicity (Izzedine et al, 2005), which poses challenges to many transplant patients
already receiving other medications including antibiotics, antifungals and medications to prevent
graft-versus-host disease (Ramsay et al, 2017). The use of cidofovir has also been associated
with kidney injury in children and nephrotoxicity was associated with high mortality (Vora et al,
2017). Brincidofovir, also known as CMX 001, is a derivative of cidofovir with increased
bioavailability due to the addition of a lipid chain to the drug and shows less accumulation in the
kidney compared to cidofovir (Florescu, 2012). It has shown promise in animal models and is in
Phase III clinical trials. Other drugs such as ganciclovir and its derivative valganciclovir have
shown efficacy against adenovirus replication in A549 cells and animal models (Toth, 2015) and
have been used in clinical trials against keratoconjunctivitis, suggesting faster and improved
response in those treated with ganciclovir, though not statistically significant (Yabiku et al,
2011; Clinical Trails.gov: NCT01349452).The limited range of therapies to treat serious
adenovirus infections calls for alternative strategies to reduce virus replication.
1.1.2 Virion structure, entry and genome delivery
Adenovirus is non-enveloped with an icosahedral capsid made up of 240 hexon capsomers, with
a penton base at each of 12 vertices. A trimeric fiber protrudes from each vertex (Figure 1.1).
Adenovirus infection begins with high-affinity attachment of the fiber knob to its receptor on the
cell surface (Figure 1.2). For HAdV-C5 and other adenovirus types belonging to species A, C, E
and F, the receptor is the Coxsackie adenovirus receptor (CAR), a tight junction protein
(Roelvink, 1998). CD46 and sialic acid have been shown to be receptors for several species B
and D adenoviruses, respectively, as discussed by Zhang and Bergelson (2005). Subsequent
interaction of the RGD motif on the penton base with integrins on the host cell surface mediates
endocytosis (Wickham et al, 1993) but, prior to endocytosis, fibres and possibly penton bases are
released at the cell surface as a result of tension between fiber attached to CAR, which can drift
within the plasma membrane (Burckhardt et al, 2011), and penton base attached to integrins,
which are immobile. Removal of fibre exposes some of the internal protein VI which inserts into
the plasma membrane, creating small lesions and signaling a repair process that contributes to
endocytosis of the virion into an endosome with a leaky membrane (Luisoni et al, 2015). Within
the endosome, more protein VI is exposed and inserts into the endosomal membrane causing it to
rupture, releasing the destabilized virion into the cytoplasm (Wiethoff et al, 2015). Once the
3
virus has escaped from the endosome, it uses the microtubule network (Suomalainen et al,1999)
and dynein
Figure 1.1 Model of adenovirus structure. The most abundant capsid protein is hexon. At each
vertex of the adenovirus capsid is the penton base; extending from the penton base is the fiber.
Inside the virus capsid is the viral genome, which has a terminal protein (TP) at each 5’ end of its
double-stranded DNA genome. Taken from: Reddy, V. S., & Nemerow, G. R. (2014). Structures
and organization of adenovirus cement proteins provide insights into the role of capsid
maturation in virus entry and infection. Proceedings of the National Academy of
Sciences, 111(32), 11715-11720.
4
Figure 1.2 The adenovirus replication cycle. The adenovirus replication cycle begins with entry
into the host cell (1-2), endosomal escape (3) followed by trafficking of the virus to the nuclear
pore (4). Once docked at the nuclear pore, the virus genome is released into the nucleus (5) and
the early gene expression program begins (6). E1A gene expression promotes the expression of
the other early genes (7), followed by viral genome replication (8). Some of these genomes are
destined to be packaged as progeny virions. As viral genome replication commences, an
intermediate phase of replication begins (9). Proteins expressed from this phase stimulate the late
phase of replication through the major late promoter. (10) The late phase of replication results in
the expression of viral capsid proteins, which are transported back into the nucleus for viral
assembly and genome packaging (11-15). Reprinted with permission from Berk, A.J. (2013).
Adenoviridae. In D. M. Knipe, & P. M. Howley (Eds.), Fields virology (6th ed., pp. 1704-1731).
Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins Health
5
molecular motors (Bremner et al, 2009) to traffic to the nuclear pore. The viral capsid is too large
to enter the nuclear pore; it docks at the nuclear pore, where the viral genome is delivered into
the nucleus and the viral capsid disintegrates. Once the viral genome is delivered, the entry phase
of the replication cycle has been completed and the next step is to begin the early phase of viral
gene expression.
1.1.3 Regulation of adenovirus gene expression
Most studies regarding adenovirus replication have been done in the context of species C
adenoviruses, namely types 2 and 5. In general, the genome organization is conserved among
human adenoviruses, though there may be differences in expression between adenoviruses of
different species. The human adenovirus genome (Figure 1.3) consists of early (E1A, E1B, E2,
E3 and E4), intermediate (IX, IVa2, L4 intermediate, and E2 late) and late genes (under control
of the MLP, the major late promoter) that are temporally regulated to successfully replicate the
genome, as well as assemble and package new particles for spread to neighbouring cells. It
should be noted that the adenovirus genome is transcribed by host RNA polymerase II, except in
the case of the virus-associated RNAs (VA RNA) that are transcribed by host RNA polymerase
III (Weinmann et al, 1974). Adenovirus gene expression is controlled at the level of promoter
activation as well as transcriptionally and post-transcriptionally at the level of splicing and polyA
site usage, respectively.
1.1.3.1 Control of early promoters during the early phase of replication
To begin the early phase of replication, activation of the E1A promoter is required. As discussed
in Schreiner et al (2012), a host protein, Daxx, represses the viral genome. This transcriptional
repression is negated by the viral protein, pVI, through its PPxY motif. The adenovirus E1A
proteins (243R and 289R) are important regulators of early gene expression. The E1A proteins
are the first proteins to be expressed after genome delivery to the nucleus and are responsible for
increasing the transcription of the E1A promoter itself and the promoters of the other early
regions E1B, E2, E3 and E4 (Berk, 2013). The large E1A protein transactivates its own and other
early viral promoters indirectly, by recruiting and binding host factors. The large E1A protein
(289R) contains a unique region that binds host factors such as Med23, a part of the host
Mediator complex, TATA-binding protein (TBP) and p300 that contribute to the strong
6
activation of early promoters. The E1B, E2 and E3 promoters all have TATA-boxes, which
contribute to the control of these promoters through interactions of the large E1A protein and
host TBP. The early promoters also contain binding sites for host transcription factors: E1B and
the E2 late promoter have binding sites for Sp1; E3 and E4 promoters have binding sites for
activating transcription factors (ATF). Unique to the E2 and E4 regions are transcription factors
E2F and E4F, respectively. The importance of E2F will be discussed later. Overall, adenovirus
E1A proteins and host transcription factors are important for regulating early gene expression at
the transcriptional level.
1.1.3.2 Splicing and polyA site usage during the early phase of replication
RNA splicing is an important strategy that many viruses use to alleviate genome size restrictions
and be able to express all the proteins necessary for viral infection. All human adenovirus
mRNAs are spliced, with exception of the pIX transcription unit (Zhao et al, 2014). Over the
course of adenovirus replication, splice site usage changes; early and late phase splicing patterns
are different. For example, in the case of E1A, three major RNAs are made: 13S, 12S and 9S.
The 13S and 12S transcripts are abundant early, while 9S is abundant at later times (Spector et
al, 1978). These transcripts share 5’ and 3’ ends and have different-sized introns spliced out. A
similar comparison can be made with the E2B RNAs (22S and 13S), where the 13S mRNA is
more abundant during the late phase of infection due to alternative splicing of the E2B transcript
(Montell et al, 1984). The E2 region is controlled by two promoters, early and late, and expresses
two sets of transcripts: E2A and E2B. E2A produces an mRNA coding for the 72kD DNA-
binding protein (DBP), though more than one mRNA is made. The E2B set of transcripts code
for two other replication proteins, precursor terminal protein (pTP) and the adenovirus DNA
polymerase (Adpol) and differ in splicing of their introns (Chow et al, 1979). Transcription of
the E3 region produces one primary transcript that is processed by alternative splicing to create
either one of two RNAs, E3A or E3B. The E3A or E3B transcripts share a common 5’ end but
differ in polyA site usage, as will be discussed later (Zhao et al, 2014). The E4 transcription unit
expresses one primary transcript that is processed by splicing to generate over 20 transcripts all
sharing a common 5’ and 3’ end (Wold et al, 1995).
Polyadenylation (polyA) involves the addition of adenosine monophosphates and is required for
mRNA maturation. Adenoviruses use splicing along with different polyA site usage to express
7
many different mRNA transcripts from one transcriptional unit. The E2 transcriptional unit has
two polyA sites; one is used to express the shorter E2A RNA earlier in infection but later in
infection this site can be bypassed, and transcription extends to a site downstream to generate the
longer E2B RNAs (Zhao et al, 2014). The E3A and E3B RNAs are differently expressed by use
of different polyA sites. The E3A transcription unit has four different polyA sites, while E3B has
one polyA site, located downstream of the others (Zhao et al, 2014).
1.1.3.3 Regulation of the early-to-late transition of adenovirus replication
The major late promoter (MLP) generates adenovirus mRNA coding for proteins essential for
packaging and assembly such as capsid proteins (including hexon, fiber nd penton base), the
viral protease and histone-like protein (pVII) (Figure 1.1). The mRNAs expressed from the MLP
are products of alternative RNA splicing and polyA site usage. The promoter controls expression
of a single transcript and generates five groups of RNAs by differential polyA site usage,
followed by splicing to give specific mRNAs belonging to each family. These groups or
“families” of RNAs are designated L1-L5. All mRNAs expressed by the MLP share the same
201-nucleotide sequence at their 5’ end; this tripartite leader sequence is non-coding and
promotes the translation of these transcripts over host mRNAs (Zhao et al, 2014).
The L1 family of RNAs consists of two products, L1-52/55K and IIIa (Biasiotto and Akusjarvi,
2015); the L2 precursor RNA is alternatively spliced to give four mRNAs (penton base, pV, pVII
and pX) (Akusjarvi et al, 1981); L3 RNA generates three major transcripts (pVI, hexon, 23K
viral protease) (Prescott et al, 1994); L4 precursor mRNA splicing gives transcripts encoding
four proteins (100K, 22K, 33K and pVIII) (Sittler et al, 1994); and L5 precursor mRNA codes
for the fiber protein (Zhao et al, 2014).
During the early phase of adenovirus replication, the major late promoter is active at a very low
level. Transcription can encompass the L1-L3 sequences (Hales et al, 1988), but only the L1
transcript encoding the 52/55K protein accumulates in the cytoplasm (Biasiotto and Akusjarvi,
2015). The L1 IIIa transcript is expressed only following the onset of genome replication
(Biasiotto and Akusjarvi, 2015). To express all late viral transcripts, the activation of a MLP-
independent promoter located within the major late transcriptional unit, specifically expressing
L4-22K and L4-33K proteins, is required (Morris et al, 2010). This promoter is activated during
viral genome replication, consistent with the observation that early to late transition requires
8
replication of the template DNA genome (Thomas and Matthews, 1980). Viral proteins such as
E1A, E4orf3 and the intermediate protein IVa2 act to stimulate transcription from this promoter
(Morris et al, 2010). L4-33K has been identified as a viral alternative splicing factor, responsible
for the selection of weak 3’ splice sites within the major late transcription unit; in other words,
L4-33K allows for the expression and accumulation of specific transcripts during the late phase
of replication (Törmänen et al, 2006). The L4-22K protein has its own role in the transition from
early to late phase of replication - a feedback mechanism where the L4-22K protein stabilizes the
IVa2 protein, thereby increasing levels of IVa2 for further activation of the L4 promoter and for
MLP activation (Pardo-Mateos et al, 2004; Backström et al, 2010; Morris et al, 2010). At high
levels, the L4-22K protein, without IVa2, binds directly to regions of the MLP, for suppression
of transcription from the MLP (Lan et al, 2017). Given that viral genome replication is required
for the expression of IVa2 and subsequently L4 promoter activity, it is not surprising that viral
DNA replication is required for late gene expression.
1.1.3.4 Role of host splicing factors in adenovirus gene expression
SR (serine-arginine rich) proteins are host factors involved in regulating alternative splicing and
require phosphorylation of their RS domain to be active. Given the important role of RNA
splicing in adenovirus replication, it is not surprising that host SR proteins regulate viral gene
expression. Kanopka et al (1996), using HeLa cell extracts, demonstrated that host SR proteins
bind to an element near the 3’ splice site of IIIa to prevent its expression in vitro. Further
investigation showed that SR proteins become hypophosphorylated during infection due to the
action of viral E4orf4 protein and its interactions with protein phosphatase 2A (PP2A) (Estmer
Nilsson et al, 2001). However, E4orf4 only interacts with a subset of SR proteins, namely
SF2/ASF and SRp30c. As would be expected, E4orf4 binds to hyperphosphorylated proteins and
manipulates these SR proteins to activate splicing of adenovirus mRNAs. Overexpression of
SF2/ASF inhibits the transition from the early to late phase of virus replication (Molin et al,
2000). Such reliance on host RNA splicing machinery and the factors regulating the process
suggests that its manipulation may offer opportunities to suppress adenovirus replication. The
human immunodeficiency virus (HIV) was shown to be reliant on host SR proteins for gene
expression (Mahiet et al, 2016). Work by Wong et al (2013) has shown that manipulation of host
SR proteins may prove to be an effective strategy to inhibit HIV RNA processing. Adenovirus
9
replication may be inhibited using a similar strategy, as suggested by inhibition of adenovirus
replication upon overexpression of SF2/ASF.
Figure 1.3 Adenovirus transcription map. Red indicates early transcripts and yellow represents
late transcripts. Intermediate transcripts are in black. For each transcription unit, the common
promoter is shown (rectangular bracket) and multiple products thereof. Taken from Roberta
Biasiotto and Göran Akusjärvi. Regulation of human adenovirus alternative RNA splicing by the
adenoviral L4-33K and L4-22K proteins. 2015. International Journal of Molecular Sciences.
(16)2: 2893-2912. Used under the terms of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).
10
1.1.4 Adenovirus early gene expression
The purpose of early gene expression is to prepare the infected cell for viral replication and to
shut off the host’s antiviral responses. There are six early transcription units: E1A, E1B, E2A,
E2B, E3 and E4.
1.1.4.1 E1A
E1A is the first viral transcriptional unit expressed (Nevins et al, 1979). The primary E1A
transcript produces three major mRNAs named for their sedimentation coefficients: 13S, 12S
and 9S. Early in infection, the 13S and 12S transcripts are the most abundant while 9S
accumulates later in infection. The 13S transcript codes for the 289-residue (amino acid) product
(289R), while 12S codes for the 243R product and 9S codes for a 55R product. Interestingly, the
latter product is not detected in vivo (Stephens et al, 1987). The most studied E1A proteins are
243R and 289R, which both act to control transcription of viral and host genes during infection
through the two exons which they share (Fig. 1.4). The unique region of the 13S product is a 46-
amino acid sequence that is the result of alternative splicing. The 243R and 289R E1A proteins
contain two and three conserved regions (CR), respectively, that are present in all human and
simian adenovirus types (Kimelman et al, 1985).
E1A proteins do not bind to DNA directly but transactivate expression of the other early regions
of the viral genome, as well as host genes whose products are needed for virus replication,
through binding to host factors. All early regions of the viral genome, including E1A itself, are
transactivated by binding of E1A 289R protein, through its unique CR3 domain, to host factors
that in turn bind at or near the corresponding viral promoter. The viral E2 promoter, and
multiple host promoters, is activated by binding of E2F which is normally sequestered in a
complex with pRb but is released when the CR2 domain, common to both 243R and 289R E1A
proteins, binds to pRb. Host genes activated by E2F include those whose products promote
transcription of genes required for unscheduled DNA synthesis and others like dihydrofolate
reductase which enhances metabolism of the infected cell, allowing for viral replication.
Activation of c-myc by E2F contributes to transforming properties of E1A.
E1A proteins can also repress transcription from host promoters, for example those controlling
expression of proteins that mediate interferon signaling, thereby downregulating the antiviral
response. E1A sequesters host histone acetyltransferase p300 (Avantaggiati et al, 1996) and co-
11
activator CREB-binding protein (CBP) (Lundblad et al, 1995) which normally recruit the histone
acetylase p/CAF to promoters for acetylation of histones thereby promoting unwinding of
chromatin to increase gene accessibility. With p300/CBP sequestered by E1A, p/CAF can no
longer be recruited to certain promoters, resulting in transcriptional repression (Yang et al,
1996). However, 289R E1A can bind p/CAF for transcriptional activation of genes via its CR3
region, as described by Pelka et al (2009). There is a clear balance of transcriptional activation
and repression carried out by E1A proteins, in association with host proteins.
Expression of the E1A proteins induces apoptosis of the cell. Depending on the cell type,
apoptosis can be induced by the 243R protein, by a p53-dependent mechanism (Debbas et al,
1993) or by the 289R protein in a p53-independent manner (Teodoro et al, 1995). Along with
apoptosis, the cell becomes sensitive to TNF-killing mechanisms (Ames et al, 1990). Although
E1A induces cell-killing mechanisms such as apoptosis and TNF-sensitivity, expression of E1B
proteins (19K and 55K) negates these effects (Debbas et al, 1993). A summary of E1A functions
and host-binding partners is shown in Figure 1.4.
12
Figure 1.4 E1A proteins and their interactions with host proteins. Diagram of E1A coding
regions, including the conserved regions (CR1, CR2 and CR3), showing binding sites common
to 289R and 243R proteins for host proteins. Reprinted by permission from Macmillan
Publishers Ltd: Nature Reviews Molecular Cell Biology. Citation: Frisch, S. M., & Mymryk, J.
S. (2002). Adenovirus-5 E1A: Paradox and paradigm. Nature Reviews Molecular Cell Biology,
3, 441 Copyright ©2002.
13
1.1.4.2 E1B
The E1B transcription unit is directly adjacent to the E1A region. E1B is expressed after
transactivation by the large E1A protein (Jones and Shenk, 1979) and also by read-through from
the E1A transcription unit (Maxfield and Spector, 1997). The E1B transcription unit encodes
two mRNAs, 13S and 22S (Montell et al, 1984) that express E1B-19K and E1B-55K proteins,
respectively. One role of E1B proteins is to inhibit apoptosis that is induced by E1A expression.
When E1A proteins act to push the host cell into S-phase, p53 is recruited to apoptotic genes
(Debbas et al, 1993). The E1B-55K protein works to inhibit p53 by directly binding to it. This
direct binding stabilizes p53 and prevents it from activating transcription of pro-apoptotic genes
(Sarnow et al, 1982). E1B-55K can also interact with p53 as a SUMO1 E3-ligase to sumoylate
1% of p53, leading to the association of p53 with PML-nuclear bodies. This association leads to
nuclear export of PML via the CRM1 pathway and accumulation in aggresomes near the
microtubule organizing enter (MTOC) (Pennella et al, 2010). In a separate mechanism, p53 can
be degraded by association with E1B-55K in complex with another adenovirus protein, E4orf6,
and host proteins Elongin B/C, Cullin 5 and RBX1. This complex functions as an E3 ubiquitin
ligase, leading to polyubiquitination of p53 and its degradation (Querido et al, 2001).
E1B-19K prevents apoptosis by a different mechanism. E1B-19K has been found to be a
homologue of host B-cell lymphoma 2 (BCL-2) proteins, specifically mimicking induced
myeloid leukemia cell differentiation protein (MCL-1), an anti-apoptotic protein (Cuconati et al,
2003). E1B-19K binds two proapoptotic proteins, BCL-2 antagonist/killer (BAK) and BCL-2-
like protein 4 (BAX), to prevent their interaction, oligomerization and pore formation in the
mitochondrial membrane. Leaky mitochondrial membranes result in the release of cytochrome c
and other factors leading to the activation of caspase-3 and -9 mediated apoptosis programs
(Cuconati et al, 2002). Therefore, binding of BAK and BAX by E1B-19K results in the
prevention of apoptosis.
E1B-55K has additional functions other than preventing apoptosis. E1B-55K binds to viral
E4orf6 protein in a complex with host proteins to form an E3 ubiquitin ligase. Additional targets
other than the p53 protein include the MRE11-RAD50-NBS1 (MRN) complex which is involved
in the double-strand break repair (DSBR) pathway. In complex with MRN, the E1B-55K/E4orf6
ubiquitin ligase can ubiquitinate subunits of the MRN complex for degradation (Schwartz et al,
14
2008). E1B-55K can also interact with the MRN complex alone and sequester it to PML-nuclear
bodies for subsequent export and accumulation in aggresomes near the MTOC (Liu et al, 2005).
These interactions with the MRN complex prevent it from being recruited to the ends of the
adenovirus genome, which it would recognize as a double-stranded break. Other targets of this
viral ubiquitin ligase are DNA ligase IV (Baker et al, 2007), Bloom helicase (Orazio et al, 2011)
and integrin α3 (Dallaire et al, 2009). The E1B-55K-E4-orf6 complex is also required for late
viral mRNA nuclear export and simultaneously inhibiting host mRNA export (Babiss et al,
1985).
1.1.4.3 E2
Viral proteins needed for adenovirus genome replication (DBP, pTP, Adpol) are products of the
E2 transcriptional unit. Expression is controlled by two promoters, an early and late E2
promoter. All three E2 proteins are expressed from the early promoter which contains two
binding sites for E2F (Kovesdi et al, 1987) and related transcription factors, a noncanonical
TATA-box (Swaminathan et al, 1993) and an EIIA-EF binding site upstream (Jalinot et al,
1986). The E2 late promoter is found about 160 bp upstream from the first transcription start site
and drives continued expression of DBP after the onset of DNA replication (Wold and Ison,
2013). The DBP, encoded by E2A transcripts, has multiple roles during viral genome replication.
As its name implies, this replication protein binds the viral genome, both when it is single- and
double-stranded. DBP has also been implicated in initiation of strand synthesis and elongation
of nascent strands and in displacement of the complementary single DNA strand as the
replication fork advances during replication (de Jong et al, 2003). DBP is expressed at a much
higher level than pre-terminal protein (pTP) and the adenovirus polymerase (Adpol) which are
encoded by the E2B transcripts. The E2B coding sequences are far downstream of the E2A
coding sequences. The E2B transcripts use a polyadenylation site that is also used by IVa2
transcripts and the primary E2 transcript is spliced such that the E2A polyA site is removed. E2A
transcripts continue to be produced at later times post infection from the E2 late promoter (Wold
and Ison, 2013).
1.1.4.4 E3
Once cells become infected, the survival of the cell is important, as premature cell death will
prevent virus replication. The virus has ways of changing the host cell transcription program to
15
prevent apoptosis. The E3 region of the adenovirus genome gives rise to alternatively spliced
mRNAs that encode viral proteins responsible for evading the host immune response. Some
examples of E3 protein functions that contribute to evading the immune response include
inhibition of MHC class I expression at the cell surface by sequestration in the endoplasmic
reticulum (Flomenberg et al, 1992), inhibition of MHC class I transport to the cell surface
(Bacik, 1994) and inhibition of TNFα (Krajcsi et al, 1996), an inflammatory cytokine. The
functional importance of E3 proteins has been reviewed extensively (Lichtenstein et al, 2004).
1.1.4.5 E4
The E4 region of the adenovirus genome expresses several different protein products. The E4
promoter has binding sites for a variety of host proteins which include E4F, E4F1 and ATF
transcription factors (Rooney et al, 1990). The E4 transcription unit is alternatively spliced to
give six mRNA products. The protein products of this region are named for the open-reading
frames (ORFs) from which they are translated starting from the 5’ end.
In contrast to the other transcriptional units of the viral genome whose protein products have
functions related to those of other proteins encoded within the same transcription unit, the
proteins expressed from the E4 transcriptional unit have different functions. Three of these
products will be discussed. E4orf6 functions in a complex with E1B-55K (Section 1.1.4.2) but,
like E1B-55K, can bind to p53 independently and inhibit transcription of genes activated by p53.
In contrast to E1B-55K, E4orf6 interacts with p73, a protein that activates transcription of genes
that interact with p53. In this way, E4orf6 prevents transcription of genes activated by p53 and
by a related protein, p73 (Wienzek et al, 2000).
E4orf3 has similar functions to E1B55K and E4orf6 in that it too inhibits the DSBR to protect
the viral genome termini from degradation and shuttles sequestered MRN complexes out of the
nucleus and forms aggresomes (Stracker et al, 2002). Similar to the E1B-55K-E4orf6 complex,
E4orf3 binds the MRN complex and p53 in the nuclei of cells and interacts with PML protein
(Evans et al, 2005). Further interactions with PML proteins and associated proteins like Daxx
allow E4orf3 to inhibit the host immune response, particularly interferon (Ullman et al, 2008).
These interactions change the morphology of PML proteins- from nuclear bodies in uninfected
cells to “track-like” structures in adenovirus-infected cells (Carvalho et al, 1995). The E4orf3
16
protein and its interactions with its targets are attributed to its function as an E3 SUMO ligase
(Sook-Young and Hearing, 2012). E4orf3 hijacks the host translation machinery by interacting
with components of cytoplasmic P-bodies, which form aggresomes to inhibit host protein
translation and facilitate late viral protein synthesis (Greer et al, 2011).
The E4orf6/7 protein functions to increase binding of E2F to its binding sites by dimerizing itself
and in turn E2F. The binding of the two dimerized proteins increases activation of the E2 early
promoter in a conformation that complements the spatial organization of the E2F binding sites
(Huang et al, 1989).
1.1.5 Adenovirus genome replication
Viral genome replication requires three adenovirus proteins: pTP, Adpol and DBP. The origin of
replication of the adenovirus genome is located at either end of the genome. At this location, the
pTP protein binds dCMP and acts as a primer for extension by the Adpol, which recognizes a
common sequence at the origin (Desiderio et al, 1981). Replication of the viral genome can be
divided into two stages. During the first stage, one strand of the double-stranded genome
becomes displaced, allowing the other strand to be continuously copied from one end to the
other. In the second stage, the displaced strand circularizes into a “panhandle” structure, due to
the self-annealing nature of the inverted terminal repeats. In this way, this single strand recruits
the replication machinery, giving rise to a completely replicated, double-stranded genome
(Lechner et al, 1977). After genome replication, the pTP remains at the termini of the genome
and at the time of virion maturation is cleaved to the mature terminal protein (TP). The genome
is packaged with the pre-terminal proteins at its termini (Webster et al, 1997). Replication of the
viral genome requires host proteins. pTP and Adpol act in complex to initiate replication but
binding of the complex is poor. It has been shown that nuclear factor 1 (NF1) and the
transcription factor Oct-1 are part of the initiation complex, where NF1 interacts with Adpol and
Oct-1 interacts with pTP. These interactions are thought to bend the viral DNA in a way that
promotes initiation of replication (Hoeben et al, 2013).
1.1.6 Adenovirus late gene expression
As discussed in Section 1.1.3.3 the major late promoter (MLP) is responsible for the expression
of genes important for the late phase of adenovirus replication. The major late promoter is active
at a low level during the early phase of replication, but transcripts are terminated prematurely at
17
the polyA site associated with the L3 family of mRNAs. Once the template genome has been
copied, full-length transcription can occur from the major late promoter and promoter activity is
increased, thus beginning the late phase of viral replication. The late genes expressed from the
MLP encode the major capsid proteins, along with accessory proteins required for regulation of
the late phase of infection and scaffolding proteins for virion assembly.
1.1.7 Assembly, packaging and release
The process of assembly refers to the process of major and minor capsid proteins arranging into a
complete capsid while packaging refers to the process of the adenovirus genome being
encapsidated. Figure 1.1 illustrates the major and minor capsid proteins and core proteins of the
virion. Currently, there are two opposing models concerning the order of events required to
assemble a complete virion. One model is that assembly of the virus capsid and packaging of the
viral genome occurs concurrently. In this model, the viral capsid would assemble around the
viral genome. Evidence to support this model includes the observations that only newly
replicated genomes have been found to be packaged (Weber et al, 1985) and that selective
inhibition of viral genome replication using a mutant with altered E2A-72K DNA-binding
protein (DBP) resulted in decreased assembly of virions (Nicolas et al, 1983). The most recent
evidence to support this model is work done by Condezo and San Martín (2017). Using electron
microscopy, it was shown that the peripheral replication zone is a location where packaging
proteins and assembly factors coincide in the nucleus. This zone is also where virions can be
found, along with viral genomes and core proteins, supporting the idea that replication, assembly
and packaging occur concurrently as hypothesized earlier by others (Weber et al, 1985). Aside
from determining the localization of virions, viral genomes, packaging and assembly factors,
electron microscopy also revealed assembly intermediates, which showed viral capsids forming
around viral cores. These observations support the idea that packaging and assembly happen at
the same time.
In the other model, assembly and packaging occur sequentially with insertion of the viral
genome into a pre-formed empty capsid. A packaging signal near the left end of the genome
directs packaging in a polarized fashion (Gräble et al, 1992). In support of this model is the
observation that purification of adenovirus particles in cesium chloride gradients shows both
complete virions and incomplete virions of lighter density which contain only DNA fragments
18
that extend from the extreme left end of the genome (Edvardsson et al, 1976),. Packaging in this
model would involve a molecular motor that uses ATP for energy. Studies by Ostapchuk et al
(2011) support the idea that viral protein IVa2 may be the molecular motor. It has been
determined that IVa2 can bind ATP and that viruses lacking IVa2 produce empty capsids. Others
have identified E4orf6 as a potential portal protein, as discussed by Ahi et al (2017). The
existence of a potential portal protein in addition to a molecular motor, along with identification
of capsids containing incomplete genomes, strongly supports the hypothesis that virus assembly
and packaging occur sequentially and not concurrently.
Regardless of the mechanism by which capsids acquire complete genomes, the newly assembled
virion must undergo maturation cleavage for the virions to be competent for genome delivery in
the target cell. Several virion proteins are synthesized as precursors and incorporated into the
virion in their precursor form. Maturation cleavage occurs as soon as the new virion acquires a
complete genome, when the virion protease is activated by DNA in association with the C-
terminal end of the protein VI precursor (Mangel et al, 2014).
The precise mechanisms for release of progeny virus are unknown although the adenovirus death
protein (ADP), expressed from the E3 transcriptional unit appears to play a role (Wold and Ison,
2013). The autophagy pathway also may be involved (Jiang et al, 2011).
1.2 Cardiotonic steroids
Cardiotonic steroids, also known as cardiac glycosides, are a group of biochemically similar
compounds that are naturally made by many plants and endogenously in the bodies of many
vertebrates (Figure 1.5 A and B). A structural component that all cardiotonic steroids share is a
steroid backbone. A lactone ring structure is found at one end, with a sugar at the other. The
lactone functional group determines the classification of a cardiotonic steroid as either a
cardenolide or bufadienolide. If the lactone is a five-membered ring, unsaturated and a
butytolactone, it is classified as a cardenolide. However, if it is a six-membered, unsaturated
pyrone ring, it is classified as a bufadienolide. These compounds, especially digoxin and
digitoxin, have been used for over 200 years for heart failure and, in recent years, for atrial
fibrillation, given their benefit to cardiac function because of binding to the Na+/K+ ATPase
(NKA) (Gheorghiade et al, 2006). Their effects on several host cell pathways are summarized in
Figure 1.5.
19
1.2.2
Figure 1.5 Cardiotonic steroids and their effects on the cell. Chemical structures of digoxin
(A) and digitoxin (B) are shown. When a cardiotonic steroid (CS) binds to the Na+/K+ ATPase
(NKA), it induces a variety of pathways such as Ras/Raf/MEK/MAPK and PKC. Some of these
pathways, like PKC, can be induced by the influx of calcium ions.. Reprinted by permission from
Macmillan Publishers Ltd: Prassas, I., & Diamandis, E. P. (2008). Novel therapeutic applications
of cardiac glycosides.7, 926, Copyright © 2011.
20
1.2.1 New potential uses for cardiotonic steroids
1.2.1.1 Use of cardiotonic steroids in the context of cancer
Investigations into the NKA as a signaling molecule have created the possibility for digoxin and
other cardiotonic steroids to be re-purposed for other medicinal uses. In the case of breast cancer,
initial observations in the 1980s suggested that the use of digoxin would be beneficial to those
with breast cancer as cancer was less likely to reoccur and showed decreased aggressiveness.
This anti-breast cancer effect was attributed to the observation that the digoxin molecule could
be an estrogen-mimic (Stenkvist et al, 1982). However, nearly 20 years later, conflicting
evidence suggested that patients who were on digoxin or digitoxin, another cardiotonic steroid,
had increased incidence of breast cancer and these cases were more likely to be estrogen
receptor-positive. These observations suggested that digoxin and digitoxin may be estrogen-like
molecules that may exacerbate estrogen receptor signaling (Ahern et al, 2008). The association
between cardiotonic steroids and breast cancer risk is still being evaluated (Karasneh et al, 2017).
Digoxin was found to be a potent inhibitor of prostate cancer in high throughput screening
assays. Men who were on digoxin long-term were found to have lower incidence of prostate
cancer, suggesting that there is in vivo evidence for anti-prostate cancer effects (Platz et al,
2011). These effects may be explained by the estrogenic properties of digoxin which may inhibit
androgenic pathways that are involved in prostate disease. A Phase II clinical trial has been done
to evaluate the use of digoxin in the context of prostate cancer in a small cohort of 16 men,
almost half of which positively responded to digoxin treatment (ClinicalTrials.gov Identifier:
NCT01162135).
Further supporting the observation that digoxin or other cardiotonic steroids could be used as
anti-cancer agents, digoxin was observed to inhibit expression of the protein hypoxia-inducible
factor 1-alpha (HIF-1α). Increased HIF-1α expression has been implicated in many cancers,
given that tumours are likely to exist in an oxygen-deprived state. It was found that digoxin was
able to decrease HIF-1α protein expression without decreasing mRNA levels and independent of
proteasomal degradation pathways. This finding led to the observation that digoxin specifically
decreased translation of HIF-1α mRNA to protein, independent of the mTOR pathways (Zhang
et al, 2008). Targets of HIF-1α, like vascular endothelial growth factor (VEGF), were decreased
at both the mRNA and protein levels. The study was done in the context of prostate cancer cell
21
lines and xenografts, however HIF-1α has been identified in many different types of cancers and
is not unique to prostate cancer alone (Talks et al, 2000). Therefore, the ability of digoxin to
specifically inhibit synthesis of this transcription factor at the level of translation may lead to a
wide spread use of digoxin in cancers expressing high levels of HIF-1α.
1.2.1.2 Cardiotonic steroids as antiviral agents
In this last decade, many studies have been published outlining digoxin, digitoxin or other
cardiotonic steroids as antiviral against diverse viruses. In 2006, Hartley et al published a paper
showing that digoxin was able to inhibit the replication of different DNA viruses including
herpes simplex virus (HSV), adenovirus and human cytomegalovirus (hCMV). Hartley attributed
the basis of the antiviral effects of digoxin to reduced potassium levels within the cell, as studies
have shown that potassium ions are required for HSV replication. Two years later, digitoxin was
shown to inhibit HSV replication during the early phase of infection, at a point before viral DNA
replication (Su et al, 2008). Similarly in 2011, Bertol,et al reported that glucoevatromonoside, a
naturally-occurring cardenolide, inhibited HSV replication by inhibiting viral protein expression
and virus release. HCMV replication can also be inhibited by cardiotonic steroids, at a point after
viral entry, before viral DNA replication (Kapoor et al, 2012). Although Hartley et al (2006) had
focused on DNA viruses and, based on his work, predicted that digoxin would not be effective
against RNA viruses, digoxin and other cardiotonic steroids have been shown to be effective
against human immunodeficiency virus (HIV) (Wong et al, 2013), influenza (Mubareka,
unpublished) and respiratory syncytial virus (RSV) (Moraes, unpublished). Most recently,
digoxin was described to inhibit chikungunya and other alphaviruses as well as reovirus and
others (Ashbrook, 2016). Given the wide range of viruses that digoxin, digitoxin or other
cardiotonic steroids are able to inhibit, developing these drugs as broad-spectrum antivirals may
be a useful strategy to combat viral infections that currently have no specific treatment.
1.2.2 Effects of cardiotonic steroids on the cell
Cardiotonic steroids are mostly known for their interactions with the sodium potassium pump to
inhibit active transport, resulting in potassium ions accumulating outside the cell and sodium
ions inside the cell. However, sodium quickly leaves the cell via the Na+/Ca2+ exchanger (NCX),
resulting in an influx of calcium ions. The influx of calcium ions results in stronger contraction
of cardiac muscle, explaining the basis by which cardiotonic steroids improve cardiac function in
22
heart failure and atrial fibrillation (Katz et al, 1985). However, investigations into novel effects
of cardiotonic steroids have led to other potential uses for them in the clinic. Their use against
many different types of cancers is based on the observation that their interaction with the sodium
potassium pump can activate certain pathways that would cause cell death in some cancer cell
lines. Firstly, the influx of calcium ions can cause apoptosis (Orrenius et al, 2003). With regard
to newly investigated effects, the cardiotonic steroids induce Src kinase which activates a variety
of factors, such as the epidermal growth factor receptor (EGFR), via phosphorylation (Prassas
and Diamandis, 2008). EGFR has been implicated as a target for cancer therapies, as it is
involved in many cell proliferation pathways (Seshacharyulu et al, 2012). Src can also activate
Ras, which is involved in pathways producing reactive oxygen species (ROS), in turn leading to
the activation of NFκB and transcriptional regulation of many genes (Xie et al, 2005).
Cardiotonic steroids can also induce the activation of MAPK pathways, also leading to
expression of many genes (Xie and Askari, 2002). Another study has shown that cardiotonic
steroids inhibit the expression of hypoxia-inducible factor-1 protein (HIF-1), which is implicated
in many cancer types and is considered a drug target (Powis et al, 2004). Many of these
observations were seen in cardiac myocytes; effects of cardiotonic steroids in other cell types
could be different.
Cardiotonic steroids have also been reported to induce alternative splicing in HEK293 cells.
Digitoxin treatment results in decreased expression of two SR proteins (Anderson et al, 2012),
thereby changing the splicing pattern of some exons. This observation was further supported by
the work of Wong et al (2013), which showed that HIV RNA processing was modulated by
digoxin treatment through the same set of SR proteins.
1.3 Research objective and rationale
Since Wong et al (2013) had shown that HIV RNA processing can be modulated using
cardiotonic steroids to inhibit HIV replication, we wanted to investigate the potential antiviral
effects of cardiotonic steroids on adenovirus replication. Since both viruses, HIV and adenovirus,
rely heavily on splicing, we predicted a negative effect of cardiotonic steroid treatment on
adenovirus replication. Experiments done prior to the work described in this thesis indicated that
two cardiotonic steroids, digoxin and digitoxin, inhibited adenovirus replication at a point after
E1A protein expression and before viral DNA replication (Grosso, 2017). This work aimed to
23
further characterize the antiviral effects of digoxin and digitoxin in both lung carcinoma A549
cells and primary human nasal epithelial cells.
24
Materials and Methods
Contributions to this work: Immunofluorescence staining using Tra2β was done by Dr. Alan
Cochrane (University of Toronto).
2.1 Drugs
Digoxin and digitoxin were purchased from Sigma-Aldrich. Drugs were dissolved in dimethyl
sulfoxide (DMSO) at a stock concentration of 10 µM and stored at -20 ̊ C.
2.2 Viruses and cells
2.2.1 Viruses
HAdV-C5 was initially obtained from the American Type Culture Collection (ATCC). Other
viruses were isolated from clinical specimens, specifically, stool of a pediatric patient with
diarrhea (HAdV-A31), lung tissue of a fatal infection in a neonate (HAdV-B35) and an eye swab
from an adult patient with uncomplicated conjunctivitis (HAdV-D).
2.2.2 Cells
A549 cells (human lung carcinoma) were obtained from the ATCC at passage level 76 and used
between passages 89 and 110. HEK 293 (human embryonic kidney) cells (Graham, 1977) were
obtained from F. Graham, McMaster University, Hamilton, Ontario, Canada, at passage 24 and
were used between passages 58 and 90. All cells were maintained in minimal essential medium
(MEM) supplemented with 10% fetal calf serum (FCS) plus penicillin (100 U/ml) and
streptomycin (100μg/ml). Hybridoma cells (2HX-2) were obtained from ATCC and cultured in
minimal essential medium (MEM), as described for A549 cells, but supplemented with
additional glucose (final concentration 0.5%). Cells were tested for mycoplasma using an e-
Myco VALiD Mycoplasma PCR Detection Kit from FroggaBio whenever a new batch of cells
was resurrected for use. Human nasal epithelial cells were received from the Moraes lab at the
Hospital for Sick Children. Cells collected from healthy donors were expanded in liquid culture
then seeded on Transwell inserts and grown at air-liquid interface for 21 days to allow for
differentiation (Cao et al, 2015). Differentiation was confirmed by measuring the transepithelial
resistance of the cell layer with an ohmmeter.
25
2.3 Cell viability assays
Cell viability was evaluated using a metabolic assay and live cell counting. For metabolic assay,
alamarBlue was used. A549 cells were seeded in a 96-well plate at a density of 8x104 cells per
well. The next day, drug was added in increasing concentrations and incubated for 24 hours. A
control for dead cells was included (50% DMSO in cell media) as well as DMSO controls. Each
condition was plated in duplicate. The next day, alamarBlue was added as per manufacturer’s
directions (10uL per 100uL) and fluorescence readings were taken every hour. DMSO-control
readings were averaged and plotted against time to determine what time point would give results
falling in the linear range (up to a maximum of 4 hours).
For live cell counts, cells were seeded in a 24-well plate and treated the next day with increasing
concentrations of drug. 24 hours post-treatment, cells were trypsinized and added to 0.3% trypan
blue (50% cell suspension, 50% trypan blue solution). Cells were counted using disposable
hemocytometers containing 10 grids.
2.4 Virus propagation and assay
All viruses were propagated in HEK293 cells and used in experiments at passage level 3
following primary isolation from clinical samples or receipt from ATCC (HAdV-C5). HEK293
cells in tissue culture flasks were infected at input multiplicity of infection (MOI) <0.1 and
harvested when cytopathic effect (CPE) was complete. Cells were collected by centrifugation for
10 min at 2000 × g, resuspended in a small volume of culture medium then disrupted with five
cycles of freezing and thawing to produce cell lysates. The lysate was clarified by centrifugation
for 10 min at 2000 × g. Clarified lysate and culture fluid were assayed for infectious virus by
endpoint dilution in HEK293 cells, using 60-well Terasaki plates (Sarstedt), as described
previously (Brown, 1985). Titres were calculated by the statistical method of Reed and Muench
(1938) and expressed as infectious units (IU)/ml (Brown, 1985). For purification of HAdV-C5,
clarified lysates were loaded on a 1.2g/ml-1.4g/ml cesium chloride step gradient for
centrifugation at 120000 x g for 1h. The virus band was collected and loaded on a continuous
preformed gradient (1.2-1.4g/ml) for centrifugation at 120000 x g rpm for 2 h. The virus band
was collected and dialysed against storage buffer (50mM Tris-HCl pH7.8, 150mM NaCl, 10mM
26
MgCl2, 10% glycerol) with three changes, 45 min each. The virus suspension was stored in
aliquots at -70 ̊ C. Absorbance was measured using a Nanodrop spectrophotometer and particle
concentration was calculated using the formula 1 OD 260 = 1.1 x 1012 vp/ml. (Maizel et al,
1968).
2.5 Effects of drugs on virus yield
A549 cells were seeded in 6-well plates at a density of 500,000 cells per well. Cells were
infected one day post-seeding at an input MOI of 100 - 400 for HAdV-C5 and -A31, 5 for -B35
and 0.1 for the conjunctival isolate HAdV-D. After one hour adsorption at 37 ̊ C, unadsorbed
inoculum was removed and replaced with fresh culture medium containing DMSO (solvent
control) or drug dissolved in DMSO (duplicate wells per condition). Progeny virus was harvested
at 24 h post infection (hpi) by scraping the cells into the culture fluid, then freeze-thawing the
suspension five times with vortexing. The lysate was clarified by centrifugation at 500 x g for 5
min and titrated by endpoint dilution in 293 cells (Brown, 1985).
2.6 Infection of primary human nasal epithelial cells
Cells were infected within 1-3 days of receipt with purified HAdV-C5 (~1012 particles/ml) at a
1:10 or 1:100 dilution in ALI culture medium. Virus inoculum was added to the apical surface
for an adsorption period of two hours at 37 ̊ C. After the adsorption period, the apical surface was
washed once with PBS; basal medium was replaced with medium containing digoxin at a final
concentration of 35nM, 75nM or 150nM or with the corresponding concentration of DMSO
without drug. Basal medium was replaced every 1-2 days with fresh medium containing DMSO
or digoxin. In some experiments, cells were fixed when cytopathic effect was suspected in cells
incubated without drug. For a time-course experiment, cells were fixed at one, two and three
days pi.
2.7 Collection of RNA and protein samples
A549 cells were seeded on glass coverslips in 6 cm plates at a density of 1x106 cells per plate or
in six-well plates (500,000 cells per well) and infected at an MOI of 100 one day post-seeding.
At 8 or 24 hr p.i., the coverslip was removed, and cells were fixed in 3.7% PFA
(paraformaldehyde in PBS) for 15 min for subsequent staining. Culture medium was removed
from the well and remaining cells were detached in PBS with 2mM EDTA, then the suspension
27
was divided into two tubes and cells pelleted by centrifugation at 800 x g for 3 min. For protein
analysis, cell pellets were re-suspended with RIPA buffer (50 mM Tris-HCl pH 7.5, 150 mM
NaCl, 1% NP- 40, 0.5% sodium deoxycholate, 0.1% SDS) then the lysate was clarified by
centrifugation at 9000 x g for 5 min and stored at -20 ̊C. For RNA analysis, cell pellets were re-
suspended in lysis buffer provided in the Aurum Total RNA extraction kit (Bio-Rad). Total RNA
was extracted as per manufacturer’s instructions and analyzed by RT-PCR or RT-qPCR, as
described later in this section.
2.8 Immunofluorescence staining
2.8.1 Staining of A549 cells
Cells were fixed at indicated times post-infection with 3.7% PFA for 15 min then washed with
PBS, permeabilized for 15 min with 0.1% Triton X in PBS (PBT), then blocked for 45 min with
5% BSA in PBT (BSA-PBT). Cells were incubated with primary antibody for 45 min at 37 ̊ C,
washed three times with PBS then incubated with secondary antibody for 45 min at 37 ̊ C. Cells
were washed twice with PBT, then twice with PBS and coverslips were mounted in PBS
containing DAPI at 0.25ug/mL. Primary antibodies were as follows: monoclonal antibody M73
(ThermoFisher) for E1A, undiluted culture fluid from 2HX-2 hybridoma cells for hexon, rabbit
polyclonal antibody for E4orf6 (a gift from G. Ketner; refer to Mohammadi et al, 2004), rabbit
polyclonal antibody for endogenous Tra2β (Abcam31353), mouse monoclonal antibodies for
E1B-55K and E2A-72K (a gift from A. Levine, PMV Pharma). Secondary antibody was either
goat anti-mouse or anti-rabbit labeled with AlexaFluor488, AlexaFluor594 or FITC and used at
1:200 dilution. Cells were viewed with a 40X objective or 63X oil immersion objective using a
Leica DMR microscope and images captured with Openlab imaging software version 2.0.7.
2.8.2 Staining of primary human nasal epithelial cells
Cells were fixed with cold methanol (-20 ̊ C) added to the apical and basal surfaces for 5 min.
Methanol was removed; cells were washed at both surfaces with PBS once and stored with fresh
PBS on both surfaces. Prior to staining, cells were treated with sodium borohydride (26µM, in
PBS) for 15 min to reduce background signal coming from the Transwell filters. Sodium
borohydride was applied only to the basal side of the filter to avoid damaging the cell layer.
Following sodium borohydride treatment, cells were washed with PBS at the basal surface then
stained using the procedure outlined above. Blocking solution and wash fluid was applied to
28
both the apical and basal sides of the filter; antibody was applied only to the apical side of the
filter, with PBS on the basal surface.
2.9 Western blot analysis
Proteins in cell lysates (prepared in RIPA buffer as described earlier in this section) were
separated on 7 or 10% SDS-PAGE gels and transferred onto PVDF membranes using the BioRad
TurboBlot system according to the manufacturer’s protocol.
For hexon analysis, blots were blocked with 5% w/v skim milk powder diluted in PBS-T (0.05%
Tween20 in PBS) for 1 h, then incubated with undiluted hydridoma (2Hx2) supernatant,
containing anti-hexon antibody, overnight at 4 ̊ C. Following three washes with PBS-T,
secondary anti-mouse antibody conjugated with HRP (diluted 1:5000 in PBS-T) was added for
1 h.
For analysis of E1A protein, blots were blocked with 5% skim milk powder diluted in TBS-T
(0.05% Tween20 in 1xTBS) for 1 h, then incubated at 4 ̊ C overnight with polyclonal E1A
antibody (Santa Cruz, sc-430) diluted 1:2000 in TBS-T. Following at least 5 washes, secondary
anti-rabbit antibody conjugated with HRP (diluted 1:5000 in TBS-T) was added for 1 h. Both
E1A and hexon bands were visualized using ECL+ chemiluminescent solution or BioRad Clarity
ECL Western Blotting Substrate and imaged with BioRad ChemiDoc. Blots were subsequently
probed for tubulin or GAPDH as loading controls.
For analysis of early proteins E1B-55K and E2-72K, blots were blocked in 3% BSA-PBS-T for
one hour, followed by incubation with specific monoclonal antibody (gift from A. Levine, PMV
Pharma) diluted 1:100 in PBS-T, for four hours at room temperature or overnight at 4 ̊C. Blots
were washed three times with PBS-T then incubated with secondary anti-mouse antibody
conjugated with HRP (diluted 1:5000 in PBS-T) for 1hr at room temperature. Blots were washed
again with PBS-T three times. Protein bands were detected using Bio-Rad Clarity ECL Western
Blotting Substrate and imaged with BioRad ChemiDoc. Blots were subsequently probed for
tubulin or GAPDH as loading controls.
29
2.10 Adenoviral DNA and RNA analysis
2.10.1 RNA analysis
To assess transcription of the E1B, E2B and E4 regions, cells were seeded in a six-well plate
(5x105 cells per well). After 19 and 21hpi, RNA was harvested and 1µg was used for reverse
transcription using MMLV-reverse transcriptase. Resulting cDNA (20µL) was diluted to a total
volume of 200µL using autoclaved, ddH2O. Primers were made to E1B, E2B and E4 transcripts
as summarized in Table 2.1. PCR mixtures for adenovirus RNA were made using 5µL of cDNA
diluted 1/10, 2.5µL of ThermolPol Buffer, 2.5µL of 2mM dNTP, 1.0µL each of forward and
reverse primer, 1.5µL of 10X SYBR Green and 11.6µL of autoclaved ddH2O. TATA-binding
protein (TBP) primers were used as a control and cDNA was used undiluted. Melt curves were
done to ensure that only one amplicon resulted from reactions.
The running conditions for each primer set were as follows:
E1B-19K: 50 ̊ C for 2 min, 95 ̊ C for 10 min, 40 cycles of 95 ̊ C for 15s, 60 ̊ C for 1 min, 72 ̊ C
for 1 min followed by 72 ̊ C for 1 min.
E1B-55K: 50 ̊ C for 2 min, 95 ̊ C for 10 min, 40 cycles of 95 ̊ C for 15s, 60 ̊ C for 1 min,72 ̊ C for
1 min followed by 72 ̊ C for 1 min.
pTP: 50 ̊ C for 2 min, 95 ̊C for 10 min, 40 cycles of 95 ̊ C for 15s, 60 ̊ C for 1 min, 72 ̊ C for 1
min followed by 72 ̊ C for 1 min.
E4orf6/7: 50 ̊ C for 2 min, 95 ̊ C for 10 min, 40 cycles of 95 ̊ C for 15s, 60 ̊ C for 1 min, 72 ̊ C for
1 min followed by 72 ̊ C for 1 min.
E4orf6: 50 ̊ C for 2 min, 95 ̊ C for 10 min, 40 cycles of 95 ̊ C for 15s, 55 ̊ C for 1 min, 72 ̊ C for 1
min followed by 72 ̊ C for 1 min.
TBP control: 50.0 ̊ C for 2 min, 95.0 °C for 10 min, 40 cycles of 95.0 ̊ C for 15s, 60.0 ̊ C for 1
min, 72.0 ̊ C for 30s.
30
Table 2.1. Primers used for RT-qPCR experiments
Forward Sequence
(5’ to 3’)
Pos. on
Genome
(nt-nt)
Reverse Sequence
(5’ to 3’)
Pos. on
Genome
(nt-nt)
E1B-
19K
AGGCTTGGGAGTGTTTGGAAG 1718-
1738
GATGAGCCCCACAGAAACCTC 1817-
1797
E1B-
55K
ACATACTGACCCGCTGTTCC 3181-
3200
AAACACCCCGTTCAGGTTCA 3314-
3295
pTP TTGTTGTGTAGGTACTCCGCC 9633-
9653
CCTTGCGACTGTGACTGGTT 9734-
9715
E4orf6 AGGCGCTGTATCCAAAGCTC 33447-
33466
TCCAGCGTGTTTATGAGGGG 33549-
33530
E4orf6/7 CACACGGTTTCCTGTCGAGC 33041-
33060
CCCGTTAAGCAACCGCAAGT 33171-
33152
31
qPCR was done using a BioRad MyiQ Single Color Real-Time PCR Detection System, Standard
Edition and data was collected with iQ5 Optical System Software, version 2.1.
2.10.2 DNA analysis
For DNA analysis, cells were seeded on 6cm plates at a density of 1x106 cells per plate. The next
day, cells were infected with HAdV-C5, MOI 100 for 1 h, inoculum was removed and replaced
with medium containing DMSO or drug at 100nM for digoxin and digitoxin. At the indicated
times post-infection, cells were lifted using 2mM EDTA and washed in PBS. Cells were lysed
using 200µL DNA lysis buffer (10mM Tris-HCl [pH 8.0], 75 mM NaCl, 0.1% SDS, 0.5% NP-
40, 0.5% Tween 20, 0.5 mg/ml proteinase K) at 56 ̊ C for 4 to 5 h and the mixture was boiled for
15 min. Samples were centrifuged at 13000 x g and supernatant was collected. All samples from
a given experiment were processed at the same time. Adenovirus DNA was amplified with
primers specific for the adenovirus E3 region (Ying et al, 2009). The reference gene encoding
TBP was amplified with F: 5’-GATGCCTTATGGCACTGGAC-3’ and R: 5’-
GCCTTTGTTGCTCTTCCAAA-3’ primers (a gift from Lucy Osborne). PCR mixtures were
prepared as follows: 0.4 μL of Taq DNA polymerase (5 U/μL, NEB, Cat. #M0267L), 2.5 μL of
ThermoPol buffer, 1.5 μL of 10X SYBR Green I (Sigma-Aldrich, Cat. #S9430), 2.5 μL of 2.5
mM dNTPs, 1.0 μL of 5' primer (0.1 ug/uL), and 1.0 μL of 3' primer (0.1 μg/μL), 11.1 μL H2O,
and 5 μL of DNA. Standard curves were made using serial 10-fold dilutions of DNA collected 20
h pi from infected cells treated with DMSO. Extracted DNA was diluted 1/10 and 1/100 for
amplification with TBP and E3 primers, respectively. Parameters for E3 primers: 50.0 ̊ C for 2
min, 95.0 ̊ C for 10 min, 35 cycles of 95.0 ̊ C for 15 sec, 65.0 ̊ C for 1 min, 72.0 ̊ C for 1 min.
Parameters for TBP primers: 50.0 ̊ C for 2 min, 95.0 ̊ C for 10 min, 40 cycles of 95.0 ̊ C for 15
sec, 60.0 ̊ C for 1 min, 72.0 ̊ C for 30 sec. qPCR was done using a BioRad MyiQ Single Color
Real-Time PCR Detection System, Standard Edition and data was collected with iQ5 Optical
System Software, version 2.1.
2.11 Transmission electron microscopy
Infected A549 cells in 6-well plates were prepared for transmission electron microscopy 24
hours pi. Primary fixation was done using “universal fixative” (4% paraformaldehyde plus 1%
glutaraldehyde in 0.1M, PBS pH 7.2), warmed to room temperature, for at least 20 min. Cells
were scraped into the fixative, and pelleted by centrifugation at 13000 x g for 5min. Supernatant
32
was removed, and fresh fixative was added to the pellet so as to not disturb the pellet. Samples
were kept at 4 ̊ C for further processing. Pellets then were washed with 0.1M PBS, pH 7.2, at
least three times, with incubations of at least 20 min each, then fixed with 1% osmium tetroxide
in PBS, 0.1M phosphate buffer, pH 7.2 for at least 60 min. Samples were washed with 0.1M
phosphate buffer twice, for 10 min each, in preparation for ethanol dehydration. PBS was
removed, and ethanol was added to the samples in increasing concentrations: 30% ethanol with
2 changes in 10 min, 50% ethanol with 2 changes in 10 min, 70% ethanol with 2 changes in 10
min, 90% ethanol with 2 changes in 15 min and 100% ethanol with three changes in 45 min.
After ethanol dehydration, pellets were washed twice with propylene oxide, 15 min each time, in
preparation for Epon resin infiltration. Resin infiltration was done using increasing parts Epon
resin to propylene oxide. The first infiltration involved 1 part Epon resin mixed with 2 parts
propylene oxide for 0.5h using an agitator, followed by 2 parts Epon resin mixed with 1 part
propylene oxide for 3h using an agitator then 100% Epon resin overnight using an agitator. The
next day, one more change was made using fresh 100% resin for 2 hours. Samples were allowed
to polymerize at 60 ̊ C for at least 48 hours. Samples were stained, sectioned and mounted on
grids for visualization using a Hitachi H-7000. Sectioning of samples was done by Steve Doyle
in the Microscopy Imaging Laboratory, Faculty of Medicine, University of Toronto.
2.12 Statistical Analysis
Where statistical analysis was done, a Student’s t-test for significance using either the Microsoft
Excel or GraphPad Prism platforms were used. Error bars are represented as standard error of the
mean (SEM) unless otherwise specified.
33
Results
3.1 Treatment with digoxin and digitoxin reduces viral yield
The efficacy of digoxin and digitoxin against adenovirus replication was evaluated in A549 (lung
carcinoma) cells. Initial screening was done by Jingwei Chen as a summer student. Cells infected
with HAdV-C5, in a 96-well plate were identified by immunodetection of hexon 24hpi
(Appendix 1.1). Results showed that the proportion of infected cells decreased as the
concentration of drugs increased, predicting a reduction in virus yield. This project began with
yield reduction assays which showed that the yield of progeny virus was reduced by at least two
logs (Figure 3.1). Interestingly, this effect was shown not only for HAdV-C5, but for three other
adenovirus types belonging to three different species: HAdV-A31, HAdV-B35 and a clinical
isolate belonging to HAdV-D. Cells treated with drug showed no obvious signs of toxicity. An
alamarBlue assay showed a modest reduction in metabolic activity of the cells but this reduction
was not indicative of cell death as the viable count as measured by trypan blue exclusion, did not
decrease (Figure 3.2).
3.2 Digoxin and digitoxin inhibit adenovirus replication prior to viral genome replication
Initial experiments to define the block in virus replication analysed expression of the E1A
protein, which is the earliest protein to be expressed, and hexon, the major capsid protein which
is synthesized late in the replication cycle. Consistent with original screening results, digoxin and
digitoxin decreased the number of hexon-positive cells - from ~80% to less than 10%. An overall
decrease in hexon expression was confirmed by western blot analysis (Figure 3.3). In contrast,
there was little effect on E1A protein expression at 8 hpi, the time at which the proportion of
E1A-positive cells was highest in untreated cultures. E1A protein levels, as determined by
western blot analysis, were reduced somewhat but the proportion of E1A positive cells (~50-
60%) was comparable in untreated and treated cultures (Figure 3.4). Late gene expression
requires the replication of the viral genome. The block in hexon expression predicted a block in
genome replication which was confirmed by qPCR using primers specific to the E3 region of
HAdV-C5 (Figure 3.5).
34
Figure 3.1 Digoxin and digitoxin suppress replication of multiple adenovirus species. A549
cells in 6-well plates were infected at 1 day post-seeding. After 60 min of adsorption at 37°C, the
inoculum was removed and replaced with culture medium containing DMSO, digoxin or
digitoxin. Cells and media were collected together at 24 hpi. for titration of total virus by
endpoint dilution in 293 cells. Data points represent average titers of duplicate samples. Error
bars represent standard error of the mean from three experiments for HAdV-C5 with each drug
and HAdV-A31 with digoxin and for two experiments for HAdV-A31 with digitoxin and HAdV-
D (conjunctivitis isolate) with digoxin. Only one experiment was done for HAdV-D with
digitoxin and HAdV-B35 with each drug.
35
Figure 3.2 Digoxin and digitoxin have minimal cytotoxic effects on A549 cells. Cells treated
with digoxin (A) or digitoxin (B) were compared to DMSO-treated cells in an alamarBlue
metabolic assay and by viable counts with trypan blue exclusion. Cells in a 96-well plate were
treated at 1 day post-seeding with digoxin or digitoxin at different concentrations (with duplicate
wells at each concentration), and alamarBlue was added 24 h later. For viable counts, cells
seeded in a 24-well plate were treated at 1 day post-seeding with DMSO or with digoxin or
digitoxin at different concentrations, in duplicate. After 24 h of drug exposure, cells were
trypsinized and the suspension diluted with culture medium and then added to an equal volume
of 0.3% trypan blue. Cells were counted with disposable hemocytometers containing 10 grids.
More than 99% of the cells excluded trypan blue. The viable cell count in each well containing
drug was normalized to the count in the DMSO-treated wells in the same experiment. Error bars
in both panels represent the standard error of the mean from three experiments
36
Figure 3.3 Effect of drug treatment on expression of hexon protein. A549 cells infected with
HAdV-C5 were harvested for immunodetection of hexon at 24 hpi by (A) fluorescence
microscopy and (C) western blot analysis. Blots were re-probed for alpha-tubulin to verify
protein loading. (B)The graph shows the proportion of cells expressing hexon as determined by
counting the number of antibody-stained cells and the total number of nuclei (DAPI) in four to
eight random fields in each of three experiments. Data are based on more than 1,000 cells per
condition. Error bars show the standard error of the mean from three experiments.
37
Figure 3.4 Effect of drug treatment on expression of E1A protein. A549 cells infected with
HAdV-C5 were harvested for immunodetection of E1A at 8 hpi by fluorescence microscopy (A)
and western blot analysis (B). Numbers below the E1A blot indicate the relative expression of
the viral protein normalized to the level in DMSO-treated cells. Blots were re-probed for
alphatubulin to verify protein loading. (C) The graphs show the proportion of cells expressing
E1A as determined by counting the number of antibody-stained cells and the total number of
nuclei (DAPI) in four or five random fields in each of two experiments. Data are based on more
than 200 cells per condition. Error bars show the standard error of the mean from two
experiments. ns, not significant in an unpaired, 2-tailed Student t test.
38
Figure 3.5 Digoxin and digitoxin block adenovirus genome replication. A549 cells were
infected with HAdV-C5 and treated with 100 nM digoxin (A) or 100 nM digitoxin (B)
immediately after the adsorption period. Cells were collected at 0, 10, 16, and 20 or 22 hpi, total
DNA was extracted, and the level of adenovirus DNA was determined by qPCR. The relative
amount of viral DNA at each time point is plotted as the ratio of viral DNA to cellular TBP DNA
(E3/TBP). The results represent three independent experiments for digoxin (A) and two
independent experiments for digitoxin (B), with each sample analyzed in duplicate. Error bars
show the standard error of the mean.
39
3.3 Digoxin and digitoxin affect early gene expression after E1A expression
The block in viral DNA replication implied a block in expression of one or more early proteins,
beyond E1A, that are required for viral genome replication. Antibodies (not commercially
available) were obtained for detection of E1B-55K, E2A-72K DBP, E4orf3 and E4orf6 proteins.
Time course experiments were done to determine optimal times for comparison of protein
expression levels between untreated and treated cells. Western blot analysis of E1B-55K at 14hpi
showed a reduction in protein expression with digoxin or digitoxin treatment, with bands being
barely detectable in three experiments (Figure 3.6). Infected, untreated cells show a variety of
staining patterns: cytoplasmic and nuclear, nuclear with cytoplasmic speckles and only nuclear
(Figure 3.6). Treated cells showed mostly nuclear staining with reduced signal intensity (Figure
3.6). E2A-72K protein levels were reduced to a lesser extent than was E1B-55K expression,
showing a reduction of only ~60% but with a clear difference in staining pattern (Figure 3.7). In
untreated cells, E2A-72K was first detected as a dull diffuse nuclear signal that became more
intense then localized to intranuclear clusters. By 14 hpi, more than half the E2A 72K-positive
cells had multiple clusters which coalesced with time into larger structures apparent at 24 hpi. In
digoxin-treated cultures, E2A-72K was still diffuse within the nucleus of most positive cells 14
hpi. Clusters were apparent in about half of the positive cells by 24 hpi but had not coalesced
into larger structures (Figure 3.7). The delayed early protein, E4orf6, which is expressed near the
end of the early phase of the replication cycle, was affected to some degree. IF staining at 8hpi
showed a decrease in the proportion of E4orf6-positive cells from ~30% to less than 15% (Figure
3.8) but by 24hpi, more than 85% of the cells were positive in both untreated and treated cultures
(Figure 3.8). Preliminary IF staining of E4orf3 suggested a decrease in protein expression, but
there was no obvious change in protein localization (Figure 3.9).
Given that several early proteins were affected in drug-treated cells, it was of interest to examine
early mRNA expression, especially for essential E2 proteins for which antibody was not
available. Figure 3.10 shows the positions of primers targeting transcripts from E1B, E2B and E4
regions for analysis by RT-qPCR. Time-course experiments were done to determine appropriate
time points to assess changes in mRNA expression (Appendix 1.2). E1B transcripts and E4
transcripts, detected with two primer sets each, decreased to 10-50 % of the levels in untreated
cells (Figure 3.11). Of particular interest was the E2B region, encoding pre-terminal protein and
40
viral DNA polymerase. Whereas relative RNA abundance in DMSO-treated cells increased ~ 4-
fold from 19 to 21 hpi, the relative RNA abundance in drug-treated cells was less than ~2% even
at 21 hpi (Figure 3.12).
41
Figure 3.6 Digoxin and digitoxin reduce E1B-55K protein expression. A549 cells infected
with HAdV-C5 were harvested for immunodetection of E1B-55K at 14 hpi by western blot
analysis (A) and fluorescence microscopy (B). Blots were re-probed for GAPDH to verify
protein loading. Images are representative of three western blot experiments and two
immunofluorescence staining experiments.
42
Figure 3.7 The replication protein E2A-72K is decreased and not localized to replication
centers after drug treatment. A549 cells infected with HAdV-C5 were harvested for western
blot analysis at 14hpi (A) and immunodetection at 8, 14 and 24 hpi by fluorescence microscopy
(B). Blots were re-probed for GAPDH to verify protein loading. In (B) the arrows represent the
staining pattern most prominent for untreated cells; red: diffuse, blue: nuclear clusters, orange:
coalesced clusters. Images are representative of three western blot experiments and two
immunofluorescence staining experiments at 14hpi.
43
Figure 3.8 Effect of digoxin and digitoxin on adenoviral E4orf6. A549 cells were infected
with HAdV-C5, treated with digoxin or digitoxin immediately after the adsorption period, and
then fixed at 8 hpi (A) and at 24 hpi (B) for immunodetection of E4orf6. Nuclei are stained with
DAPI. The proportion of cells positive for E4orf6 protein expression is based on total counts of
500 cells in 12 to14 random fields for each condition in panel A and 300 to 400 cells in 8 or 9
random fields for each condition in panel B. Error bars show the standard error of the mean from
two (B) or three (A) experiments. An unpaired, 2-tailed Student t test was used to determine P
values.
44
Figure 3.9 Effect of digoxin and digitoxin on adenoviral E4orf3 protein expression. A549
cells were infected with HAdV-C5, treated with digoxin or digitoxin immediately after the
adsorption period, and then fixed at 24 hpi for immunodetection of E4orf3. Nuclei are stained
with DAPI. Red arrow indicates staining pattern typical of E4orf3, visualized as “track-like
structures” due to interactions with PML proteins. The green arrow indicates a nuclear dot
staining, typical of PML protein in uninfected cells.
45
Figure 3.10 Positioning of primer sets used for RT-qPCR. Primers were made to the E1B (A),
E4 (B) and E2B (C) regions of the adenovirus genome. indicates the location of the
resulting amplicon. Two primer sets were made to the E1B and E4 regions; one was made to the
E2B region. Blue text indicates nucleotide positions of exons; green text indicates nucleotide
positions of splice sites. Information regarding nucleotide positions and splice sites is taken from
Zhao et al. (2014). A new look at adenovirus splicing. Virology, 456-457:329-41.
46
Figure 3.11 E1B and E4 mRNA levels are decreased after digoxin and digitoxin treatment.
A549 cells were infected with HAdV-C5 for one hour and treated with 100nM digoxin or
digitoxin after the adsorption period. RNA was collected at 19 (A) and 21 hpi (B) and 1μg was
used in a reverse transcription reaction. cDNA was diluted and used for qPCR analysis as
described in Materials and Methods. Primers to TBP were used as a control and RNA amounts for
each adenovirus region was measured, relative to TBP levels and normalized to RNA abundance
in the DMSO, 19hpi condition. Data is representative of four experiments and error bars are
representative of the standard error of the mean.
47
Figure 3.12 Digoxin and digitoxin reduce E2B mRNA levels dramatically. A549 cells were
infected with HAdV-C5 for one hour and treated with 100nM digoxin or digitoxin after the
adsorption period. RNA was collected at 19 (A) and 21 hpi (B) and 1μg was used in a reverse
transcription reaction. cDNA was diluted and used for qPCR analysis as described in Materials
and Methods. Primers to TBP were used as a control and RNA amounts for E2B were measured,
relative to TBP levels and normalized to the DMSO 19hpi condition. Data is representative of
four experiments and error bars are representative of the standard error of the mean.
48
3.4 Digoxin and digitoxin induce nuclear changes in treated cells
Hexon production was blocked in most drug-treated cells because the input viral genomes had
not undergone replication, likely reflecting insufficient levels of essential early proteins. It was
of interest to determine whether ultrastructural changes were evident in the drug-treated cells,
despite the block in genome replication and therefore virion production. Nuclear changes usually
are evident by transmission electron microscopy at early times post-infection, prior to genome
replication (Leung and Brown, 2011). The most prominent characteristic is a “blotchy”
appearance of the nuclei, with densely staining material throughout the nucleus (Appendix 1.3).
Cells fixed at one, two and three days post-infection were scored based on their appearance:
uninfected, or with nuclear changes or with progeny virions (Table 3.1). In DMSO-control
cultures, ~20% of the cells showed virions by 1 dpi whereas ~40% showed nuclear changes
without virions and ~40% looked uninfected. By 3 dpi, all DMSO-treated cells looked infected
and only 4% were without virions (Table 3.1). In digoxin-treated cultures, the proportion of
uninfected cells (10-15%) did not change from 2 to 3 dpi. Most cells showed nuclear changes by
2 dpi but very few cells (4%) progressed to virion production by 3 dpi. About half of the cells
with nuclear changes had an atypical appearance, with a visible nucleolus but “blotchy” nucleus
(Figure 3.13).
Although digoxin interferes with expression of early proteins needed for genome replication,
there is sufficient expression to induce changes within the nucleus of infected cells. Without
DNA replication centers, however, the staining pattern of nuclear splicing factor Tra2β remained
diffuse within the nucleus (Figure 3.14).
3.5 Time of addition of digoxin and digitoxin affects their efficacy
It was interesting that addition of drug at the end of the adsorption period affected expression of
early E1B, E2 and E4 proteins but not the E1A protein. Pre-treatment for 4, 12 or 24 hours prior
to infection did reduce E1A expression but only if drug was maintained post-infection. If drug
was removed from pre-treated cells at the time of infection, E1A levels were comparable to the
untreated control (Figure 3.15). Other time of addition experiments showed that drugs had an
inhibitory effect when added up to 8 hpi. Cells were treated at 0, 2 and 8hpi infection and
assayed by immunofluorescent detection of hexon at 24hpi. Drugs added at 2 hpi compromised
hexon production to the same degree as drugs added at the end of the adsorption period. When
49
added at 8 hpi, drugs still blocked hexon expression in some cells, though in fewer cells than
when drug was added immediately after virus adsorption or at 2 hpi (Figure 3.16).
50
Figure 3.13 Infected cells display
abnormal nuclear appearance after
digoxin or digitoxin treatment. A549 cells
were infected with HAdV-C5 for one hour
and treated with 100nM digoxin or
digitoxin after the adsorption period. Cell
pellets were collected and fixed at one, two
and three days pi and processed for
transmission electron microscopy as
described in Materials and Methods. Red
asterisk (*) indicates a cell displaying an
atypical nuclear appearance. Image is
representative of cells collected at 2 dpi and
treated with digoxin.
Table 3.1 Proportion of cells with nuclear changes by transmission electron microscopy
Infected cells were scored either as positive for virions or as having “nuclear changes”.
characteristic of an infected nucleus at early times pi. Note that most treated cells display
evidence of nuclear changes but do not become virion-positive.
*Within this count, about half of these cells had “atypical” nuclei.
51
Figure 3.14 Digoxin and digitoxin change the localization of splicing factor Tra2β in
infected cells. A549 cells were infected with HAdV-C5 for one hour and treated with 100nM
digoxin or digitoxin after the adsorption period. Cells were fixed 24 hpi and immunofluorescence
staining was used to detect Tra2β. Done by Dr. Alan Cochrane.
52
Figure 3.15 Effect of pre-treatment with digoxin and digitoxin on adenovirus E1A protein
expression. A549 cells were incubated with 100 nM drug for 24 h (24PT), 12 h (12PT), or 4 h
(4PT) prior to infection with HAdV-C5. After 60 min of adsorption, the inoculum was replaced
with medium containing DMSO alone (no posttreatment) or containing 100 nM digoxin or
digitoxin (posttreatment). Cells were harvested at 8 hpi and levels of E1A expression were
determined by western blot analysis. The relative expression levels of E1A, normalized to the
level in untreated cells, are shown below the blot.
53
Figure 3.16 Effect of time of addition of digoxin and digitoxin on hexon protein expression.
Cells were infected with HAdV-C5, and drugs (100 nM) were added (A) either immediately
after the adsorption period (t =0 h p.i.) or at 2 or 8 h p.i. (t =2 h p.i. or t =8 h p.i.). Cells were
fixed at 24 hpi for immunodetection of hexon by fluorescence microscopy. (B) The proportion of
cells staining positive for hexon is based on counts from four to eight random fields in each of
three independent experiments, except for digitoxin added at 2 h p.i. (six random fields in one
experiment). The total cell count for each condition exceeded 1,000. Error bars represent the
standard error of the mean from three experiments.
54
3.6 Potassium ions can counter the antiviral effects of digoxin and digitoxin
The known target of digoxin and digitoxin is the Na+/K+-ATPase (NKA). Inhibition of the pump
results in the accumulation of calcium ions (Ca2+) inside the cell and potassium ions (K+) outside
the cell. Hartley et al. (2006) attributed the reduced adenovirus yield in digoxin-treated cells to
reduced activity of the viral DNA polymerase in the altered ionic environment of the cell,
showing that addition of potassium chloride to the culture medium abrogated the effect of
digoxin. It was of interest to determine whether the reduced viral yield in drug-treated cells in
the current study could be affected by manipulating K+ concentration. Culture medium (MEM)
containing digoxin or digitoxin, and supplemented with increasing concentrations of potassium
chloride (KCl), was added to cells after virus adsorption. Western blot analysis showed rescue of
hexon expression 24 hpi, in a dose-dependent manner, in cells treated with digoxin but not with
digitoxin (Figure 3.17). Titration of progeny virus showed a 3-log reduction in yield in drug-
treated cells (Figure 3.18A and B) consistent with earlier results (Figure 3.1). When digoxin was
diluted in MEM containing an additional 50mM KCl, its efficacy was reduced. At most, a one-
log reduction was seen with the highest concentration of digoxin (100 nM) (Figure 3.18A). In
contrast, additional 50 mM KCl had no effect at 100nM digitoxin but did reduce the inhibitory
effect of 25 nM digitoxin (Figure 3.18B). Addition of 50mM KCl without digoxin or digitoxin
reduced hexon expression and compromised virus yield (Figure 3.18A and B), indicating a
delicate relationship between ion balance and adenovirus replication. This reduction in virus
yield with 50mM KCl alone was not due to decreased cell viability, as shown in Figure 3.18C.
3.7 Human nasal epithelial cells can be used as a model for adenovirus replication and assaying drug effects
Having shown that digoxin and digitoxin inhibit adenovirus replication in the continuous A549
(lung epithelial) cell line, it was important to determine whether these drugs also inhibit
adenovirus replication in primary human airway cells. Human nasal epithelial cells (hNEC),
growing at air-liquid interface (ALI) in 24-well plates, were infected from the apical surface with
1/10 or 1/100 dilutions of purified HAdV-C5 (~1012 vp/ml). Cells were pre-treated with IL-8 or
with EGTA or were left untreated, then infected for a two-hour adsorption period, washed once
with PBS and incubated for 7dpi. CPE is difficult to assess by light microscopy, given the
multiple layers of cells on the inserts. Hexon positive cells were detected on all inserts by
55
confocal microscopy, regardless of pre-treatments (Figure 3.19). Almost all cells were infected
in the culture pre-treated with EGTA and infected with the 1:100 dilution (data not shown). In
cultures with fewer cells infected at the outset, foci of infected cells indicated spread of progeny
virus within the culture (Figure 3.19). Since pre-treatment with IL-8 did not increase the number
of infected cells compared to untreated cultures, subsequent experiments were done without pre-
treating the cells. In the next experiment, cells were infected from the apical surface for an
adsorption period of two hours, then inoculum was removed, and cells were washed once with
PBS. Media on the basal surface was replaced with media containing DMSO or digoxin at 35, 75
or 150nM. Cells were fixed when CPE was suspected in the DMSO control cells (4dpi). DMSO
control cells had foci of hexon-positive cells (Figure 3.20). In cells treated with at least 75nM of
digoxin, the presence of only individual hexon-positive cells suggested that spread was inhibited
in these cultures (Figure 3.20). To analyse the kinetics of virus replication in these cultures, a
time-course experiment was done in which cells were fixed one, two and three dpi and stained
for both hexon and E1A. Unfortunately, there were too few cells infected for a comparison of
untreated and treated cultures. Infected cells were not distributed evenly throughout each culture
but, even so, it was evident by 2 dpi that individual, hexon-positive cells had made progeny virus
that had spread to neighbouring cells, as shown by hexon positive cells surrounded by E1A
positive cells (Figure 3.21). By 3 dpi, E1A positive foci appeared to be larger, consistent with
outward spread of progeny virus from infected cells, and hexon positive foci were evident
(Figure 3.21).
56
Figure 3.17 KCl addition to media containing digoxin, but not digitoxin, rescues hexon
protein expression in a dose-dependent manner. Cells were infected with HAdV-C5, and
drugs (100 nM) were added immediately after the adsorption period with media containing
increasing concentrations of KCl (15mM, 30mM and 50mM). Cells were collected at 24 hpi for
immunodetection of hexon by western blotting.
57
Figure 3.18 KCl addition to media containing digoxin and digitoxin rescues viral yields,
dependent on drug concentration. Cells were infected with HAdV-C5, and drugs were added
immediately after the adsorption period at 0, 25nM and 100nM in media containing 50mM KCl.
Cells and media were collected together at 24 hpi for titration of total virus by endpoint dilution
in HEK293 cells (A and B). Data points represent average titers of duplicate samples and error
bars indicate the standard deviation. (C) Cells in a 96-well plate were treated at 1 day post-
seeding with KCl at different concentrations (with duplicate wells at each concentration) and
alamarBlue was added 24 h later. Colorimetric readings were taken using a spectrophotometer
and metabolic rate was calculated relative to untreated controls. Error bars show the standard
deviation of data collected from one experiment.
58
Figure 3.19 Primary human nasal epithelial cells are susceptible to human adenovirus
infection. Cells were infected with different dilutions of purified HAdV-C5 for a two-hour
adsorption period at the apical surface. Inoculum was removed, and the apical surface was
washed once with PBS. IL-8 was used to pre-treat cells four hours before infection, as described
by Lutschg et al (2011), to stimulate exposure of CAR and integrins at the apical surface of the
cells. Cells were fixed 7 days pi with cold methanol for immunodetection of hexon. Images were
collected using confocal microscopy (Zeiss LSM 510). Images show hexon (green) and DAPI
(blue). No obvious differences were found between cells that were untreated or pre-treated with
IL-8; subsequent experiments did not use pre-treatment with IL-8.
59
Figure 3.20 Digoxin is effective in primary human nasal epithelial cells. Cells were infected
with purified HAdV-C5, at a 1:100 dilution for a two-hour adsorption period; the apical surface
was washed once with PBS, and the medium on the basal surface was replaced with medium
containing DMSO or digoxin. Cells were fixed 4 days pi with cold methanol for immunodection
of hexon. Bright foci were seen in the DMSO-control cells, however individual hexon-positive
cells were seen in the treated cells, consistent with a decrease in virus spread.
60
Figure 3.21 Adenovirus infection kinetics in human nasal epithelial cells. Cells were
infected with purified HAdV-C5, at a 1:100 dilution for a two-hour adsorption period and the
apical surface was washed once with PBS. Cells were fixed 1, 2 or 3 dpi with cold methanol
for immunodection of E1A and hexon. 1dpi data is not shown, as very little signal was
detectable for either protein. Images captured for each condition belong to the same field (ie
2dpi images for E1A and hexon belong to the same field).
61
Discussion
Digoxin and digitoxin are cardiotonic steroids, used for centuries to treat heart-related illnesses
such as heart failure (Rietbrock and Woodcock, 1985) and were approved by the US Food and
Drug Administration in 1998 for the treatment of atrial fibrillation (Gheorghiade et al, 2006).
Despite their use in the clinic for many years, they had not been investigated for antiviral
potential until the last decade, beginning with a study reported by Hartley et al (2006). Hartley et
al had predicted the antiviral potential of digoxin against multiple DNA viruses based on the
observation that K+ is required for HSV DNA polymerase (1993). Since then, at least five studies
have described digoxin, digitoxin or other cardiotonic steroids to be antiviral against different
viruses, including hCMV (Kapoor et al, 2012), HSV (Bertol et al, 2011; Su et al, 2008), CHIKV,
alphaviruses and reovirus (Ashbrook et al, 2016). The work presented in this thesis evaluates the
effects of digoxin and digitoxin on adenovirus replication, identifying a block affecting
transcription of early regions beyond E1A expression, in both the continuous A549 cell line and
in primary nasal epithelial cells grown in ALI culture.
4.1 Evaluating the effect of digoxin and digitoxin on adenovirus replication
In all experiments, except those done specifically to look at the effect of pre-treatment, drugs
were added at the end of a one-hour adsorption period at 37 ̊ C. Under these conditions, virion
entry and genome delivery were not blocked, as shown by normal expression of viral E1A
protein, the first viral protein to be expressed post infection. Given that most HAdV-C5 virions
deliver their genome to the nucleus within one hour of uptake (Wang et al, 2013; Greber et al,
1997), it is likely that genome delivery had taken place before the drugs had taken effect. In fact,
it seems that expression of E1A protein escapes inhibition because sufficient transcripts likely
are made before the drugs take effect. Pre-treatment for four hours was sufficient to block E1A
protein expression (Fig 3.15), showing that the drug takes effect within four hours or less
whereas E1A protein was first detected, by immunofluorescent staining, as early as 2 hpi (data
not shown). In pre-treatment experiments, it is possible that a block in entry/genome delivery
contributed to the reduction in E1A expression. The minimum pre-treatment time was four hours
before infection and the effect of 4 hour pre- and post-treatment was the same as pre-treatment
for 24 hours with post-treatment. These results suggest that the drugs take little time to take
62
effect and are easily reversible, given that lack of post-treatment results in E1A protein
expression despite pre-treatment.
Time of addition experiments showed that addition of drug 2 hpi had an effect comparable to
addition of drug at the end of the adsorption period (Figure 3.16). Drug added 8 hpi had less
inhibitory effect than drug added 2 hpi. These results are consistent with a block subsequent to
E1A expression, which was first detectable by immunofluorescence staining at 2 hpi (data not
shown), but affecting expression of other early genes such as E4orf6 (Figure 3.8) as well as E1B
55K (Figure 3.6) and E2A 72K (Figure 3.7) beyond 8 hpi.
The block in hexon (late protein) expression, identified in the original screen, is secondary to the
block in genome replication, as determined by qPCR (Fig 3.5). To switch the virus replication
program from early to late phase, DNA replication is required. Normally, the major late
promoter is active at a low level during the early phase of replication and requires replication of
the template viral genome to switch to the late phase of gene expression (Thomas and Matthews,
1980). Barely detectable levels of progeny viral DNA in drug-treated cells (Figure 3.5) might
reflect an inhibition of viral DNA polymerase activity, as predicted by Hartley et al (2006). Their
“ionic contraviral therapy” approach is based on the hypothesis that low intracellular K+ levels,
due to impaired activity of the NKA, should compromise activity of adenoviral DNA
polymerase, as shown for low potassium levels and activity of HSV DNA polymerase in infected
cell extracts (Hartley et al., 1993, 2006). An alternative possibility is that the drugs inhibit
expression of proteins, other than DNA polymerase, that are necessary for viral genome
replication. To distinguish between these possibilities, antibodies to E1B-55K, E2A-72K, E4orf6
and E4orf3 were used to compare expression of these proteins in infected cells with and without
drug treatment.
E1B-55K was markedly reduced in treated cells as determined by western blotting.
Immunofluorescence staining showed that cells were positive for E1B-55K, but signal intensity
was decreased and detectable only in the nucleus (Figure 3.6). The apparent difference in
localization could be due to the reduction in overall expression of the protein. It is also possible
that the absence of cytoplasmic staining in treated cells reflects an earlier stage of infection in
untreated cells, consistent with delayed E1B-55K expression in treated cells. A time-course
experiment with cells infected and treated with DMSO or digoxin/digitoxin could be done to
63
determine whether untreated cells have E1B-55K localized only to their nuclei earlier in
infection.
Expression of two products of the adenovirus E4 region, E4orf3 and E4orf6, was analysed using
immunofluorescence staining. The proportion of cells positive for E4orf6 protein was decreased
at 8 hpi but recovered at 24 hpi, though cells seemingly expressed lower amounts of protein
(Figure 3.8).
Given the block in viral DNA replication, it was predicted that expression of the DNA binding
protein, E2A-72K, would be compromised. Overall levels of E2A-72K were reduced by about
40%, as determined by western blotting, but the striking difference between treated and untreated
cells was in the staining pattern (Figure 3.7). Immunofluorescence revealed that the localization
was changed; in untreated adenovirus-infected cells, E2A-72K is localized to replication centers
and visualized as clusters. However, most treated cells had a diffuse staining pattern, while a
small proportion of cells showed nuclear clusters, though in some cells the clusters were lower in
number and/or smaller in size. Results of the time-course experiment indicated that the diffuse
staining pattern seen in treated cells is representative of early E2A-72K protein expression. This
staining pattern continues to be seen in most treated cells as long as 24 hpi indicating a
prolonged block in protein localization, as most untreated cells show clusters that have fused by
24 hpi. E2A-72K protein staining is commonly used as a proxy for identifying replication centers
since this protein binds to adenovirus genomes. A diffuse staining pattern may reflect a lack of
viral DNA replication, with few, if any, genomes for the protein to bind to. This idea is
supported by a lack of detection of viral genomes at later times of infection (Fig 3.5).
Immunofluorescence staining of a nuclear splicing factor, Tra2β at 24 hpi showed a similar
staining pattern in untreated cells to that of E2A-72K at the same time post-infection. In treated
cells, Tra2β was diffuse throughout the nucleus, similar to E2A-72K. Lack of Tra2β clusters is
consistent with the absence of viral replication centers.
Antibodies were not available for analysis of other early proteins; selected mRNA transcripts
were analyzed by RT-qPCR instead. Primers were made to amplify coding sequences within
E1B, E2B and E4 transcripts. A time-course experiment using RNA collected from infected
A549 cells was used for RT-qPCR to determine appropriate time points to assess changes in
expression of the chosen transcripts. All mRNA experiments assessed expression at 19 and 21
64
hpi. E1B transcripts were decreased in abundance in treated cells, compared to untreated
controls, at both 19 and 21 hpi, regardless of which primer set was used. Two major transcripts,
22S and 13S, encoding the 55K and 19K proteins, respectively, are expressed from the E1B
region but the two primer sets could not distinguish the two splice variants. Decreased levels of
E1B transcripts at 19 and 21 hpi are consistent with decreased levels of E1B-55K protein at 14
hpi. The E4 region produces eight different mRNAs generated by alternative splicing. The
transcripts encoding E4orf6 and E4orf6/7 proteins contain common 5’ and 3’ sequences but the
E4orf6/7 transcript has an additional region spliced out to generate a smaller RNA. One primer
set was made to amplify a region of the E4orf6 transcript that would be spliced out of the
E4orf6/7 RNA. The second primer set was made to a region that belongs to both transcripts. RT-
qPCR analysis showed that relative RNA abundance was decreased in treated cells, as measured
with both primer sets. With regard to correlation with E4orf6 protein expression data (Figure
3.8), immunofluorescence staining at 24hpi would be consistent with most cells producing lower
levels of E4orf6 RNA and subsequently decreased levels of protein.
There was interest in E2B mRNAs that encode the viral proteins, DNA polymerase (Adpol) and
the terminal protein precursor (pTP), that are essential for viral DNA replication. Terminal
protein (TP) is covalently bound to the 5’ end of each DNA strand of the double-stranded viral
genome. The precursor form (pTP) acts as a primer for DNA replication (Webster et al , 1997).
Primers were made to amplify sequence within the pTP transcript. RT-qPCR analysis showed
that the level of pTP transcript was decreased to a greater degree than the levels of E1B and E4
transcripts in drug-treated cells (Figures 3.11 and 3.12). With reduced levels of E2B RNAs, the
corresponding pTP and Adpol could be at insufficient levels to support viral DNA replication.
The functional significance of reduced E1B and E4 protein levels is less clear. E4 proteins are
important for multiple events, including late viral mRNA expression, transport of late mRNA,
protein synthesis and inhibition of the host response. E1B 55K, in complex with E4orf6, is
important for controlling levels of host proteins p53 and MRE11 to prevent p53-mediated
apoptosis (Querido et al, 2001) and joining of double-stranded DNA ends of viral genomes
(Lakdawala et al, 2008), respectively, and then later to facilitate export of late viral mRNAs, at
the expense of host mRNAs, to the cytoplasm for translation (Gonzalez et al, 2006; Babiss et al,
1985). Nonetheless, digoxin and digitoxin compromise transcription of early regions E1B, E2
and E4, after E1A expression. The block in genome replication likely reflects insufficient levels
65
of the essential pTP and Adpol. It is interesting to speculate that reduced transcription,
particularly of E2B transcripts, may reflect reduced processivity of the host RNA polymerase in
the altered ionic environment of the cell, as predicted by Hartley et al (2006) for viral DNA
polymerase. Although E1A protein is made at apparently normal levels in drug-treated cells
(Figure 3.4), it may not function efficiently to activate transcription of E1B, E2A and E4
promoters and/or it may not displace E2F from pRb for activation of the E2 and necessary host
promoters. Though the precise reason for reduced expression of E1B, E2 and E4 transcripts has
not been determined, the block in virus replication can be overcome by increasing intracellular
potassium concentration.
4.2 Determining the importance of K+ in the antiviral effects of digoxin and digitoxin
When digoxin and digitoxin inhibit the NKA, K+ builds up outside the cell and Ca2+
concentration increases inside the cell. Hartley et al (2006) predicted that low K+ levels in
digoxin-treated cells would compromise viral DNA polymerase and, in turn, adenovirus
replication. In their hands, addition of extracellular K+ counteracted the antiviral effect of the
drug, showing the importance of K+ concentration for digoxin-mediated inhibition of adenovirus
replication. In our experiments, additional extracellular K+ also counteracted the antiviral effect
of digoxin. Adding increasing amounts of KCl to MEM at the time of treatment resulted in the
rescue of hexon expression, in a dose-dependent manner, in cells treated with digoxin but not
digitoxin. Subsequent assay of progeny virus yield showed that the increase in hexon protein
expression seen in digoxin-treated cells correlated with a rescue of progeny virus yield. In the
case of digitoxin, rescue of progeny virus yield was seen only at 25nM but not 100nM digitoxin.
This apparent differential effect of potassium ion supplementation in digoxin and digitoxin-
treated cells was unexpected, given that the mechanism of action of the two drugs is said to be
the same. However, the affinities of K+, digoxin and digitoxin for the NKA are different.
Affinities of digoxin, digitoxin and other cardiotonic steroids were determined for different alpha
subunits of the pump and digitoxin was shown to have a higher affinity than digoxin (Katz et al,
2010). The rescue of late protein expression in digoxin-treated cells suggests that K+ can
compete effectively with digoxin for binding to the pump. In contrast, K+ cannot compete
effectively with digitoxin unless the drug is present at low concentration. It is possible that by
supplementing media with KCl, a concentration gradient is made, allowing K+ to enter the cell
66
via diffusion, though this scenario seems unlikely given the differences seen between digoxin
and digitoxin. It is interesting that the addition of excess K+ in media to untreated cells resulted
in a decrease in viral yield that was not due to cell toxicity (Figure 3.18). This observation shows
the dependence of virus replication on optimal concentrations of specific ions within the cell,
with increased or decreased concentrations of potassium ions having a detrimental effect.
Whereas Hartley et al (2006) attributed the antiviral effect of digoxin to reduced activity of the
viral DNA polymerase at low K+ concentration, our experiments have identified the block at an
earlier stage, specifically affecting early transcription, mediated by host RNA polymerase.
Reduced viral RNA levels likely provide insufficient concentrations of the viral pTP and DNA
polymerase needed for viral DNA replication. If low levels of pTP and Adpol are produced, viral
DNA polymerase activity too may be compromised by low K+ concentration, as suggested by
Hartley et al (2006). Viral DNA replication depends not only on viral proteins but on multiple
host factors whose expression is stimulated by E2F that is displaced from pRb by E1A protein
(Bagchi et al, 1990). Since drug treatment affects early viral transcription by host RNA
polymerase, it might be expected that host transcription might also be compromised.
4.3 Assessing nuclear changes in response to drug treatment
Expression levels of early proteins in drug-treated cells, though reduced, were sufficient to
induce nuclear changes consistent with infection. When cells were examined by transmission
electron microscopy, most of the cells showed nuclear changes consistent with infection by 24
hpi. Typically, infected nuclei lose their nucleoli and acquire a blotchy appearance prior to virion
assembly (Leung and Brown, 2011; Morgan et al, 1960). About half of the infected drug-treated
nuclei developed an appearance like that of infected nuclei in untreated cultures while many
infected drug-treated nuclei had a somewhat atypical appearance (Fig 3.13). It appeared that the
drug-treated cells were responding to the limited viral gene expression that was taking place. An
early cell response seems to be initiated but, without viral DNA synthesis, viral factories do not
form. Uninfected drug-treated cells should be examined to confirm that the nuclear changes are
not due to drug treatment alone, but the atypical nuclear appearance of infected drug-treated cells
is consistent with atypical diffuse staining of E2-72K DBP. It is interesting to note that crystal-
like structures present in both treated and untreated cells, have been observed previously in
adenovirus-infected cells (Morgan et al, 1957; 1960, Franqueville et al, 2008). It has been
determined that such crystal structures, similar in appearance to those in the current study,
67
contain capsid proteins, specifically pentons (fiber bound to penton base) (Figure 3.13), in the
study by Franqueville et al (2008). Observations from the current study, that treated cells show
crystals at 48 and 72hpi, suggest that those cells were expressing capsid proteins and that the
absence of virions in those cells may reflect a block in virion assembly. Immunofluorescence
staining of cells from the same experiment showed that ~25% of drug-treated cells were hexon
positive 2 days pi (data not shown) yet none of the 89 cells examined by EM contained virions at
2 days pi. It is possible that treatment with digoxin creates multiple roadblocks for adenovirus
replication; passing one roadblock (viral DNA replication) may result in inhibition by another
mechanism later in infection (virion assembly).
4.4 Using hNEC as a model for adenovirus infection
All the experiments discussed thus far were done in A549 cells, a cancer cell line. It is important
to test the effect of the two drugs in non-cancer cells. Human primary nasal epithelial cells were
used in this work as a model for adenovirus infection in humans. Cells from the nose are taken
from healthy individuals, amplified in submerged culture and seeded on filters using air-liquid
interface (ALI) culturing techniques. Primary respiratory epithelial cells (from trachea or
bronchi, obtained from surgical specimens or accident victims; available from www.lonza.com)
have been used to study adenovirus infection (for example: Kotha et al, 2015; Lam et al, 2015);
however primary nasal epithelial cells have not been used to date to model adenovirus infection.
Fluorescence microscopy was used to detect cells expressing viral antigen (hexon or E1A
protein). In one experiment, apical washes were assayed for infectious virus but unadsorbed
input virus trapped in the mucus led to high background levels of virus that masked the
appearance of progeny virus. Any progeny virus released into the basal compartment was below
detectable levels. Based on immunofluorescent staining of infected cultures 4 days pi, 75 nM
digoxin was effective in blocking the spread of virus from cells infected at the outset (Fig 3.20).
For better comparison with A549 cell cultures, which usually were fixed 1 dpi, a time-course
experiment was done to determine the kinetics of infection in hNEC cultures with and without
digoxin, by staining for hexon and E1A protein at 1.2 and 3 dpi. Few infected cells were found,
apart from a small region on a filter with cells in the absence of digoxin, fixed 2 dpi. The lack of
more hexon positive cells may reflect donor differences as cells from this donor may not be as
susceptible to infection as cells that were used in previous experiments. The virus inoculum is
68
less likely to account for so few infected cells since all experiments with hNEC cultures were
done with aliquots of the same gradient-purified virus inoculum, stored at -70 ̊ C and not thawed
more than once. Although there were no data from the drug-treated cultures in this experiment,
we opted to examine the untreated cells for a better understanding of the progression of infection
in these cells. Several hexon-positive cells were surrounded by E1A-positive cells 2 dpi.,
showing spread of progeny virus from infected cells to neighbouring cells within 2 days (Fig
3.21). Further spread of progeny virus was evident. Foci of E1A-positive cells were seen as
early as 2 dpi. These foci originated from an isolated, hexon-positive cell that was infected at the
outset, then released progeny virus which spread to neighbouring cells, inducing E1A synthesis
in those cells. At 3 dpi, very small hexon-positive foci of about three cells were seen with 10 or
more E1A-positive cells surrounding them. These observations gave us insight into the
progression of adenovirus infection in hNEC cells.
This work shows the potential but also the challenges of using differentiated hNEC ALI cultures
as a working model for adenovirus infection and assay of antivirals. One challenge is the high
cost and labour intensive care of these cultures from collection to use in experiments.
Differentiation of cultures begins after seeding the cells onto filters, under ALI culture conditions
to facilitate the differentiation process, which typically takes about three weeks. This period does
not include the time it takes to recruit healthy donors, then collect and amplify the cells in
submerged culture. Cost for specialized media, filters and other requirements for maintenance of
these cells can become quite expensive. Secondly, differentiated hNEC produce and excrete
mucus, which can be seen with or without a microscope. This mucus can impede the infection
process, with virus inoculum getting trapped in the mucus, thereby reducing the efficiency of
infection. Thirdly, the receptors required for adenovirus entry in differentiated cells are mostly
unavailable at the apical surface of these cells where virus inoculum is being applied. CAR, the
receptor for HAdV-C5, is a tight junction protein which is located between neighbouring cells
rather than at the apical or basal surface of cells. The absence of CAR and integrins at the apical
surface limit the infection efficiency in these cultures. Lastly, donor variability poses a challenge
to using hNEC for modeling adenovirus infection as there is no way to predict which donor will
provide infection-sensitive or insensitive cells until the experiment is done, potentially wasting
reagents to amplify, differentiate and maintain hNEC that will not be susceptible to infection.
69
The paradigm of drug discovery and development follows a linear progression where a
successful hit in tissue culture is further characterized in animal models, then human subjects.
The observation that digoxin and digitoxin are effective against adenovirus replication and other
viruses, as described in the literature, support evaluating these drugs in the context of an animal
model to determine potential antiviral efficacy in a human. Developing animal models that
accurately mimic human adenovirus infections has been difficult, given the fact that
adenoviruses are species-specific. Human adenoviruses do not replicate in animals commonly
used as models in scientific investigations. However, there are currently two animal models used
to study adenovirus infection- a Syrian hamster model (Thomas et al, 2007) and a mouse model
using murine adenovirus (Lenaerts et al, 2005). In some studies, cotton rats have been used as
well (Prince et al, 1993). Syrian hamsters are documented to be permissive to human adenovirus
infection and show liver pathologies consistent with hepatitis caused by a systemic infection.
Otherwise, these animals do not display respiratory, ocular or digestive pathology that can be
caused by adenovirus infection in humans. In contrast, when mice are inoculated intranasally
with murine adenovirus type 1 (MAdV-1), they show signs of respiratory infection (Lenaerts,
2005). A challenge to using mouse models for studies with digoxin is that the mouse NKA has a
poor response to cardiotonic steroids due to differences in the beta-subunit structure (Akera,
1969). Significantly larger concentrations of drugs would have to be used in mice to elicit a
response, making the study of cardiotonic steroids in mice, and potentially other rodents, difficult
and unreliable. A transgenic animal expressing the human NKA may alleviate cardiotonic
steroid dosing issues seen in mice and other rodents. In the absence of a good animal model and
despite multiple challenges, primary human nasal cells prove to be an attractive alternative,
functionally representing the nasal milieu and can be used to evaluate antiviral potential of
compounds against adenovirus infections at the cellular level.
4.5 Cardiotonic steroids as a pan-antiviral
The observation that digoxin, digitoxin and other cardiotonic steroids are antiviral against
diverse viruses is interesting. Although the specific mechanism of action for the antiviral effects
of cardiotonic steroid treatment on most of these viruses has not been determined, it is possible
that a common set of antiviral mechanisms affecting diverse viruses may exist. Multiple virus
studies have found the same importance of K+ concentrations for the antiviral effects of
cardiotonic steroids, particularly against adenovirus (Hartley et al, 2006 and the current study),
70
hCMV (Kapoor et al, 2012), HSV (Bertol et al, 2011; Su et al, 2008), CHIKV (Ashbrook et al,
2016) and RSV (unpublished). It is important to note that these viruses belong to different
families and therefore have different strategies to replicate in the host. Despite these genomic
differences, all these viruses depend on ionic equilibria to replicate efficiently; disruption of this
equilibrium by the addition of cardiotonic steroids to drive Ca2+ influx and K+ efflux, proves
detrimental to virus replication. Given the importance of ion concentration for replication of
different viruses, manipulating the ionic concentration may prove to be an effective strategy to
combat multiple viruses as Hartley et al first suggested (2006).
The time at which the antiviral activity of cardiotonic steroids takes effect seems to be similar for
the viruses tested. Consistent with our results, experiments with hCMV and HSV showed that
viral DNA replication was inhibited (Su et al, 2008; Bertol et al, 2011; Kapoor et al, 2012). In
the case of CHIKV, antiviral activity was seen after entry of the virus had taken place and before
genome replication (Ashbrook et al, 2016). Comparable timing of the block mediated by digoxin,
digitoxin or other cardiotonic steroids, for different viruses, may suggest a similar mechanism of
action.
Though reports of cardiotonic steroids as antivirals describe specific effects on different viruses,
the modulation of host events has not been assessed for contribution to the antiviral effects. The
initial rationale of the current project was based on the observation that digoxin perturbed HIV
RNA splicing by manipulating host splicing factors SRp20 and Tra2β (Wong et al, 2013). This
observation correlated with the work of Anderson et al (2012) who identified digitoxin as a
modulator of splicing of specific host transcripts through host splicing factors. With splicing
modulation as a strategy to inhibit virus replication, it was hypothesized that digoxin and
digitoxin might inhibit adenovirus replication in this manner. Slight perturbations in splicing of
E1A mRNA were seen but these changes did not translate to any change in E1A protein
expression (Grosso et al, 2017). The primary mechanism of action of cardiotonic steroids against
virus replication cannot be modulation of splicing, as several viruses including CHIK, reovirus
and other viruses described in Ashbrook et al (2016) whose replication is inhibited by digoxin,
do not use splicing as a replication strategy. The effects on splicing of HIV RNA and adenovirus
E1A mRNA are most likely secondary to a signaling event in the host cell that may be
detrimental to different viruses through downstream effects. As discussed previously, the NKA
has been studied as a signaling molecule, inducing specific signaling events that may contribute
71
to the antiviral effect of the cardiotonic steroids. Studies of NKA signaling in cells infected with
the different viruses may uncover common signaling pathways that are essential for replication
of multiple viruses. Supporting this idea is the recently accepted paper by Wong et al (2018)
showing that cardiotonic steroids use the MEK/ERK pathway to elicit modulation of HIV RNA
splicing. Investigation of these pathways in other infected and treated cells may uncover similar
signaling pathways that may reveal new host drug targets to combat virus infections.
Future directions
This project focused on identifying how the cardiotonic steroids, digoxin and digitoxin, inhibit
adenovirus replication. It was shown that these drugs inhibit adenovirus replication early in the
replication cycle, immediately after E1A expression. It is of interest to investigate specifically
how digoxin and digitoxin inhibit transcription of the early genes after E1A protein expression.
As previously discussed, the host protein E2F plays a major role in activation of the E2
transcription units. Reduced transcription from the E2 region may indicate a problem with E2F
binding, which may allude to a problem with E1A function. E1A is responsible for
transactivation of the E2 region by displacing E2F from pRb. It is possible that when digoxin or
digitoxin is added to infected cells, E1A no longer binds to pRb, therefore no longer releasing
E2F. In another mechanism, the larger E1A protein can bind TBP and enhance its transactivation
activity. Similarly, drug addition may prevent binding of E1A and TBP. To address changes in
binding of E1A to known host binding partners, a co-immunoprecipitation assay (co-IP), coupled
with mass spectrometry, can be done to determine the host factors bound by E1A in treated and
untreated cells.
Further investigation into the E2 transcription unit is required. Whereas protein analysis showed
only a modest decrease in E2A-72K mRNA analysis showed that pTP RNA was barely
detectable, suggesting a differential effect of digoxin and digitioxin on E2A and E2B gene
expression. This apparent differential effect of digoxin and digitoxin on E2A or E2B transcripts
needs to be confirmed and investigated. RT-qPCR should be done to specifically amplify E2A-
72K mRNA to confirm that sufficient mRNA is expressed from the E2A transcriptional unit and
antibodies to either pTP or Adpol should be acquired to confirm that these proteins indeed
cannot be detected. These experiments should be done to determine if digoxin or digitoxin affect
transcription elongation or mRNA splicing possibly explaining the effects on the E2 transcription
72
unit. There is a possibility that early transcription, particularly of the region encoding E2B
transcripts, is inhibited due to a lack of RNA polymerase II processivity. All adenovirus
transcriptional units are transcribed by the host RNA polymerase II except for the virus-
associated RNAs (VA RNAs) which are transcribed by host RNA polymerase III. It would be
interesting to do RT-qPCR to quantify VA RNA expression to determine if RNA polymerase III
is also inhibited, suggesting that global adenovirus transcription after E1A protein expression
may be inhibited if VA RNA expression is also decreased. If it is not, it may suggest specific
inhibition of RNA polymerase II and not III.
Lastly, it would be of significance to investigate known signaling pathways induced by
cardiotonic steroid binding to the NKA. The NKA can activate an extensive signaling network
and examining how these pathways may contribute to the antiviral effects of cardiotonic steroids
in the context of adenovirus replication would provide insight into potential new antiviral drug
targets that may be transferable to other virus infections.
Conclusions
This thesis work describes digoxin and digitoxin as inhibitors of adenovirus replication, affecting
expression of early genes subsequent to expression of the immediate E1A protein. Given the
reported effectiveness of these drugs against multiple viruses, re-purposing digoxin and digitoxin
as broad-spectrum antivirals may be an attractive solution to combat viral infections.
73
References
Ahern, T. P., Lash, T. L., Sørensen, H. T., & Pedersen, L. (2008). Digoxin treatment is
associated with an increased incidence of breast cancer: A population-based case-control study.
Breast Cancer Research: BCR, 10(6), R102-R102.
Ahi, Y. S., Hassan, A. O., Vemula, S. V., Li, K., Jiang, W., Zhang, G. J., & Mittal, S. K. (2017).
Adenoviral E4 34K protein interacts with virus packaging components and may serve as the
putative portal.Scientific Reports, 7(1), 7582.
Akera, T., Larsen, F. S., & Brody, T. M. (1969). The effect of ouabain on sodium-and potassium
activated adenosine triphosphatase from the hearts of several mammalian species. Journal of
Pharmacology and Experimental Therapeutics, 170(1), 17.
Akusjärvi, G., & Persson, H. (1981). Gene and mRNA for precursor polypeptide VI from
adenovirus type 2. Journal of Virology, 38(2), 469-482.
Ames, R. S., Holskin, B., Mitcho, M., Shalloway, D., & Chen, M. J. (1990). Induction of
sensitivity to the cytotoxic action of tumor necrosis factor alpha by adenovirus E1A is
independent of transformation and transcriptional activation. Journal of Virology, 64(9), 4115-
4122.
Anderson, E. S., Lin, C., Xiao, X., Stoilov, P., Burge, C. B., & Black, D. L. (2012). The
cardiotonic steroid digitoxin regulates alternative splicing through depletion of the splicing
factors SRSF3 and TRA2B. RNA, 18(5), 1041-1049.
Ashbrook, A. W., Lentscher, A. J., Zamora, P. F., Silva, L. A., May, N. A., Bauer, J. A.,
Dermody, T. S. (2016). Antagonism of the sodium-potassium ATPase impairs chikungunya virus
infection. Mbio, 7(3), e00693-16.
Avantaggiati, M. L., Carbone, M., Graessmann, A., Nakatani, Y., Howard, B., & Levine, A. S.
(1996). The SV40 large T antigen and adenovirus E1a oncoproteins interact with distinct
isoforms of the transcriptional co-activator, p300. The EMBO Journal, 15(9), 2236-2248.
74
Babiss, L. E., Ginsberg, H. S., & Darnell, J. E. (1985). Adenovirus E1B proteins are required for
accumulation of late viral mRNA and for effects on cellular mRNA translation and transport.
Molecular and Cellular Biology, 5(10), 2552-2558.
Bacik, I., Cox, J. H., Anderson, R., Yewdell, J. W., & Bennink, J. R. (1994). TAP (transporter
associated with antigen processing)-independent presentation of endogenously synthesized
peptides is enhanced by endoplasmic reticulum insertion sequences located at the amino- but not
carboxyl-terminus of the peptide. The Journal of Immunology, 152(2), 381.
Backström, E., Kaufmann, K. B., Lan, X., & Akusjärvi, G. (2010). Adenovirus L4-22K
stimulates major late transcription by a mechanism requiring the intragenic late-specific
transcription factor-binding site. Virus Research, 151(2):220-8. doi:
https://doi.org/10.1016/j.virusres.2010.05.013
Bagchi, S., Raychaudhuri, P., & Nevins, J. R. (1990). Adenovirus E1A proteins can dissociate
heteromeric complexes involving the E2F transcription factor: A novel mechanism for E1A
trans-activation. Cell, 62(4), 659-669.
Baker, A., Rohleder, K. J., Hanakahi, L. A., & Ketner, G. (2007). Adenovirus E4 34k and E1b
55k oncoproteins target host DNA ligase IV for proteasomal degradation. Journal of Virology,
81(13), 7034-7040.
Barouch, D. H., Kik, S. V., Weverling, G. J., Dilan, R., King, S. L., Maxfield, L. F., Goudsmit, J.
(2011). International seroepidemiology of adenovirus serotypes 5, 26, 35, and 48 in pediatric and
adult populations. Vaccine, 29(32), 5203-5209.
Berk, A. J. (2013). Adenoviridae. In D. M. Knipe, & P. M. Howley (Eds.), Fields virology (6th
ed., pp. 1704-1731). Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins Health.
Bertol, J. W., Rigotto, C., de Pádua, R. M., Kreis, W., Barardi, C. R. M., Braga, F. C., & Simões,
C. M. O. (2011). Antiherpes activity of glucoevatromonoside, a cardenolide isolated from a
Brazilian cultivar of Digitalis lanata. Antiviral Research, 92(1):73-80 doi:
https://doi.org/10.1016/j.antiviral.2011.06.015
75
Biasiotto, R., & Akusjärvi, G. (2015). Regulation of human adenovirus alternative RNA splicing
by the adenoviral L4-33K and L4-22K proteins. International Journal of Molecular Sciences,
16(2), 2893-2912.
Bremner, K. H., Scherer, J., Yi, J., Vershinin, M., Gross, S. P., & Vallee, R. B. (2009).
Adenovirus transport through a direct cytoplasmic dynein-hexon interaction. Cell Host &
Microbe, 6(6), 523-535.
Brown, M. (1985). Selection of nonfastidious adenovirus species in 293 cells inoculated with
stool specimens containing adenovirus 40. Journal of Clinical Microbiology, 22(2), 205-209.
Burckhardt, C., Suomalainen, M., Schoenenberger, P., Boucke, K., Hemmi, S., & Greber, U.
(2011). Drifting motions of the adenovirus receptor CAR and immobile integrins initiate virus
uncoating and membrane lytic protein exposure. Cell Host & Microbe, 10(2), 105-117.
Cao, H., Ouyang, H., Ip, W., Du, K., Duan, W., Avolio, J., Moraes, T. J. (2015). Testing gene
therapy vectors in human primary nasal epithelial cultures. Molecular Therapy - Methods &
Clinical Development, 2(Supplement C), 15034.
Carvalho, T., Seeler, J. S., Ohman, K., Jordan, P., Pettersson, U., Akusjärvi, G., Dejean, A.
(1995). Targeting of adenovirus E1A and E4-ORF3 proteins to nuclear matrix-associated PML
bodies. Journal of Cell Biology, 131(1), 45.
Chow, L. T., Broker, T. R., & Lewis, J. B. (1979). Complex splicing patterns of RNAs from the
early regions of adenovirus-2. Journal of Molecular Biology, 134(2), 265-303.
Condezo, G. N., & San Martín, C. (2017). Localization of adenovirus morphogenesis players,
together with visualization of assembly intermediates and failed products, favor a model where
assembly and packaging occur concurrently at the periphery of the replication center. PLOS
Pathogens, 13(4), e1006320.
Cuconati, A., Mukherjee, C., Perez, D., & White, E. (2003). DNA damage response and MCL-1
destruction initiate apoptosis in adenovirus-infected cells. Genes & Development, 17(23), 2922-
2932. doi:10.1101/gad.1156903
76
Cuconati, A., & White, E. (2002). Viral homologs of BCL-2: Role of apoptosis in the regulation
of virus infection. Genes & Development, 16(19), 2465-2478. doi:10.1101/gad.1012702
Dallaire, F., Blanchette, P., & Branton, P. E. (2009). A proteomic approach to identify candidate
substrates of human adenovirus E4orf6-E1B55K and other viral cullin-based E3 ubiquitin
ligases. Journal of Virology, 83(23), 12172-12184.
Daniell, E., & Mullenbach, T. (1978). Synthesis of defective viral DNA in HeLa cells infected
with adenovirus type 3. Journal of Virology, 26(1), 61-70.
de Oliveira, C. B. R., Stevenson, D., LaBree, L., McDonnell, P. J., & Trousdale, M. D. (1996).
Evaluation of cidofovir (HPMPC, GS-504) against adenovirus type 5 infection in vitro and in a
New Zealand rabbit ocular model. Antiviral Research, 31(3):165-72. doi: https://doi-
org.myaccess.library.utoronto.ca/10.1016/0166-3542(95)00962-0
de Jong, R., van der Vliet, PC., & Brenkman, A. (2003). Adenovirus DNA replication: Protein
priming, jumping back and the role of the DNA binding protein DBP. Current Topics in
Microbiology and Immunology, 272, 187-211.
Debbas M, W. E. (1993). Wild-type p53 mediates apoptosis by E1A, which is inhibited by E1B.
Genes & Development, 7(4), 546-554. doi:10.1101/gad.7.4.546
Desiderio, S. V., & Kelly, T. J. (1981). Structure of the linkage between adenovirus DNA and
the 55,000 molecular weight terminal protein. Journal of Molecular Biology, 145(2):319-37. doi:
https://doi.org/10.1016/0022-2836(81)90208-4
Dyson, N., Guida, P., McCall, C., & Harlow, E. (1992). Adenovirus E1A makes two distinct
contacts with the retinoblastoma protein. Journal of Virology, 66(7), 4606-4611.
Echavarría, M. (2008). Adenoviruses in immunocompromised hosts. Clinical Microbiology
Reviews, 21(4), 704-715.
Edvardsson, B., Everitt, E., Jörnvall , H., Prage, L., & Philipson, L. (1976). Intermediates in
adenovirus assembly. Journal of Virology, 19(2), 533-547.
77
Egan, C., Jelsma, T. N., Howe, J. A., Bayley, S. T., Ferguson, B., & Branton, P. E. (1988).
Mapping of cellular protein-binding sites on the products of early-region 1A of human
adenovirus type 5. Molecular and Cellular Biology, 8(9), 3955-3959.
Estmer Nilsson, C., Petersen-Mahrt, S., Durot, C., Shtrichman, R., Krainer, A. R., Kleinberger,
T., & Akusjärvi, G. (2001). The adenovirus E4-ORF4 splicing enhancer protein interacts with a
subset of phosphorylated SR proteins. The EMBO Journal, 20(4), 864-871.
Evans, J. D., & Hearing, P. (2005). Relocalization of the Mre11-Rad50-Nbs1 complex by the
adenovirus E4 ORF3 protein is required for viral replication. Journal of Virology, 79(10), 6207-
6215.
Flomenberg, P., Szmulewicz, J., Gutierrez, E., & Lupatkin, H. (1992). Role of the adenovirus
E3-19k conserved region in binding major histocompatibility complex class I molecules. Journal
of Virology, 66(8), 4778-4783.
Florescu, D. F., Pergam, S. A., Neely, M. N., Qiu, F., Johnston, C., Way, S., van der Horst, C.
(2012). Safety and efficacy of CMX001 as salvage therapy for severe adenovirus infections in
immunocompromised patients. Biol Blood Marrow Transplant. 2012 May;18(5):731-8 doi:
https://doiorg.myaccess.library.utoronto.ca/10.1016/j.bbmt.2011.09.007
Franqueville, L., Henning, P., Magnusson, M., Vigne, E., Schoehn, G., Blair-Zajdel, M. E.,
Boulanger, P. (2008). Protein crystals in adenovirus type 5-infected cells: Requirements for
intranuclear crystallogenesis, structural and functional analysis. Plos One, 3(8), e2894.
Gastaldelli, M., Imelli, N., Boucke, K., Amstutz, B., Meier, O., & Greber, U. F. (2008).
Infectious adenovirus type 2 transport through early but not late endosomes. Traffic, 9(12), 2265-
2278. doi:10.1111/j.1600-0854.2008.00835.x
Geisberg, J. V., Lee, W. S., Berk, A. J., & Ricciardi, R. P. (1994). The zinc finger region of the
adenovirus E1A transactivating domain complexes with the TATA box binding protein.
Proceedings of the National Academy of Sciences of the United States of America, 91(7), 2488-
2492.
78
Gheorghiade, M., van Veldhuisen, D. J., & Colucci, W. S. (2006). Contemporary use of digoxin
in the management of cardiovascular disorders. Circulation, 113(21), 2556-2564.
doi:10.1161/CIRCULATIONAHA.105.560110
Gonzalez, R., Huang, W., Finnen, R., Bragg, C., & Flint, S. J. (2006; 2005; 2005). Adenovirus
E1B 55-kilodalton protein is required for both regulation of mRNA export and efficient entry
into the late phase of infection in normal human fibroblasts. Journal of Virology, 80(2), 964-974.
Gräble, M., & Hearing, P. (1992). Cis and trans requirements for the selective packaging of
adenovirus type 5 DNA. Journal of Virology, 66(2), 723-731.
Graham, F. L., Smiley, J., Russell, W. C., & Nairn, R. (1977). Characteristics of a human cell
line transformed by DNA from human adenovirus type 5. Journal of General Virology, 36(1),
59-72. doi: https://doi.org/10.1099/0022-1317-36-1-59
Greber, U. F., Suomalainen, M., Stidwill, R. P., Boucke, K., Ebersold, M. W., & Helenius, A.
(1997). The role of the nuclear pore complex in adenovirus DNA entry. The EMBO Journal,
16(19), 5998-6007.
Greber, U. F., Willetts, M., Webster, P., & Helenius, A. (1993). Stepwise dismantling of
adenovirus 2 during entry into cells. Cell, 75(3):477-86. doi: https://doi.org/10.1016/0092-
8674(93)90382-Z
Greer, A. E., Hearing, P., & Ketner, G. (2011). The adenovirus E4 11k protein binds and
relocalizes the cytoplasmic P-body component Ddx6 to aggresomes. Virology, 417(1), 161-168.
Grosso, F., Stoilov, P., Lingwood, C., Brown, M., & Cochrane, A. (2017). Suppression of
adenovirus replication by cardiotonic steroids. Journal of Virology, 91(3), e01623-16.
Hales, K. H., Birk, J. M., & Imperiale, M. J. (1988). Analysis of adenovirus type 2 L1 RNA 3'-
end formation in vivo and in vitro. Journal of Virology, 62(4), 1464-1468.
Hartley, C., Hartley, M., Pardoe, I., & Knight, A. (2006). Ionic contra-viral therapy (ICVT); a
new approach to the treatment of DNA virus infections. Archives of Virology, 151(12), 2495-
2501.
79
Hartley, C. E., Buchan, A., Randall, S., Skinner, G. R. B., Osborne, M., & Tomkins, L. M.
(1993). The effects of lithium and potassium on macromolecular synthesis in herpes simplex
virus-infected cells. Journal of General Virology, 74(8), 1519-1525. doi:
https://doi.org/10.1099/0022-1317-74-8-1519
Hiebert, S. W., Blake, M., Azizkhan, J., & Nevins, J. R. (1991). Role of E2F transcription factor
in E1A-mediated trans activation of cellular genes. Journal of Virology, 65(7), 3547-3552.
Hiebert, S. W., Lipp, M., & Nevins, J. R. (1989). E1A-dependent trans-activation of the human
MYC promoter is mediated by the E2F factor. Proceedings of the National Academy of Sciences
of the United States of America, 86(10), 3594-3598.
Hoeben, R. C., & Uil, T. G. (2013). Adenovirus DNA replication. Cold Spring Harbor
Perspectives in Biology, 5(3), a013003.
Huang, M., & Hearing, P. (1989). The adenovirus early region 4 open reading frame 6/7 protein
regulates the DNA binding activity of the cellular transcription factor, E2F, through a direct
complex. Genes & Development, 3(11), 1699-1710. doi:10.1101/gad.3.11.1699
Izzedine, H., Launay-Vacher, V., & Deray, G. (2005). Antiviral drug-induced nephrotoxicity.
American Journal of Kidney Diseases, 45(5):804-17. doi: https://doi-
org.myaccess.library.utoronto.ca/10.1053/j.ajkd.2005.02.010
Jalinot, P., & Kédinger, C. (1986). Negative regulatory sequences in the EIa-inducible enhancer
of the adenovirus-2 early EIIa promoter. Nucleic Acids Research, 14(6), 2651-2669.
Jiang, H., White, E. J., Ríos-Vicil C. I., Xu, J., Gomez-Manzano, C., & Fueyo, J. (2011) Human
adenovirus type 5 induces cell lysis through autophagy and autophagy-triggered caspase activity.
Journal of Virology, 85(10), 4720-4729.
Jones, N., & Shenk, T. (1979). An adenovirus type 5 early gene function regulates expression of
other early viral genes. Proceedings of the National Academy of Sciences of the United States of
America, 76(8), 3665-3669.
80
Kapoor, A., Cai, H., Forman, M., He, R., Shamay, M., & Arav-Boger, R. (2012). Human
cytomegalovirus inhibition by cardiac glycosides: Evidence for involvement of the hERG gene.
Antimicrobial Agents and Chemotherapy, 56(9), 4891-4899.
Karasneh, R. A., Murray, L. J., & Cardwell, C. R. (2017). Cardiac glycosides and breast cancer
risk: A systematic review and meta-analysis of observational studies. International Journal of
Cancer, 140(5), 1035-1041. doi:10.1002/ijc.30520
Katz, A., Lifshitz, Y., Bab-Dinitz, E., Kapri-Pardes, E., Goldshleger, R., Tal, D. M., & Karlish,
S. J. D. (2010). Selectivity of digitalis glycosides for isoforms of human Na,K-ATPase. The
Journal of Biological Chemistry, 285(25), 19582-19592.
Katz, A. M. (1985). Effects of digitalis on cell biochemistry: Sodium pump inhibition. Journal of
the American College of Cardiology, 5(5 Suppl A):16A-21A doi: https://doi.org/10.1016/S0735-
1097(85)80459-9
Kent, R., Emanuel, B., Neriah, Y., Levenson, R., & Housman, D. (1987). Ouabain resistance
conferred by expression of the cDNA for a murine Na+, K+-ATPase alpha subunit. Science,
237(4817), 901-903. doi:10.1126/science.3039660
Kimelman, D., Miller, J. S., Porter, D., & Roberts, B. E. (1985). E1a regions of the human
adenoviruses and of the highly oncogenic simian adenovirus 7 are closely related. Journal of
Virology, 53(2), 399-409.
Kotha, P. L. N., Sharma, P., Kolawole, A. O., Yan, R., Alghamri, M. S., Brockman, T. L.,
Excoffon, K. J. D. A. (2015). Adenovirus entry from the apical surface of polarized epithelia is
facilitated by the host innate immune response. PLoS Pathogens, 11(3), e1004696.
Kovesdi, I., Reichel, R., & Nevins, J. R. (1987). Role of an adenovirus E2 promoter binding
factor in E1A-mediated coordinate gene control. Proceedings of the National Academy of
Sciences of the United States of America, 84(8), 2180-2184.
Krajcsi, P., Dimitrov, T., Hermiston, T. W., Tollefson, A. E., Ranheim, T. S., Vande Pol, S. B.,
Wold, W. S. (1996). The adenovirus E3-14.7K protein and the E3-10.4K/14.5K complex of
proteins, which independently inhibit tumor necrosis factor (TNF)-induced apoptosis, also
81
independently inhibit TNF-induced release of arachidonic acid. Journal of Virology, 70(8), 4904-
4913.
Lakdawala, S. S., Schwartz, R. A., Ferenchak, K., Carson, C. T., McSharry, B. P., Wilkinson, G.
W., & Weitzman, M. D. (2008;). Differential requirements of the C terminus of Nbs1 in
suppressing adenovirus DNA replication and promoting concatemer formation. Journal of
Virology, 82(17), 8362-8372.
Lam, E., Ramke, M., Warnecke, G., Schrepfer, S., Kopfnagel, V., Dobner, T., & Heim, A.
(2015) Effective apical infection of differentiated human bronchial epithelial cells and induction
of proinflammatory chemokines by the highly pneumotropic human adenovirus type 14p1. Plos
One, 10(7), e0131201.
Lan, S., Kamel, W., Punga, T., & Akusjärvi, G. (2017). The adenovirus L4-22K protein regulates
transcription and RNA splicing via a sequence-specific single-stranded RNA binding. Nucleic
Acids Research, 45(4), 1731-1742.
Lechner, R. L., & Kelly, T. J. (1977). The structure of replicating adenovirus 2 DNA molecules.
Cell, 12(4):1007-20. doi: https://doi.org/10.1016/0092-8674(77)90165-9
Lenaerts, L., Verbeken, E., De Clercq, E., & Naesens, L. (2005). Mouse adenovirus type 1
infection in SCID mice: An experimental model for antiviral therapy of systemic adenovirus
infections. Antimicrobial Agents and Chemotherapy, 49(11), 4689-4699.
Leung, T. K. H., & Brown, M. (2011). Block in entry of enteric adenovirus type 41 in HEK293
cells. Virus Research, 156(1-2):54-63. doi: https://doi.org/10.1016/j.virusres.2010.12.018
Lion, T. (2014). Adenovirus infections in immunocompetent and immunocompromised patients.
Clinical Microbiology Reviews, 27(3), 441-462.
Lichtenstein, D., Toth, K., Doronin, K., Tollefson, A., & Wold, W. (2004). Functions and
mechanisms of action of the adenovirus E3 proteins. International Reviews of Immunology,
23(1-2), 75-111.
82
Liu, Y., Shevchenko, A., Shevchenko, A., & Berk, A. J. (2005). Adenovirus exploits the cellular
aggresome response to accelerate inactivation of the MRN complex. Journal of Virology, 79(22),
14004-14016.
Luisoni, S., Suomalainen, M., Boucke, K., Tanner, L., Wenk, M., Guan, X., Greber, U. (2015).
Co-option of membrane wounding enables virus penetration into cells. Cell Host & Microbe,
18(1), 75-85.
Lundblad, J. R., Kwok, R. P. S., Laurance, M. E., Harter, M. L., & Goodman, R. H. (1995).
Adenoviral ElA-associated protein p300 as a functional homologue of the transcriptional co-
activator CBP. Nature, 374(6517), 85-88.
Mahiet, C., & Swanson, C. M. (2016). Control of HIV-1 gene expression by SR proteins.
Biochemical Society Transactions, 44(5), 1417.
Maizel, J. V., White, D. O., & Scharff, M. D. (1968). The polypeptides of adenovirus: I.
evidence for multiple protein components in the virion and a comparison of types 2, 7A, and 12.
Virology, 36(1), 115-125.
Mangel, W. F., & San Martín, C. (2014). Structure, function and dynamics in adenovirus
maturation. Viruses, 6(11), 4536-4570.
Maxfield, L. F., & Spector, D. J. (1997). Readthrough activation of early adenovirus E1b gene
transcription. Journal of Virology, 71(11), 8321-8329.
Mohammadi, E. S., Ketner, E. A., Johns, D. C., & Ketner, G. (2004). Expression of the
adenovirus E4 34k oncoprotein inhibits repair of double strand breaks in the cellular genome of a
293-based inducible cell line. Nucleic Acids Research, 32(8), 2652-2659.
Molin, M., & Akusjärvi, G. (2000). Overexpression of essential splicing factor ASF/SF2 blocks
the temporal shift in adenovirus pre-mRNA splicing and reduces virus progeny formation.
Journal of Virology, 74(19), 9002-9009.
Montell, C., Fisher, E. F., Caruthers, M. H., & Berk, A. J. (1984). Control of adenovirus E1B
mRNA synthesis by a shift in the activities of RNA splice sites. Molecular and Cellular Biology,
4(5), 966-972.
83
Morgan, C., Godman, G. C., Breitenfeld, P. M., & Rose, H. M. (1960). a correlative study by
electron and light microscopy of the development of type 5 adenovirus: I. electron microscopy.
The Journal of Experimental Medicine, 112(2), 373-382.
Morgan, C., Godman, G. C., Rose, H. M., Howe, C., & Huang, J. S. (1957). Electron
microscopic and histochemical studies of an unusual crystalline protein occurring in cells
infected by type 5 adenovirus. preliminary observations. The Journal of Biophysical and
Biochemical Cytology, 3(3), 505-508.
Morris, S. J., Scott, G. E., & Leppard, K. N. (2010). Adenovirus late-phase infection is
controlled by a novel L4 promoter. Journal of Virology, 84(14), 7096-7104.
Nevins, J. R., Ginsberg, H. S., Blanchard, J., Wilson, M. C., & Darnell, J. E. (1979). Regulation
of the primary expression of the early adenovirus transcription units. Journal of Virology, 32(3),
727-733.
Nicolas, J. C., Sarnow, P., Girard, M., & Levine, A. J. (1983). Host range temperature-
conditional mutants in the adenovirus DNA binding protein are defective in the assembly of
infectious virus. Virology, 126(1), 228-239.
Orazio, N. I., Naeger, C. M., Karlseder, J., & Weitzman, M. D. (2011). The adenovirus
E1b55K/E4orf6 complex induces degradation of the bloom helicase during infection. Journal of
Virology, 85(4), 1887-1892.
Pardo-Mateos, A., & Young, C. S. H. (2004). Adenovirus IVa2 protein plays an important role in
transcription from the major late promoter in vivo. Virology, 327(1), 50-59 doi:
https://doi.org/10.1016/j.virol.2004.06.011
Pelka, P., Ablack, J. N. G., Shuen, M., Yousef, A. F., Rasti, M., Grand, R. J., Mymryk, J. S.
(2009). Identification of a second independent binding site for the pCAF acetyltransferase in
adenovirus E1A. Virology, 391(1), 90-98
Pennella, M., Liu Y, W., JL, Kim, C., & Berk, A. (2010). Adenovirus E1B 55-kilodalton protein
is a p53-SUMO1 E3 ligase that represses p53 and stimulates its nuclear export through
84
interactions with promyelocytic leukemia nuclear bodies. Journal of Virology, 84(23), 12210--
12225. doi:10.1128/JVI.01442-10
Platz, E. A., Yegnasubramanian, S., Liu, J. O., Chong, C. R., Shim, J. S., Kenfield, S. A.,
Nelson, W. G. (2011). A novel two-stage, transdisciplinary study identifies digoxin as a possible
drug for prostate cancer treatment. Cancer Discovery, 2011(1), 68-77.
Potter, R. N., Cantrell, J. A., Mallak, C. T., & Gaydos, J. C. (2012). Adenovirus-associated
deaths in US military during postvaccination period, 1999-2010. Emerging Infectious Diseases,
18(3), 507-509.
Powis, G., & Kirkpatrick, L. (2004). Hypoxia inducible factor-1α as a cancer drug target.
Molecular Cancer Therapeutics, 3(5), 647.
Prassas, I., & Diamandis, E. P. (2008). Novel therapeutic applications of cardiac glycosides.
Nature Reviews Drug Discovery, 7(11), 926-35. doi: 10.1038/nrd2682
Prescott, J., & Falck-Pedersen, E. (1994). Sequence elements upstream of the 3' cleavage site
confer substrate strength to the adenovirus L1 and L3 polyadenylation sites. Molecular and
Cellular Biology, 14(7), 4682-4693.
Prince, G. A., Porter, D. D., Jenson, A. B., Horswood, R. L., Chanock, R. M., & Ginsberg, H. S.
(1993). Pathogenesis of adenovirus type 5 pneumonia in cotton rats (Sigmodon hispidus). Journal
of Virology, 67(1), 101-111.
Querido, E., Blanchette, P., Yan, Q., Kamura, T., Morrison, M., Boivin, D., Branton, P. (2001).
Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism
involving a cullin-containing complex. Genes & Development, 15(23), 3104-3117.
doi:10.1101/gad.926401
Ramsay, I. D., Attwood, C., Irish, D., Griffiths, P. D., Kyriakou, C., & Lowe, D. M. (2017).
Disseminated adenovirus infection after allogeneic stem cell transplant and the potential role of
brincidofovir – case series and 10 year experience of management in an adult transplant cohort.
Journal of Clinical Virology, 96:73-79. doi: https://doi-
org.myaccess.library.utoronto.ca/10.1016/j.jcv.2017.09.013
85
Reddy, V. S., & Nemerow, G. R. (2014). Structures and organization of adenovirus cement
proteins provide insights into the role of capsid maturation in virus entry and infection.
Proceedings of the National Academy of Sciences, 111(32), 11715-11720.
Reed, L., & Muench, H. (1938). A simple method of estimating fifty percent end-points. The
American Journal of Hygiene, 27(3), 493-497.
Rietbrock, N., & Woodcock, B. G. (1985). Two hundred years of foxglove therapy Digitalis
purpurea 1785–1985. Trends in Pharmacological Sciences, 6, 267-269.
Roelvink, P. W., Lizonova, A., Lee, J. G. M., Li, Y., Bergelson, J. M., Finberg, R. W., Wickham,
T. J. (1998). The Coxsackievirus-adenovirus receptor protein can function as a cellular
attachment protein for adenovirus serotypes from subgroups A, C, D, E, and F. Journal of
Virology, 72(10), 7909-7915.
Rooney, R., Raychaudhuri, P., & Nevins, J. (1990). E4F and ATF, two transcription factors that
recognize the same site, can be distinguished both physically and functionally: A role for E4F in
ElA trans activation. Molecular and Cellular Biology, 10, 5138-5149.
Sarnow, P., Ho, Y. S., Williams, J., & Levine, A. J. (1982). Adenovirus E1b-58kd tumor antigen
and SV40 large tumor antigen are physically associated with the same 54 kd cellular protein in
transformed cells. Cell, 28(2), 387-394. doi: 10.1016/0092-8674(82)90356-7.
Schreiner, S., Martinez, R., Groitl, P., Rayne, F., Vaillant, R., Wimmer, P., Wodrich, H. (2012).
Transcriptional activation of the adenoviral genome is mediated by capsid protein VI. PLOS
Pathogens, 8(2), e1002549.
Schwartz, R. A., Lakdawala, S. S., Eshleman, H. D., Russell, M. R., Carson, C. T., & Weitzman,
M. D. (2008). Distinct requirements of adenovirus E1b55K protein for degradation of cellular
substrates. Journal of Virology, 82(18), 9043-9055.
Seshacharyulu, P., Ponnusamy, M. P., Haridas, D., Jain, M., Ganti, A., & Batra, S. K. (2012).
Targeting the EGFR signaling pathway in cancer therapy. Expert Opinion on Therapeutic
Targets, 16(1), 15-31.
86
Sittler, A., Gallinaro, H., & Jacob, M. (1994). Upstream and downstream cis-acting elements for
cleavage at the L4 polyadenylation site of adenovirus-2. Nucleic Acids Research, 22(2), 222-
231.
Sohn, S., & Hearing, P. (2012). Adenovirus regulates sumoylation of Mre11-Rad50-Nbs1
components through a paralog-specific mechanism. Journal of Virology, 86(18), 9656-9665.
Spector, D. J., McGrogan, M., & Raskas, H. J. (1978). Regulation of the appearance of
cytoplasmic RNAs from region 1 of the adenovirus 2 genome. Journal of Molecular Biology,
126(3), 395-414.
Stenkvist, B., Bengtsson, E., Dahlqvist, B., Eriksson, O., Jarkrans, T., & Nordin, B. (1982).
Cardiac glycosides and breast cancer, revisited. New England Journal of Medicine, 306(8), 484-
484. doi:10.1056/NEJM198202253060813
Stephens, C., & Harlow, E. (1987). Differential splicing yields novel adenovirus 5 E1A mRNAs
that encode 30 kd and 35 kd proteins. The EMBO Journal, 6(7), 2027-2035.
Stracker, T. H., Carson, C. T., & Weitzman, M. D. (2002). Adenovirus oncoproteins inactivate
the Mre11-Rad50-NBS1 DNA repair complex. Nature, 418 (6895):348-52.
Su, C., Hsu, J. T., Hsieh, H., Lin, P., Chen, T., Kao, C., Chang, S. (2008). Anti-HSV activity of
digitoxin and its possible mechanisms. Antiviral Research, 79(1):62-70. doi:
https://doi.org/10.1016/j.antiviral.2008.01.156
Suomalainen, M., Nakano, M. Y., Keller, S., Boucke, K., Stidwill, R. P., & Greber, U. F. (1999).
Microtubule-dependent plus- and minus end–directed motilities are competing processes for
nuclear targeting of adenovirus. The Journal of Cell Biology, 144(4), 657-672.
doi:10.1083/jcb.144.4.657
Talks, K. L., Turley, H., Gatter, K. C., Maxwell, P. H., Pugh, C. W., Ratcliffe, P. J., & Harris, A.
L. (2000). The expression and distribution of the hypoxia-inducible factors HIF-1α and HIF-2α
in normal human tissues, cancers, and tumor-associated macrophages. The American Journal of
Pathology, 157(2), 411-421.
87
Teodoro, J., Shore, G., & Branton, P. (1995). Adenovirus E1A proteins induce apoptosis by both
p53-dependent and p53-independent mechanisms. Oncogene, 11(3), 467-474.
Thomas, G.P., & Mathews, M. B. (1980). DNA replication and the early to late transition in
adenovirus infection. Cell, 22(2), 523-533.
Thomas, M., Spencer, J., & Wold, W. (2007). Use of the Syrian hamster as an animal model for
oncolytic adenovirus vectors. Methods in Molecular Medicine, 130, 169-183.
Törmänen, H., Backström, E., Carlsson, A., & Akusjärvi, G. (2006). L4-33K, an adenovirus-
encoded alternative RNA splicing factor. Journal of Biological Chemistry, 281(48), 36510-
36517.
Toth, K., Ying, B., Tollefson, A. E., Spencer, J. F., Balakrishnan, L., Sagartz, J. E., Wold, W. S.
M. (2015). Valganciclovir inhibits human adenovirus replication and pathology in permissive
immunosuppressed female and male Syrian hamsters. Viruses, 7(3), 1409-1428.
Ullman, A. J., & Hearing, P. (2008). Cellular proteins PML and Daxx mediate an innate antiviral
defense antagonized by the adenovirus E4 ORF3 protein. Journal of Virology, 82(15), 7325-
7335.
Vora, S. B., Brothers, A. W., & Englund, J. A. (2017). Renal toxicity in pediatric patients
receiving cidofovir for the treatment of adenovirus infection. Journal of the Pediatrics Infectious
Diseases Society, 6(4):399-402. doi: 10.1093/jpids/pix011
Wang, I., Suomalainen, M., Andriasyan, V., Kilcher, S., Mercer, J., Neef, A., Greber, U. (2013).
Tracking viral genomes in host cells at single-molecule resolution. Cell Host Microbe,
14(4):468-80. doi: 10.1016/j.chom.2013.09.004.
Weber, J., & Stich, H. F. (1969). Electron microscopy of cells infected with adenovirus type 2.
Journal of Virology, 3(2), 198-204.
Weber, J. M., Déry, C. V., Amin Mirza, M., & Horvath, J. (1985). Adenovirus DNA synthesis is
coupled to virus assembly. Virology, 140(2), 351-359.
88
Webster, A., Leith, I. R., Nicholson, J., Hounsell, J., & Hay, R. T. (1997). Role of preterminal
protein processing in adenovirus replication. Journal of Virology, 71(9), 6381-6389.
Weinmann, R., Raskas, H. J., & Roeder, R. G. (1974). Role of DNA-dependent RNA
polymerases II and III in transcription of the adenovirus genome late in productive infection.
Proceedings of the National Academy of Sciences of the United States of America, 71(9), 3426-
3430.
Wickham, T. J., Mathias, P., Cheresh, D. A., & Nemerow, G. R. (1993). Integrins αvβ3 and αvβ5
promote adenovirus internalization but not virus attachment. Cell, 73(2), 309-319. doi:
https://doi.org/10.1016/0092-8674(93)90231-E
Wienzek, S., Roth, J., & Dobbelstein, M. (2000). E1B 55-kilodalton oncoproteins of adenovirus
types 5 and 12 inactivate and relocalize p53, but not p51 or p73, and cooperate with E4orf6
proteins to destabilize p53. Journal of Virology, 74(1), 193-202.
Wold, W., & Ison, M. (2013). Adenoviruses. In D. M. Knipe, & P. M. Howley (Eds.), Fields
virology (6th ed., pp. 1732-1767). Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins
Health
Wold, W. S. M., Tollefson, A. E., & Hermiston, T. W. (1995). E3 transcription unit of
adenovirus. In W. Doerfler, & P. Böhm (Eds.), The molecular repertoire of adenoviruses I:
Virion structure and infection (pp. 237-274). Berlin, Heidelberg: Springer Berlin Heidelberg.
Wong, R. W., Balachandran, A., Ostrowski, M. A., & Cochrane, A. (2013). Digoxin suppresses
HIV-1 replication by altering viral RNA processing. PLoS Pathogens, 9(3): e1003241.
Wong, R.W. et al (2018). Cardiac glycoside/aglycones inhibit HIV-1 gene expression by a
mechanism requiring MEK1/2-ERK1/2 signaling. Nature Scientific Reports (in press).
Xie, Z., & Xie, J. (2005). The Na/K-ATPase-mediated signal transduction as a target for new
drug development. Frontiers in Bioscience, (10), 3100-3109.
Yabiku, Simone Tiemi, Yabiku, Mariann Midori, Bottós, Kátia Mantovani, Araújo, Aline Lutz,
Freitas, Denise de, & Belfort Jr., Rubens. (2011). Ganciclovir 0.15% ophthalmic gel in the
89
treatment of adenovirus keratoconjunctivitis. Arquivos Brasileiros de Oftalmologia, 74(6), 417-
421. https://dx.doi.org/10.1590/S0004-27492011000600007
Yang, X., Ogryzko, V. V., Nishikawa, J., Howard, B. H., & Nakatani, Y. (1996). A p300/CBP-
associated factor that competes with the adenoviral oncoprotein E1A. Nature, 382(6589), 319-
324.
Ying, B., Toth, K., Spencer, J., Meyer, J., Tollefson, A., Patra, D., Lichtenstein, D. (2009).
INGN 007, an oncolytic adenovirus vector, replicates in Syrian hamsters but not mice:
Comparison of biodistribution studies. Cancer Gene Therapy, 16(8), 625-637.
Yoshitomi, H., Sera, N., Gonzalez, G., Hanaoka, N., & Fujimoto, T. (2017). First isolation of a
new type of human adenovirus (genotype 79), species human mastadenovirus B (B2) from
sewage water in Japan. Journal of Medical Virology, 89(7), 1192-1200. doi:10.1002/jmv.24749
Zhang, H., Qian, D. Z., Tan, Y. S., Lee, K., Gao, P., Ren, Y. R., Semenza, G. L. (2008). Digoxin
and other cardiac glycosides inhibit HIF-1α synthesis and block tumor growth. Proceedings of
the National Academy of Sciences of the United States of America, 105(50), 19579-19586.
Zhang, Y., & Bergelson, J. M. (2005). Adenovirus receptors. Journal of Virology, 79(19),
12125-12131.
Zhao, H., Chen, M., & Pettersson, U. (2014). A new look at adenovirus splicing. Virology, 456-
457:329-41. doi: 10.1016/j.virol.2014.04.006.
90
Appendix 1.1 Initial screening identifies digoxin and digitoxin as potential adenovirus
inhibitors. A549 cells in 96-well plates were infected at one day postseeding with HAdV-
C5 and treated with different concentrations of digoxin or digitoxin at the end of the
adsorption period, in parallel with uninfected cells. Cells were fixed at 24 hpi for
immunodetection of hexon (green) as described in Materials and Methods. Nuclei were
stained with DAPI (blue). The image shown is representative of cells treated with digitoxin.
Appendices
91
0200400600800
1000120014001600
Fo
ld I
ncr
ease
(Rel
ativ
e to
Unin
fect
ed)
Hours post-infection (hpi)
E4orf6/7
0
500
1000
1500
2000
2500
3000F
old
Incr
ease
(Rel
ativ
e to
Unin
fect
ed)
Hours post-infection (hpi)
pTP
0
100
200
300
400
500
600
Un
infe
cted 1.5 3 4 5 6 7 8 9
13
19
21
min
us
RT
NT
C
Fold
Incr
ease
(Rel
ativ
e to
Un
infe
cted
)
Hours post-infection (hpi)
E1B-19K
Appendix 1.2 Time course of adenovirus RNA expression. A549 cells were infected with
HAdV-C5 and RNA was harvested at the following times post infection: 1.5hpi, every hour from
3-9hpi, 13, 19 and 21hpi. RNA was reverse transcribed, and cDNA was used for qPCR analysis
using primers to E1B, E2 and E4 transcripts. Shown are the graphs for pTP, E4orf6/7 and E1B-
19K as representative reactions. Peak expression was seen to be at 19-21hpi; these time points
were chosen for comparison of untreated and treated cells. No RT refers to a negative control
without reverse transcriptase added to the RT reaction; NTC refers to a negative control without
cDNA in the qPCR reaction.
92
Appendix 1.3 Representation of uninfected and adenovirus- infected nuclei visualized by
transmission electron microscopy. HEK293 cells were infected with HAdV-41. Cell pellets
were fixed, processed, stained and imaged using transmission electron microscopy. Nu:
nucleolus. Asterisk (*) indicates an infected cell. Image taken by Thomas Leung.