REGULATION OF MRNA TRANSLATION BY HUMAN CYTOMEGALOVIRUS PTRS1
Heather Ashley Vincent
A dissertation submitted to the faculty at the University of North Carolina at Chapel Hill in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Microbiology and Immunology in the School of Medicine.
Chapel Hill 2017
Approved by:
Nathaniel Moorman
Mark Heise
Marty Ferris
Stanley Lemon
Bill Marzluff
Cary Moody
© 2017 Heather Ashley Vincent ALL RIGHTS RESERVED
ii ABSTRACT
Heather Ashley Vincent: Regulation of mRNA Translation by Human Cytomegalovirus pTRS1 (Under the direction of Nathaniel Moorman)
Human cytomegalovirus (HCMV) is a major public health burden. Acute infection during pregnancy can lead to congenital birth defects, and reactivation of a latent infection in immune compromised individuals can cause significant morbidity and mortality. HCMV does not encode its own ribosomes, and is therefore completely reliant on the host translation machinery for viral protein synthesis. HCMV also does not induce host translational shutoff upon infection, thus viral mRNAs must compete with cellular mRNAs to efficiently translate viral proteins. The HCMV protein TRS1 (pTRS1) plays an integral role in translation regulation during HCMV replication by antagonizing the antiviral kinase PKR. Activated PKR phosphorylates eIF2α, which causes an overall inhibition of protein synthesis that inhibits HCMV replication. pTRS1 also increases overall levels of protein synthesis and enhances the translation of reporter mRNAs in a
PKR-independent manner, showing that pTRS1 regulates mRNA translation through multiple mechanisms. In this dissertation I sought to define the mechanisms used by pTRS1 to stimulate translation. In chapter 1 I show that pTRS1 inhibits PKR activation by binding to PKR and inhibiting PKR kinase activity. pTRS1 binds PKR residues that are conserved across eIF2α kinases, suggesting that pTRS1 can antagonize multiple eIF2α kinases. In chapter 2 I show that pTRS1 stimulates cap-independent translation.
iii pTRS1 enhances the activity of both host and viral internal ribosome entry sites (IRESs) and stimulates translation of a circular mRNA reporter. These pTRS1 functions were independent of its ability to antagonize PKR, but dependent on its ability to bind double- stranded RNA. To understand how pTRS1 stimulates translation, in chapter 4 I identify ribosome-associated, cellular proteins that bind pTRS1. I found that pTRS1 interacts with active protein phosphatase 1 (PP1) catalytic subunits. Rather than affect PP1 catalytic activity, pTRS1 changes the complement of proteins that interact with the PP1 alpha catalytic subunit, possibly to regulate PP1 substrate specificity. Together these data further characterize the mechanisms used by pTRS1 to regulate mRNA translation and reveal how pTRS1 may contribute to the efficient translation of viral mRNAs during
HCMV infection.
iv To my little man. Your smiling face brightens even the roughest days.
v ACKNOWLEDGEMENTS
I first want to thank my family for their continued love and support. You have supported me in every step of my journey to becoming a scientist and I am eternally grateful for that.
I also want to thank my friends for keeping me sane during my time in graduate school. This was not an easy process, but you were always there for me. I could not have made it without you.
I cannot express how grateful I am to have worked with such a wonderful group of people in the Moorman lab. You guys are amazing, and became like family. Thank you for your support and friendship throughout the years.
Lastly, none of this would have been possible without the incredible mentorship I received from Nathaniel Moorman. You taught me how to think, and very importantly, to always include proper controls. I am so grateful for the time and effort you put into training me. I am the scientist I am today because of you. Thank you.
vi TABLE OF CONTENTS
LIST OF TABLES ...... xi
LIST OF FIGURES ...... xii
LIST OF ABBREVIATIONS AND SYMBOLS ...... xiv
CHAPTER 1: INTRODUCTION ...... 1
Herpesvirus replication and pathogenesis ...... 1
Human cytomegalovirus pathogenesis ...... 5
Human cytomegalovirus replication ...... 8
Regulation of mRNA translation during HCMV infection ...... 9
Overview of translation regulation ...... 10
Overview of translation regulation during HCMV infection ...... 18
The role of the eIF4F complex in HCMV replication ...... 19
Alternate modes of translation initiation during infection ...... 21
HCMV and eIF2α kinases ...... 23
Role for HCMV pTRS1 in translation regulation ...... 27
Outstanding questions in the field ...... 31
Figure ...... 35
References ...... 37
CHAPTER 2: MECHANISM OF PKR INHIBITION BY HCMV PTRS1 ...... 66
Overview ...... 66
vii Introduction ...... 67
Materials and Methods ...... 70
Results ...... 77
Discussion ...... 85
Acknowledgements ...... 90
Figures ...... 91
Tables...... 98
References ...... 104
CHAPTER 3: HUMAN CYTOMEGALOVIRUS PTRS1 STIMULATES CAP- INDEPENDENT TRANSLATION ...... 111
Overview ...... 111
Introduction ...... 112
Materials and Methods ...... 115
Results ...... 118
Discussion ...... 123
Acknowledgements ...... 125
Figures ...... 126
Tables...... 131
References ...... 132
CHAPTER 4: HCMV PTRS1 INTERACTS WITH ACTIVE PP1 CATALYTIC SUBUNITS ...... 139
Overview ...... 139
Introduction ...... 140
Materials and Methods ...... 142
viii Results ...... 149
Discussion ...... 154
Figures ...... 158
Tables...... 164
References ...... 165
CHAPTER 5: DISCUSSION ...... 170
Summary of my findings ...... 170
Outstanding questions related to pTRS1 and PKR ...... 173
What are the dsRNA ligands that activate PKR during HCMV infection? ...... 173
Does pTRS1 prevent PKR binding to eIF2α? ...... 174
Does pTRS1 antagonize multiple eIF2α kinases during HCMV infection? ...... 174
What drives eIF2α phosphorylation during the late stage of HCMV replication? ...... 175
What is the role of pIRS1 during infection? ...... 176
How does pTRS1 stimulate translation in a PKR- independent manner? ...... 178
Outstanding questions pertaining to translation regulation during HCMV infection ...... 184
What is the complement of full-length transcripts produced during HCMV replication? ...... 184
What are the RNAs packaged within the HCMV virion, and are they functional? ...... 186
Does mRNA stability regulate mRNA translation during HCMV infection? ...... 188
What is the role of 3’UTR binding proteins in HCMV mRNA translation regulation? ...... 191
ix Do mRNA modifications affect translation regulation during HCMV infection? ...... 192
How are mRNAs translated in the presence of eIF2α phosphorylation during HCMV infection? ...... 193
Conclusion ...... 197
References ...... 198
x LIST OF TABLES
Table 2.1 - Primer and gBlock Gene Fragment Sequences...... 98
Table 2.2 - pTRS1 and PKR Mutations and Functional Descriptions...... 101
Table 3.1 - Primer and gBlock Gene Fragment Sequences...... 131
Table 4.1 - Primer Sequences...... 164
xi LIST OF FIGURES
Figure 1 - Schematic of human cytomegalovirus (HCMV) manipulation of translation initiation and elongation...... 35
Figure 2.1 - Domains of pTRS1 necessary for PKR binding and inhibition...... 91
Figure 2.2 - PKR binding by full-length pTRS1 is necessary for inhibition of PKR activation...... 92
Figure 2.3 - pTRS1 does not prevent PKR dimerization...... 93
Figure 2.4 - pTRS1 binds and inhibits a constitutively active PKR kinase domain...... 94
Figure 2.5 - pTRS1 interacts with the PKR kinase domain to prevent PKR autophosphorylation...... 95
Figure 2.6 - pTRS1 inhibits PKR kinase activity...... 96
Figure 2.7 - pTRS1 binds HRI and limits its activation...... 97
Figure 3.1 - pTRS1 stimulates poliovirus IRES activity...... 126
Figure 3.2 - pTRS1 increases the activity of multiple viral IRESs...... 127
Figure 3.3 - pTRS1 stimulates the translation of a circular GFP mRNA...... 128
Figure 3.4 - pTRS1 stimulates cellular IRES activity...... 129
Figure 3.5 - Stimulation of IRES activity by pTRS1 is not dependent on PKR antagonism...... 129
Figure 3.6 - pTRS1 dsRNA binding is necessary to promote cap- independent translation...... 130
Figure 4.1 - pTRS1 interacts with specific ribosome-associated proteins...... 158
Figure 4.2 - Ribosome-associated proteins that specifically interact with pTRS1...... 159
Figure 4.3 - Validation of pTRS1-interacting proteins...... 160
Figure 4.4 - pTRS1 interacts with all three PP1 catalytic subunits...... 161
Figure 4.5 - pTRS1 does not affect PP1Cα phosphatase activity...... 162
xii Figure 4.6 - pTRS1 changes the complement of proteins that interact with PP1Cα....163
xiii LIST OF ABBREVIATIONS AND SYMBOLS
4EBP1 eukaryotic initiation factor 4E-binding protein 1
4SU 4-thiouridine aa amino acid
ABC ammonium bicarbonate
ActD actinomycin D
AKT protein kinase B
AMP adenosine monophosphate
AMPK adenosine monophosphate-regulated protein kinase
ATF6 activating transcription factor 6
ATP adenosine triphosphate
BioID proximity-dependent biotin identification
BiP binding immunoglobulin protein bp basepair
C celsius
Cas9 clustered regularly interspaced short palindromic repeats associated protein 9 cDNA complementary deoxyribonucleic acid
CHX cyclohexamide
Ci curie circRNA circular ribonucleic acid
CPEB cytoplasmic polyadenylation element binding partner
CRISPR clustered regularly interspaced short palindromic repeats
xiv CrPV cricket-paralysis virus
DDX3 DEAD-Box helicase 3
DDX5 DEAD-Box helicase 6
DHX9 DExH-Box helicase 9
DHX29 DExH-Box helicase 29
DNA deoxyribonucleic acid dsDNA double-stranded deoxyribonucleic acid dsRBD double-stranded ribonucleic acid binding domain dsRNA double-stranded ribonucleic acid
DTT dithiothreitol
E early
EBV Epstein-Barr virus
EDTA ethylenediaminetetraacetic acid eEF1 eukaryotic elongation factor 1 eEF2 eukaryotic elongation factor 2 eEF2K eukaryotic elongation factor 2 kinase eIF1 eukaryotic initiation factor 1 eIF1A eukaryotic initiation factor 1A eIF2 eukaryotic initiation factor 2 eIF2A eukaryotic initiation factor 2A eIF2α eukaryotic initiation factor 2 alpha subunit eIF2B eukaryotic initiation factor 2B eIF2Bα eukaryotic initiation factor 2B alpha subunit
xv eIF2D eukaryotic initiation factor 2D eIF3 eukaryotic initiation factor 3 eIF4A eukaryotic initiation factor 4A eIF4B eukaryotic initiation factor 4B eIF4E eukaryotic initiation factor 4E eIF4F eukaryotic initiation factor 4F eIF4G eukaryotic initiation factor 4G eIF5 eukaryotic initiation factor 5
ER endoplasmic reticulum eRF-1 eukaryotic release factor 1 eRF-3 eukaryotic release factor 3
EV empty vector
FLuc firefly luciferase
GADD34 growth arrest and deoxyribonucleic acid-damage inducible protein 34
GAP guanosine diphosphatase activating protein
GCN2 general control nonderepressible 2
GDP guanosine diphosphate
GEF guanine nucleotide exchange factor
GFP green fluorescent protein
GMP-PNP 5’guanylyl imidodiphosphate
GST glutathione-S-transferase
GTP guanosine triphosphate
xvi HCl hydrochloric acid
HCMV human cytomegalovirus
HCV hepatitis C virus
HFF human foreskin fibroblast
HHV human herpesvirus
HITS-CLIP high-throughput sequencing of ribonucleic acids isolated by crosslinking immunoprecipitation
HIV human immunodeficiency virus
HPV human papilloma virus
HRI heme-regulated inhibitor
HSV-1 human simplex virus-1
HSV-2 human simplex virus-2
IE immediate early
IFN interferon
IRE-1 inositol-requiring enzyme 1
IRES internal ribosome entry site
ISG interferon stimulate gene
ITAF internal ribosome entry site trans-activating factor kB kilobase
KD kinase domain
KO knockout
KSHV Kaposi’s sarcoma-associated herpesvirus l liter
L late
xvii La lupus autoantigen
M molar m6A N6-methyladenosine m7G 7-methyl guanosine
MCMV murine cytomegalovirus min minute miRNA micro ribonucleic acid ml milliliter mM millimolar
MNK mitogen-activated protein kinase
MOI multiplicity of infection mRNA messenger ribonucleic acid mTOR mammalian target of rapamycin kinase
NaCl sodium chloride
NADPH nicotinamide adenine dinucleotide phosphate ng nanogram
ORF open reading frame p70S6K 70 kilodalton ribosomal protein S6 kinase
PAA phosphonoacetic acid
PABP poly(A) binding protein
PAGE polyacrylamide gel electrophoresis
PAN poly(A)-nuclease
PBD protein kinase R binding domain
xviii PBS phosphate-buffered saline
PEI polyethylenimine
PERK protein kinase R-like endoplasmic reticulum kinase
PI3K phosphoinositide 3-kinase
PIC preinitiation complex
PKR protein kinase R
PP1 protein phosphatase 1
PP1Cα protein phosphatase 1 catalytic subunit alpha
PP1Cβ protein phosphatase 1 catalytic subunit beta
PP1Cγ protein phosphatase 1 catalytic subunit gamma
PP2A protein phosphatase 2A
PTB polypyrimidine tract binding protein pTRS1i pTRS1-inducible
RAPTOR regulatory-associated protein of mammalian target of rapamycin
RBD ribonucleic acid binding domain
RBP ribonucleic acid binding protein
RICTOR rapamycin-insensitive companion of mammalian target or rapamycin
RIP-Seq ribonucleic acid immunoprecipitation sequencing
RLuc renilla luciferase
RNA ribonucleic acid
RNA-Seq ribonucleic acid sequencing
ROS reactive oxygen species
RT room temperature
xix RT-qPCR real-time quantitative polymerase chain reaction
SDS sodium dodecyl sulfate
SHAPE-MaP selective 2’-hydroxyl acylation analyzed by primer extension and mutational profiling shRNA short hairpin ribonucleic acid siRNA short interfering ribonucleic acid ssDNA single-stranded deoxyribonucleic acid
TCA trichloroacetic acid
TOR target of rapamycin tRNA transfer ribonucleic acid
Met tRNAi initiator methionyl-transfer ribonucleic acid tRNAMet methionyl-transfer ribonucleic acid
TSC tuberous sclerosis complex
TSC1 tuberous sclerosis complex 1
TSC2 tuberous sclerosis complex 2
TSS transcription start site
UL unique long uORF upstream open reading frame
UPR unfolded protein response
US unique short
UTR untranslated region
VAI adenovirus virus-associated ribonucleic acid-I
VV vaccinia virus
VZV varicella zoster virus
xx YTHDC2 YT521-B homology domain containing 2
YTHDF2 YT521-B homology domain family protein 2
YTHDF3 YT521-B homology domain family protein 3
α alpha
β beta
γ gamma
δ delta
ε epsilon
Δ delta
µ micro
xxi CHAPTER 1: INTRODUCTION1
Herpesvirus replication and pathogenesis
Herpesviruses are double-stranded DNA viruses that infect and cause disease in many animal species, including humans (1-3). Several different herpesviruses infect a majority of the human population (4-6). Thus, nearly every person is infected with at least one herpesvirus. The host immune response initially controls acute infection, and the virus then enters a latent state and is harbored by the host for the remainder of their life. The majority of herpes infections are latent and therefore asymptomatic (6-8).
However, lytic herpesvirus replication after primary infection, or reactivation of a latent infection, in an immune-compromised or immune naïve individual can cause significant morbidity and mortality (7, 9, 10). There are no vaccines for herpesviruses, and current antiviral drugs have toxic side effects (11). Therefore, herpesvirus infections are a significant global public health burden.
There are nine types of human herpesviruses (HHVs) (1). HHV-1 and HHV-2 are both herpes simplex viruses (HSV-1 and HSV-2, respectively). Both HSV-1 and HSV-2 are transmitted through contact with infected skin or via fluid exchange, and establish lifelong latency in the neural ganglia. Reactivation of an HSV-1 or HSV-2 infection can lead to cold sores or genital warts depending on the site of infection (4, 7). Primary infection with HHV-3, or varicella zoster virus (VZV), causes chickenpox, while
1 Part of this chapter previously appeared as an article in Virology. The original citation is as follows: Vincent HA, Ziehr B, Moorman NJ. “Human Cytomegalovirus Strategies to Maintain and Promote mRNA Translation.” Virology. 2016 Apr 13;8(4):97.
1 reactivation of latent VZV from trigeminal or basal root ganglia leads to shingles (12).
Epstein-Barr virus (EBV), or HHV-4, infects a majority of the population. Nearly 90% of adults in the United States have been infected with EBV, and primary infection during adolescence can lead to infectious mononucleosis. Reactivation of latent EBV infection from B cells in immune-compromised individuals is associated with several cancers including Hodgkin’s lymphoma, Burkitt’s lymphoma, and nasopharyngeal carcinoma (6,
13). HHV-5, or human cytomegalovirus (HCMV), also infects a majority of the population. The virus establishes latency in hematopoietic progenitor cells, however
HCMV can infect many different cell types during a lytic infection. Thus, reactivation of
HCMV in immune-compromised individuals can lead to disease in multiple organ systems, and even result in death (8, 14). In addition, lytic HCMV infection can be spread from mother to child during pregnancy causing congenital birth defects (10). The roseolovirus genus contains three herpesviruses, HHV-6A, HHV-6B, and HHV-7, which infect nearly 100% of the human population. Infection with HHV-6A, HHV-6B or HHV-7 infection as a child can cause roseola rash and fever, and lead to transplant complications in adults (15). While initial infection with Kaposi’s sarcoma-associated herpesvirus (KSHV or HHV-8) is typically asymptomatic, reactivation of KSHV is the causative agent of Kaposi’s sarcoma, multicentric Castleman’s disease, and primary effusion lymphoma (9). Because herpesvirus infection is ubiquitous, the disease burden caused by human herpesviruses is staggering, and highlights the need for development of herpesvirus vaccines and more efficient therapeutics.
The nine herpesviruses are further classified into three subfamilies: alpha-, beta- and gammaherpesviruses. These classifications were originally based on the speed of
2 replication and range of cell types infected by each virus (2). However, herpesviruses are now classified based on the sequence and structure of their viral genome (16).
Alphaherpesvirus genomes contain a unique long (UL) and unique short (US) region that is flanked by short inverted terminal repeat regions. Betaherpesvirus genomes also contain both a UL and US region, but are flanked by long inverted terminal repeat regions. Gammaherpesvirus genomes contain tandem direct repeat regions at the terminal ends of the genome, and in the case of EBV, tandem direct repeat regions interspersed throughout the genome (1). Each subfamily contains conserved genes that are not present within the other two subfamilies. Thus viruses are classified into subfamilies based on these conserved features. Each virus also contains unique genes that distinguish it from other viruses within that subfamily (17).
All herpesviruses have a large, double-stranded DNA genome that is packaged in the viral nucleocapsid. A layer of protein and RNA, the tegument, surrounds the nucleocapsid, and together the nucleocapsid and tegument layer are packaged within a phospholipid bilayer envelope (3). Viral glycoproteins in the virion envelope mediate cell entry by binding receptors on the cell surface, and promoting fusion of the viral envelope and cell plasma membrane. Following fusion, the viral tegument and nucleocapsid are released into the cytosol. The nucleocapsid transits to the nuclear envelope, and the viral DNA is then released into the nucleus (17, 18).
After deposition of the viral DNA in the nucleus, the virus can enter into a phase of lytic replication. A large number viral proteins are expressed during lytic infection that blunt the host immune response, replicate the viral genome, and produce progeny virus.
Herpesvirus genes are transcribed in three classes: immediate early (IE), early (E), and
3 late (L). Immediate early genes are the first transcribed and encode viral proteins that block innate immune responses and transactivate expression of early genes (19, 20).
Early viral proteins are involved in replication of the viral genome, induction of metabolic changes necessary for efficient viral replication, and transactivation of late viral gene expression (21). Late viral proteins primarily function in virion assembly and release (22-
24). This coordinated expression of viral proteins allows herpesviruses to efficiently hijack multiple cellular pathways to ensure the production of progeny virus.
Following an initial lytic infection the virus establishes a latent infection within specific cell types. During latency, viral gene expression is very limited as compared to lytic replication. Only genes that encode viral proteins that antagonize the host immune response and maintain the virus as a chromatinized episome are expressed (6, 8, 25).
In addition, several herpesviruses express non-coding RNAs and miRNAs that blunt the innate immune response during latency (26-28). Repressive chromatin structure on IE gene promoters ensures that lE genes are not expressed, and many infections will remain latent for the remainder of the host’s lifespan. However conditions that remove the repressive chromatin structure from IE promoters, such as stress, can induce IE expression and reactivation of lytic replication (29, 30).
Although this general replication program is conserved, herpesviruses vary widely in their cellular tropism, which greatly impacts the specific disease caused by each virus. In cell culture, alphaherpesviruses infect a wide range of cell types (31). In vivo, however, HSV-1 or HSV-2 lytic replication occurs in epithelial cells at mucosal surfaces or areas of broken skin, and latency is established in trigeminal ganglion (7).
VZV initially infects epithelial cells in the upper respiratory mucosa, and then infects T
4 cells, which disseminate the virus to large areas of skin epithelium, causing chickenpox.
VZV establishes latency in dorsal root ganglia (32, 33). Reactivation of alphaherpesviruses leads to anterograde transport of virions to initial sites of infection, where lytic replication ensues (34).
The betaherpesviruses show a more limited range of cell tropism than alphaherpesviruses in vitro, though they infect many cell types in vivo (35, 36). Initial lytic HCMV infection is thought to occur within epithelial cells of the genital tract or upper respiratory tract. The infection then spreads to the epithelium of multiple organ systems including the salivary glands, liver and spleen, and latency is established within myeloid progenitor cells (14). HHV-6 and HHV-7 replicate in CD4+ lymphocytes, but can also infect epithelial cells. HHV-6 DNA has been found in lymph nodes, macrophages, the salivary gland and the central nervous system (37). HHV-6B is thought to establish latency in hematopoietic progenitor cells, though the exact cell types supporting HHV-
6A and HHV-7 latency are unknown (15).
Gammaherpesviruses have the most limited cell tropism (6). KSHV replicates in epithelial cells and B cells after primary infection, and establishes latency within B cells and monocytes (38). KSHV infection of endothelial cells is also the cause of Kaposi sarcoma (39). Similarly, EBV infects both epithelial cells and B cells and establishes latency in memory B cells (40).
Human cytomegalovirus pathogenesis
Human cytomegalovirus (HCMV) is a pervasive herpesvirus that infects nearly
100% of individuals in developing countries and >60% of individuals in developed
5 nations (5, 41). Primary infection generally occurs during childhood and is transmitted via the saliva and urine of infected individuals, most often toddlers. Primary infection in children is largely asymptomatic, but may lead to fever and malaise (42). In rare cases primary infection of healthy teenagers and adults can cause infectious mononucleosis, hepatitis, thrombosis, retinitis and colitis (43-47). Infection can be spread through close contact with infected bodily fluids (e.g. saliva, urine, breast milk, semen and cervical secretions) (48-51). When this is the case, epithelial cells in the upper respiratory mucosa or genital tract are thought to be the initial site of HCMV lytic replication.
Infiltrating neutrophils and monocytes are then infected and aid in the dissemination of the virus to vascular endothelial cells. Infection can then spread to multiple organ systems, including the kidney, salivary gland, liver, inner ear, and central nervous system (CNS) (41, 44). In immune competent individuals, lytic replication in tissues is controlled and resolved by virus-specific CD8+ T cells (52, 53). Latency is then established within bone marrow resident CD34+ myeloid progenitor cells (54-56).
Differentiation of latently infected myeloid progenitor cells into macrophages or dendritic cells leads to reactivation of lytic replication, likely due to changes in chromatin structure that result in expression of immediate early genes (29, 57-59). Reactivation events seed further CD34+ progenitor cells with virus, and are cleared by the host immune system, often causing no disease. However, reactivation of HCMV lytic replication in immune- compromised individuals can lead to disease in multiple organ systems (8, 44).
Reactivation of lytic HCMV replication is a major issue for whole organ transplant recipients. Kidney and cardiac allografts are often rejected as a result of HCMV infection, with a higher rate in seronegative recipients who receive organs from
6 seropositive donors (60, 61). HCMV disease is most common, averaging 50%, in patients receiving bone marrow transplants. In addition, immunosuppressive therapy administered before transplant leads to reactivation in ~40% of patients (41, 62, 63).
This often leads to not only graft rejection, but also the disease states described above.
Another population highly affected by HCMV reactivation is cancer patients undergoing immunosuppressive chemotherapy and AIDS patients, as diminishment of protective T cells results in reactivation of lytic replication (64). Active HCMV replication in these populations can lead to retinitis, pneumonia, CNS damage, and even death (65, 66).
HCMV infection is also one of the leading causes of congenital birth defects. A primary infection, reinfection with an additional strain, or reactivation of a latent HCMV infection during pregnancy can lead to infection of the uterus and the fetus (10, 42).
Approximately 4-5 children out of every 1,000 live births worldwide have congenital
HCMV infection, and approximately 1 out of 5 infected children will develop disease as a result of the infection (67, 68). Hearing loss is the most common sequela of congenital
HCMV infection. However congenital infection can also cause vision loss, microcephaly, and CNS damage leading to severe developmental and learning disabilities (69, 70).
There is currently no vaccine to prevent HCMV infection. Further, the drug most commonly used to treat HCMV disease (acyclovir and its derivatives) is highly toxic, and drug-resistant HCMV strains are becoming increasingly common (11, 71). Thus more research defining the molecular mechanisms that regulate HCMV replication are needed to define novel targets for antiviral therapeutics and new strategies for the design of HCMV vaccines.
7 Human cytomegalovirus replication
HCMV cell entry is mediated by the viral glycoproteins gH, gB, and gL, which mediate fusion of the virion envelope with the envelope of the cell. Upon membrane fusion the viral nucleocapsid and tegument are released into the cytoplasm (72-75).
Several tegument proteins function to blunt initial immune responses (19, 76). The
HCMV proteins pTRS1 and pIRS1 are both packaged within the tegument and inhibit activation of the antiviral kinase PKR (77-79). Additionally, the viral protein UL83 inhibits activation of the innate DNA sensor IFI16 (80). The viral nucleocapsid then transits to the nuclear envelope and the HCMV genome is released into the nucleus.
The HCMV genome is the largest of the human herpesviruses (>230kb) and encodes multiple miRNAs, long non-coding RNAs (lncRNAs), and >750 ORFs (1, 81-
86). Viral mRNAs are transcribed from both strands of the viral genome, and throughout the course of lytic infection almost the entirety of the HCMV genome is transcribed (87).
Immediately after deposition of the viral genome in the nucleus, viral DNA is circularized and chromatinized, at which time the transcription of immediate early (IE) genes begins
(19). IE genes are defined as genes that are transcribed in the absence of de novo viral protein synthesis (i.e. in the presence of the protein synthesis inhibitor cycloheximide)
(88, 89). Expression of the IE genes IE1 and IE2 initiates lytic replication (90, 91). IE1 also inhibits cellular histone deacetylases (HDACs) and transactivates promoter elements, and IE2 acts as a general transcription factor and regulates cell cycle progression (92-96). Thus together IE1 and IE2 promote the expression of early viral genes.
8 Early (E) viral genes are primarily responsible for replication of the viral genome, and are defined as genes whose expression requires prior IE protein synthesis (21, 88).
Thus, immediate early viral proteins must be expressed for early viral genes to be transcribed. The viral DNA polymerase (UL54), ssDNA binding protein (UL57), and DNA polymerase processivity factor (UL44) are all expressed with early kinetics and are necessary for viral DNA replication (97-100). In addition viral proteins that prevent apoptosis (UL36), promote DNA replication (UL21.5), and regulate cell cycle progression (UL33) are also expressed with early kinetics (101-105).
Early viral proteins then transactivate the expression of late (L) viral genes (22).
DNA replication is necessary for true late gene expression, as transcription of these genes is inhibited in the presence of phosphoacetic acid (PAA), a viral DNA polymerase inhibitor (88). Expression of some late viral genes is dampened in the presence of PAA, but not prevented; these genes are characterized as “leaky late” genes (106). Late gene products (e.g. UL32 and UL52), in general, regulate the formation and egress of mature virions (107, 108).
Regulation of mRNA translation during HCMV infection
Although much is known about the transcriptional regulation of HCMV genes, relatively little is known about the translational regulation of viral mRNAs (19, 21, 22, 29,
88, 106). Viruses are completely reliant on the host translation machinery for the synthesis of viral proteins, since no virus encodes a ribosome. As a result, viral and host mRNAs must compete for access to ribosomes. This is especially relevant to HCMV, as unlike most viruses, HCMV does not induce any host transcriptional or translation shut-
9 off. Therefore, both host and viral mRNAs are translated throughout infection (86, 109-
113). Translation initiation, the rate-limiting step of translation, is heavily regulated (113-
115). Multiple antiviral signaling pathways inhibit translation initiation once activated, preventing the synthesis of viral proteins (116). Therefore HCMV must simultaneously counteract host defenses that inactivate the translation machinery upon sensing viral infection, and compete with host mRNAs for limited translation machinery. HCMV effectively antagonizes host defenses that limit viral protein expression (113, 117). In addition, HCMV manipulates multiple host signaling pathways to ensure the continued synthesis of both host and viral proteins needed for virus replication (80, 118, 119).
Thus, the interface between viral mRNAs and the host translation machinery serves as a critical determinant for successful HCMV infection.
This thesis focuses on understanding the regulation of translation by the HCMV protein TRS1 (pTRS1). Therefore a brief overview of the steps in translation, especially those most relevant to infection, is provided below.
Overview of translation regulation
Translation of mRNAs occurs in three steps: initiation, elongation and termination
(120, 121). Translation initiation begins with the assembly of the eukaryotic initiation factor 4F (eIF4F) translation initiation complex on the 7-methyl guanosine cap (m7G cap) at the 5’ end of the mRNA. ((122-124) reviewed in (121)). The bound eIF4F complex recruits the 43S preinitiation complex (PIC), consisting of the 40S ribosomal subunit, the ternary complex and multiple initiation factors, to form the 48S initiation complex. The 48S complex then scans the 5’ untranslated region (UTR) of the mRNA
10 until reaching the translation start codon, whereupon multiple initiation factors are released and the 60S ribosomal subunit is recruited (125). Joining of the 40S and 60S ribosomal subunits to form the 80S ribosome marks the end of the initiation phase, and the beginning of elongation. Elongation is regulated by the eukaryotic elongation factor
1 (eEF1), which promotes binding of aminoacyl-tRNAs to the A-site of the ribosome, and eukaryotic elongation factor 2 (eEF2), which facilitates the translocation of peptidyl- tRNA from the A-site to the P-site of the ribosome (126-128). Peptide elongation continues until a translation termination codon is encountered, whereupon translation ceases, the nascent peptide chain is displaced from the ribosome, and the ribosome disassembles (129, 130). The first step of translation, initiation, is the most regulated step ((131-133) reviewed in (134)) and is described in more detail below.
Formation of the eIF4F complex, which consists of eukaryotic initiation factors 4E
(eIF4E), 4G (eIF4G) and 4A (eIF4A), mediates the initiation of translation (121, 135). eIF4F assembly begins with binding of eIF4E to the m7G cap (122, 136). After binding the m7G cap, eIF4E recruits eIF4G, which acts as a scaffold protein that mediates the recruitment of the eIF4A RNA helicase, completing the assembly of the eIF4F complex
(137, 138). eIF4G also coordinates the interaction of the eIF4F complex with additional factors bound to the mRNA, such as the poly(A) binding protein (PABP), that enhance translation initiation (121, 135, 137-139). Prior to binding the eIF4F complex, multiple initiation factors associate with the 40S ribosomal subunit to prepare for translation initiation. Together, the 40S subunit and its associated factors are referred to as the 43S
PIC ((140, 141) reviewed in (115)). The 43S PIC contains the 40S ribosomal subunit, multiple initiation factors (e.g., eIF1, eIF1A, eIF3, and eIF5), and a ternary complex
11 composed of the eukaryotic initiation factor 2 (eIF2), guanosine triphosphate (GTP) and a charged methionyl-tRNA (tRNAMet) (142). eIF3 plays a critical role during 43S assembly, acting as a scaffold to recruit multiple initiation factors to the 40S ribosomal subunit (143). These factors affect ribosome scanning, eIF4A helicase activity, and fidelity of start codon recognition once associated with the eIF4F complex on the 5’UTR of the mRNA (143, 144).
Recruitment of the 43S PIC to the mRNA is mediated via an interaction between the eIF4G subunit of the eIF4F complex and eIF3 within the 43S PIC (145). Together, the 43S PIC, the eIF4F complex, and the bound mRNA constitute the 48S initiation complex. Once assembled, the 48S complex scans the 5’UTR of the mRNA until reaching the translation “start” codon (146). The eIF4A helicase, whose activity is stimulated by binding to eIF4B (147, 148), unwinds RNA structures that would otherwise impede 48S scanning (146, 149). Additional RNA helicases such as DDX3, DHX9 and
DHX29 also facilitate scanning through areas with high RNA secondary structure (150-
153). Scanning ceases upon recognition of a translation start site, which almost always consists of an AUG methionine codon in a favorable sequence context (i.e., Kozak sequence (154)). The combined actions of the eIF1, eIF1A and eIF5 initiation factors position the tRNAMet over the AUG codon, triggering hydrolysis of eIF2-bound GTP and release of a subset of initiation factors (125). eIF5 directs subsequent joining of the 40S and 60S ribosomal subunits to form a functional 80S ribosome, after which peptide elongation commences (155). Elongation continues until a translation termination or
“stop” codon is encountered, at which point eukaryotic release factor 1 (eRF-1) together
12 with eukaryotic release factor 3 (eRF-3) terminates elongation by displacing the nascent peptide from the ribosome (129, 130).
The target of rapamycin (TOR) kinase is conserved throughout eukaryotes, where it plays a critical role in modulating protein synthesis in response to the intracellular and extracellular environment ((156, 157) reviewed in (120)). In mammalian cells, mTOR (mammalian TOR) is found in two complexes: the mTORC1 and mTORC2 complexes (158, 159). While the two complexes share several subunits, each complex has unique defining components. mTORC1 contains the regulatory-associated protein of mTOR (raptor), while the presence of rapamycin-insensitive companion of mTOR
(rictor) defines the mTORC2 complex (160, 161). mTORC1 plays a central role in regulating translation initiation by controlling the assembly of the eIF4F complex (162).
In the hypo-phosphorylated state, the eIF4E-binding protein 1 (4EBP1) binds eIF4E and prevents binding to eIF4G, thereby limiting eIF4F formation (163). Phosphorylation of
4EBP1 by mTORC1 reduces the affinity of 4EBP1 for eIF4E, allowing efficient eIF4F complex formation and increasing overall levels of protein synthesis (Figure 1A) (164).
mTORC1 also promotes protein synthesis by phosphorylating and activating the
70 kDa ribosomal protein S6 kinase (p70S6K). Active p70S6K phosphorylates several factors involved in translation, such as eIF4B and eukaryotic elongation factor 2 kinase
(eEF2K) (165, 166). When phosphorylated, eIF4B binds eIF4A and increases its helicase activity, resulting in more efficient 48S scanning through 5’UTRs with significant secondary structure (167). Like mTORC1 itself, p70S6K also phosphorylates and deactivates a translation repressor, eEFK2 (166, 168). eEFK2 phosphorylates and inactivates eukaryotic elongation factor 2 (eEF2), which stimulates the incorporation of
13 amino acids into the growing peptide chain (169, 170). eEF2K phosphorylation of eEF2 prevents its association with the ribosome, thereby slowing the rate of elongation (126).
Thus, protein synthesis is induced through multiple mechanisms upon mTORC1 activation.
Protein synthesis is one of the most energy intensive cellular processes.
Therefore mTORC1 is unsurprisingly subject to both positive and negative regulation in response to environmental cues. This regulation converges on the tuberous sclerosis
(TSC) complex, which negatively regulates mTORC1 activity (171, 172). The heterodimeric TSC consists of the tuberous sclerosis 1 (TSC1) and tuberous sclerosis 2
(TSC2) proteins that together act as a GTPase activating protein (GAP) for the mTORC1 cofactor Rheb (173). mTORC1 is activated by association with Rheb-GTP and repressed by Rheb-GDP. Increased TSC GAP activity stimulates hydrolysis of GTP bound to Rheb, and therefore inhibits mTORC1 activity (174). Decreased nutrient availability leads to elevated AMP to ATP ratios that activate the AMP-regulated protein kinase (AMPK) (175). Active AMPK phosphorylates TSC2, stimulating TSC GAP activity and inhibiting mTORC1 activity (176). Conversely, when nutrients are plentiful, growth factor receptor signaling activates the phosphoinositide 3-kinase (PI3K), which phosphorylates and inhibits the TSC complex, thereby promoting mTORC1 activity and eIF4F assembly (177). Thus the TSC complex integrates upstream signaling pathways to modulate mTORC1 activity and protein synthesis to match the nutrient availability within the cellular environment.
Translation of some viral mRNAs does not require 40S ribosomal subunit scanning.
Instead, ribosomes are recruited to the site of translation initiation, often with limited or
14 no ribosome scanning, by specific RNA sequences/structures called internal ribosome entry sites (IRESs) ((178, 179) reviewed in (180)). Ribosome recruitment to IRESs often requires only a subset of translation initiation factors. For example, translation initiation from the poliovirus IRES is independent of the eIF4E cap binding protein, but requires eIF4A, eIF4G and the 43S PIC (181). On the other end of the spectrum, the cricket paralysis virus (CrPV) IRES requires only 40S and 60S ribosomal subunits to initiate translation (182, 183). Although these and other IRESs require differing factors to initiate translation, RNA secondary and tertiary structure is indispensable for the recruitment of ribosomal subunits to IRES elements (184).
In addition to initiation factors, IRES activity can be enhanced through binding of
IRES trans-activating factors, or ITAFs (185). Two cellular factors consistently function as ITAFs, the polypyrimidine tract binding protein (PTB) and the lupus autoantigen (La).
PTB is an RNA binding protein that directly interacts with RNA secondary structures to promote RNA folding and maintain IRES structures (186, 187). The La protein binds the
5’UTR of multiple viruses, including poliovirus, human immunodeficiency virus (HIV), hepatitis C virus (HCV), and influenza virus (188). Binding of La to IRES sequences stimulates recruitment of the 40S ribosomal subunit, possibly to promote correct start codon usage (189). Both proteins are critical for the activity of multiple viral IRESs
((190, 191) reviewed in (192)) and promote IRES driven translation during HCV, poliovirus and rhinovirus infection (193, 194).
The length of the 5’UTR and the extent of RNA secondary structure also affect mRNA translation efficiency. mRNAs that contain very short 5’UTRs (<20 nucleotides) are not efficiently translated, as the AUG start codon is too close to the cap structure,
15 and initiation is inhibited by eIF1 (146, 179, 195-197). Lengthening of an mRNA 5’UTR does not inhibit translation efficiency, but increases the likelihood that the UTR will contain RNA secondary structures that can impede 43S PIC scanning (149, 198). The
RNA helicases DHX29, DDX3 and eIF4A are crucial for the efficient translation of cellular mRNAs containing highly structured 5’UTRs (199-202). In addition to complex
RNA secondary structure, the presence of upstream ORFs (uORFs) in the 5’UTR can inhibit translation of the downstream protein coding region (203). Translation initiation on the uORF produces a short peptide and inhibits initiation at the main, downstream
ORF. When levels of ternary complex are limiting, such as when eIF2α is phosphorylated in response to stress (see below), leaky scanning through the uORF occurs and allows for translation initiation at the downstream ORF (204). Nearly 50% of cellular mRNAs contain uORFs, potentially allowing for complex regulation of translation of these mRNAs in response to changes in the cellular environment (205, 206).
Another focal point in the regulation of translation initiation is the phosphorylation of the alpha subunit of eukaryotic initiation factor 2 (eIF2α). eIF2α is part of the trimeric eIF2 complex, which together with GTP and tRNAMet, form the ternary complex that associates with the 43S PIC (207). Recognition of the AUG initiation codon by tRNAMet stimulates the hydrolysis of eIF2-associated GTP. The resulting release of free phosphate triggers eIF2-GDP release from the ribosome (208). Eukaryotic initiation factor 2B (eIF2B) then exchanges GDP for GTP in the eIF2 complex, allowing eIF2-
GTP to form a new ternary complex and participate in subsequent rounds of initiation
(209). Phosphorylation of eIF2α by one of four eIF2α kinases (see below) greatly increases the affinity of eIF2B for eIF2, preventing eIF2B release and GTP exchange
16 (210, 211). As eIF2α is present at a much higher concentration than eIF2B, even small increases in eIF2α phosphorylation can sequester essentially all of the available eIF2B, preventing the formation of new ternary complexes and leading to a dramatic (>80%) decrease in protein synthesis (212).
Similar to the mTOR signaling pathway, eIF2α kinases play a critical role in matching the cell’s protein synthesis capacity to the cellular environment. Four eIF2α kinases have been identified, each of which are activated in response to a distinct cellular stress (reviewed in (213)). The heme-regulated inhibitor (HRI) kinase phosphorylates eIF2α in response to high levels of reactive oxygen species, linking the rate of protein synthesis to the respiratory capacity of the cell (214). Accumulation of unfolded proteins within the endoplasmic reticulum (ER) activates the PKR-like endoplasmic reticulum kinase (PERK) (215), temporarily inhibiting protein synthesis to allow the cell to properly fold or degrade accumulated unfolded proteins. The eIF2α kinase general control nonderepressible 2 (GCN2) is activated by binding to uncharged tRNAs that accumulate during amino acid deprivation, directly linking amino acid availability to the rate of protein synthesis (216, 217). Particularly relevant to viral infections is protein kinase R (PKR), which is activated by binding to viral double stranded RNAs (dsRNAs) (218-221) and potently inhibits viral protein synthesis, and thus virus replication (211). As discussed above, eIF2α phosphorylation can enhance the translation of mRNAs containing uORFs in the 5’UTRs. For example, a uORF in the
5’UTR of the GADD34 mRNA allows for efficient GADD34 protein synthesis when eIF2α is phosphorylated (222). GADD34 binds and recruits protein phosphatase (PP1) to eIF2α to mediate eIF2α dephosphorylation. This regulation ensures that eIF2α
17 phosphorylation is reversed after stress is resolved, and inhibition of translation initiation is relieved (223).
The mechanisms used by HCMV to regulate mRNA translation during infection, and the signaling pathways that control translation, are discussed below.
Overview of translation regulation during HCMV infection
Unlike many viruses, HCMV does not limit host translation. Instead, overall levels of protein synthesis are maintained in HCMV-infected cells (114). As both host and viral mRNAs rely on the same pool of ribosomes for their translation, HCMV mRNAs must efficiently compete with host transcripts for access to the translation machinery to ensure synthesis of viral proteins. A recent ribosome profiling study showed both host and viral mRNAs are efficiently translated as infection progresses. However the overall translation efficiency of viral mRNAs exceeds that of host mRNAs late in infection (86,
109). This suggests that viral mRNAs compete more efficiently for cellular translation machinery than host mRNAs. Additionally, HCMV must circumvent antiviral defenses that otherwise limit protein synthesis upon infection. Infection activates both antiviral defenses and stress response pathways, yet viral protein synthesis and replication remain unaffected (see below). While the full spectrum of strategies employed by
HCMV to regulate host responses is beyond the scope of this introduction, examples of the strategies HCMV uses to maintain viral protein synthesis by counteracting host defenses and stress responses are described below.
18 The role of the eIF4F complex in HCMV replication
As discussed above, the eIF4F complex plays a critical role in translation initiation by recruiting the 43S PIC to the 5’ end of the mRNA. Recruitment of the eIF4F complex, particularly binding of the eIF4E subunit to the mRNA m7G cap, is thought to be the rate-limiting step of translation initiation (224). Overall levels of eIF4F often directly correlate with the total level of protein synthesis within the cell. The sustained eIF4F-dependent translation of host mRNAs during HCMV infection suggests that the virus manipulates cellular signaling pathways to maintain eIF4F activity (Figure 1A).
Consistent with this idea, Walsh et al. found that HCMV infection increases the abundance of eIF4F subunits and promotes eIF4F complex formation (114). In addition, elevated levels of PABP stimulate eIF4F formation during infection (225). The increase in eIF4F abundance is important for virus replication, as disrupting or inhibiting the eIF4F complex early during infection profoundly limits virus replication (113, 119, 226).
Similarly, an eIF4A helicase inhibitor suppresses HCMV replication in vitro when added at the time of infection (113). This reflects the necessity for eIF4F-dependent translation of host mRNAs during infection, as depletion of several host proteins that require eIF4F for their expression reduces the production of progeny virus (227).
In addition to increasing the abundance of eIF4F subunits, HCMV activates signaling pathways that stimulate eIF4F complex formation. mTORC1 activity is increased in HCMV infected cells, promoting eIF4F formation by phosphorylating and inactivating the 4EBP family of translational repressors (Figure 1A) (114, 226, 228).
During infection, mTORC1 activity is refractile to cellular stresses such as AMPK activation that typically decrease its activity (229-231). In fact HCMV paradoxically
19 requires increased activation of both AMPK and mTOR during infection for efficient replication (118). Thus HCMV uncouples mTORC1 activity from negative regulatory cues to promote virus replication. The dichotomy of continued mTORC1 activity despite
AMPK activation is explained by the fact that the HCMV UL38 protein (pUL38) binds and inhibits the host TSC2 protein, thus preventing inhibition of mTORC1 by activated
AMPK (80). pUL38 also stimulates mTORC1 activity in a TSC2-independent manner, although the mechanism remains unclear (Figure 1A) (232). Thus HCMV pUL38 severs the connection between AMPK and mTOR signaling, allowing for eIF4F formation and maintained levels of protein synthesis.
HCMV infection also stimulates additional signaling pathways that potentially enhance translation during infection. The PI3K signaling pathway is stimulated during infection and increases mTORC1 activity (233, 234). Chemical inhibitors of PI3K limit
HCMV replication (233), suggesting that PI3K signaling could play a role in stimulating translation in infected cells. HCMV infection also activates the MNK kinases (114), which phosphorylate eIF4E (Figure 1A). Phosphorylation of eIF4E is suggested to increase the rate of translation through an unknown mechanism. Inhibitors of the MNK kinase reduce HCMV replication (114), suggesting that MNK-dependent eIF4E phosphorylation regulates protein synthesis in infected cells. Infection also increases the abundance of the critical translation elongation factor eEF2 in a UL38-dependent manner (Figure 1B). While the role of the above signaling changes in HCMV translation has not been demonstrated, their association with the control of translation in other contexts suggests a potential role in translation regulation during HCMV infection.
20 As suggested by the multiple mechanisms HCMV employs to induce and maintain mTORC1 activity, decreased mTOR activity or expression inhibits virus replication (113, 119, 226). Depletion of mTOR, rictor or raptor decreases HCMV replication, as do ATP-competitive inhibitors of mTOR kinase activity (228, 235).
However, mTOR inhibitors also prevent metabolic remodeling induced by HCMV during infection, thus the effects of the inhibitors are likely pleiotropic (236).
Interestingly, mTOR inhibitors have little effect on viral protein synthesis when added later in infection (119, 235). The association of viral mRNAs with polysomes is largely unaffected by treatment with Torin1, which inhibits mTOR activity and prevents the formation of the eIF4F complex. In contrast, the translation efficiency of most host mRNAs is greatly reduced in infected cells treated with Torin1. In addition, hippuristanol, an inhibitor of the RNA helicases eIF4AI and eIF4AIII, does not affect viral replication later in infection, nor does it inhibit IE1 and IE2 protein expression at immediate early times after infection. These data suggest that HCMV uses a unique set of factors (viral and/or host) to regulate translation initiation during infection and ensure that viral mRNAs remain efficiently translated when eIF4F availability is limited.
Alternate modes of translation initiation during infection
Many viruses use IRES elements to ensure translation of viral mRNAs under stress conditions that limit host protein synthesis. The best characterized IRES-like element identified to date in the HCMV genome is located adjacent to the UL138 open reading frame (ORF). The UL138 ORF is the most 3’ of four ORFs encoded on a polycistronic mRNA, suggesting that cap-mediated translation initiation of UL138 would
21 be exceedingly inefficient. The UL138 IRES-like element directs internal initiation on the polycistronic pUL138 transcript and allows for increased UL138 protein synthesis under conditions of cell stress (237). While the role of the UL138 IRES during infection is unknown, it likely plays a role in regulating HCMV latency, as pUL138 acts as a molecular switch that regulates virus reactivation (238). A recent study identified ~100
HCMV 5’UTRs that contain IRES activity (239). However, whether or not these IRES- like elements allow for non-canonical translation initiation during infection events remains to be determined.
HCMV also stimulates the translation of at least one IRES-containing cellular mRNA encoding a protein needed for virus replication. Cells induce a coordinated response to the accumulation of unfolded proteins in the ER, aptly named the unfolded protein response (UPR). While induction of the UPR generally suppresses protein synthesis, the translation of a subset of mRNAs involved in resolving cell stress is selectively increased. HCMV infection induces the UPR, but selectively modulates downstream signaling pathways to support virus replication (240, 241). Induction of the
UPR leads to increased levels of the ER chaperone BiP (binding immunoglobulin protein) in infected cells, which likely supports HCMV replication by increasing the protein folding capacity of the ER (242, 243). The BiP mRNA contains an IRES element, and HCMV stimulates BiP mRNA translation in part by increasing the expression of the
La protein, a known ITAF for the BiP IRES (244). Given the increase in La abundance during infection and the wide range of cellular and viral mRNAs that La interacts with,
HCMV infection may stimulate the translation of additional IRES-containing cellular mRNAs.
22 HCMV and eIF2α kinases
HCMV infection generates cellular stresses that are potent activators of eIF2α kinases. However very little eIF2α phosphorylation is observed in infected cells, and only during the late stage of the virus lytic cycle (245), suggesting that HCMV actively prevents eIF2α kinase activation. The mechanism(s) used by HCMV to inhibit or counteract activation of two eIF2α kinases, PKR and PERK, is well described (78, 117,
245-248). The potential role of the other eIF2α kinases, HRI and GCN2, during HCMV infection is less clear, however they could play beneficial or inhibitory roles in virus replication.
The antiviral kinase PKR is the best characterized eIF2α kinase during HCMV infection. PKR is activated upon binding to double-stranded RNAs (dsRNAs) generated during the early stage of HCMV infection (249). Binding to dsRNA induces PKR homodimerization and subsequent activating autophosphorylation. Activated PKR then binds and phosphorylates eIF2α, leading to a significant decrease in the translation of both cellular and viral mRNAs, and thus decreased virus replication. dsRNA ligands for
PKR accumulate during HCMV replication (250), likely due to transcription from overlapping regions of both strands of the viral genome. Yet PKR activation and eIF2α phosphorylation are not observed, suggesting HCMV actively prevents PKR activation.
Almost all viruses express factors that inhibit PKR activation or limit eIF2α phosphorylation in infected cells. Some viruses produce RNAs that bind PKR and prevent its recognition of dsRNA ligands, such as the adenovirus VAI RNA (251) and
Epstein-Barr virus (EBV) EBER RNAs (252). Other viruses encode proteins that inhibit
PKR activation or reverse eIF2α phosphorylation. The herpes simplex virus 1 (HSV-1)
23 US11 protein prevents PKR activation (250, 253), while the HSV-1 ICP34.5 protein recruits the host protein phosphatase 1 (PP1) to de-phosphorylate eIF2α in order to maintain sufficient levels of active ternary complex (254). Thus inhibition of PKR is a conserved strategy used by many viruses to ensure efficient viral protein synthesis and virus replication.
Using a screen to identify HCMV genes that rescue growth of a vaccinia virus mutant lacking its PKR antagonist E3L, the laboratory of Dr. Adam Geballe identified the
HCMV TRS1 and IRS1 proteins (pTRS1 and pIRS1, respectively) as potent PKR antagonists (79). Expression of either pTRS1 or pIRS1 is necessary for HCMV replication, as a recombinant virus lacking both proteins fails to replicate (249).
Replication of a ΔTRS1/ΔIRS1 double mutant is rescued through depletion of PKR mRNA or disruption the PKR gene expression (77, 78). Thus a major function of pTRS1/pIRS1 during infection is to prevent inhibition of viral protein synthesis by PKR
(Figure 1B). Despite pTRS1/pIRS1 inhibition of PKR, phosphorylated eIF2α accumulates in infected cells during the late stage of HCMV infection (240, 255).
The eIF2α kinase PERK also plays a critical role in HCMV replication. High levels of protein synthesis, as during the late phase of HCMV infection, can overwhelm the folding capacity of the endoplasmic reticulum (ER). The presence of unfolded proteins in the ER initiates a coordinated cellular response, called the unfolded protein response
(UPR). The UPR is activated when unfolded proteins accumulate in the ER, competing the ER chaperone BiP away from its normal binding partners, the ER sensors ATF6,
IRE-1 and PERK (256). Loss of BiP binding activates each sensor and initiates a series of events designed to re-establish cellular homeostasis. Dissociation from BiP also
24 activates PERK, which then phosphorylates eIF2α. This results in a global reduction in translation, preventing the synthesis of new proteins into the already overtaxed ER. In addition eIF2α phosphorylation enhances the translation of a specific subset of mRNAs that encode proteins involved in resolving ER stress (257). Thus eIF2α phosphorylation by PERK allows the cell to both prevent further accumulation of unfolded proteins and reduce their levels within the ER.
As mentioned above, HCMV manipulates many aspects of the UPR, including the activity of the eIF2α kinase PERK. PERK levels increase during infection, and
PERK expression is necessary for efficient HCMV replication (240). Depletion of PERK prevents the increase in lipid synthesis observed during infection, and inhibits viral growth (248). Despite activation of the UPR and increased PERK abundance, PERK remains inactive until late in infection. The factors responsible for the regulation of
PERK activity during HCMV infection are not known. However PERK activation is not responsible for eIF2α phosphorylation late in infection, as eIF2α phosphorylation is observed when PERK expression is knocked down (258). These data again demonstrate how HCMV manipulates complex signaling pathways to ensure successful virus replication.
HCMV-induced ER stress also increases the levels of reactive oxygen species
(ROS) and overall levels of oxidative stress, which activate the eIF2α kinase heme- regulated inhibitor kinase (HRI) (259). Infection also stimulates NADPH oxidase, the enzyme responsible for production of superoxide (231). As HCMV induces oxidative stress and limits eIF2α phosphorylation, it seems likely that the virus regulates HRI
25 activation or its ability to phosphorylate eIF2α. However the effect of HCMV-induced oxidative stress on HRI expression and activity has not been examined.
High levels of translation increase the levels of deacetylated, or uncharged, tRNAs in the cell. The eIF2α kinase general control nonderepressible 2 (GCN2) is activated by binding to uncharged tRNAs (259). Thus as levels of translation and uncharged tRNAs increase, so do the levels of eIF2α phosphorylation. The resulting inhibition of translation allows the cell to restore a sufficient pool of charged tRNAs to support normal levels of protein synthesis. The high levels of viral protein synthesis together with the ongoing synthesis of host proteins would be expected to decrease the available pool of charged tRNAs, resulting in GCN2 activation. To date the role of GCN2 during HCMV infection has not been examined, however it seems likely that HCMV regulates GCN2 signaling to allow for continued viral protein expression. The related murine cytomegalovirus (MCMV) causes more severe disease in mice lacking GCN2 compared to wild type mice (260), demonstrating an antiviral role for GCN2 in vivo.
Because HCMV induces multiple stresses that could activate GCN2, and GCN2 plays an antiviral role in MCMV infection, it is likely that HCMV has mechanisms in place to limit GCN2 activation.
While infection induces stress that should activate multiple eIF2α kinases, eIF2α is not phosphorylated until late in infection (261, 262). Interestingly, both host and viral mRNAs are translated efficiently at this time despite eIF2α phosphorylation (86, 109). It is unclear if eIF2α phosphorylation no longer restricts translation initiation late in infection, or if the extent of eIF2α phosphorylation is insufficient to suppress protein synthesis. Perhaps the increased abundance and activity of the host PP1 and PP2A
26 phosphatases during infection (263) in conjunction with viral inhibitors of eIF2α kinases allow for efficient translation in the presence of multiple eIF2α stresses. Regardless, continued protein synthesis in the face of significant cellular stress suggests that HCMV encodes viral proteins that limit eIF2α kinase activity, or actively limit eIF2α phosphorylation in response to virus-induced stress.
Role for HCMV pTRS1 in translation regulation
As stated above, HCMV has multiple mechanisms in place to antagonize antiviral signaling events that inhibit translation initiation to ensure the continued synthesis of viral proteins. Two viral proteins involved in this process, pTRS1 and pIRS1, antagonize activation of the antiviral kinase PKR (78, 117). This prevents phosphorylation of PKR’s substrate, eIF2α, to ensure continued translation initiation. Thus pTRS1 and pIRS1 are pivotal in the regulation of translation during HCMV infection.
HCMV pTRS1 and pIRS1 were first shown to enhance the transactivation of viral promoters by IE1 and IE2 in vitro (264-266). However, when the TRS1 gene was removed from HCMV (HCMVΔTRS1), viral genomes and mRNAs accumulated to similar levels as wild-type virus. A defect in the amount of cell-free virus released late in infection was observed when primary human foreskin fibroblasts (HFFs) were infected with HCMVΔTRS1 at a low multiplicity of infection (267). It was later shown that pTRS1 played a minor role in the proper formation of DNA-containing capsids (268).
Later studies by Geballe et. al. demonstrated a role for pTRS1 and pIRS1 in blocking antiviral signaling. Either protein alone was sufficient to rescue replication of a vaccinia virus mutant lacking its PKR antagonist, E3L (VVΔE3L). Expression of pTRS1
27 or pIRS1 restored overall levels of protein synthesis, inhibited phosphorylation of eIF2α, and prevented activation of an additional antiviral protein, RNaseL following VVΔE3L infection (79). E3L binds and sequesters dsRNA ligands to prevent activation of PKR, suggesting pTRS1 might prevent PKR activation through a similar mechanism (269).
The subsequent finding that pTRS1 and pIRS1 both contain a non-canonical RNA binding domain in their conserved amino terminus further supported this hypothesis. pTRS1 binds both poly(I:C), a dsRNA ligand that activates PKR, and dsRNA hairpins of varying size (20, 29, and 39bp) (270). Three amino acids in the pTRS1 RNA binding domain (R121, R124, K125) are necessary for pTRS1 binding to poly(I:C). A pTRS1 mutant with these residues mutated to alanines (pTRS1 triple) did not rescue replication of VVΔE3L, suggesting that in this system the ability of pTRS1 and pIRS1 to bind dsRNA is important to inhibit PKR activation. Although the affinity of pTRS1 for dsRNA is much lower than that of PKR, the molar ratio of pTRS1 and pIRS1 together is greater than PKR across infection, suggesting that together pTRS1 and pIRS1 could efficiently compete with PKR for dsRNA ligands (246).
An HCMV mutant lacking both TRS1 and IRS1 (ΔTRS1/ΔIRS1) does not replicate in human foreskin fibroblasts (HFFs). Robust eIF2α phosphorylation is seen early after infection with ΔTRS1/ΔIRS1, which correlates with reduced overall levels of protein synthesis (271). dsRNA accumulates in HCMV infected cells, leading to the assumption that PKR was activated in the absence of pIRS1 and pTRS1. This would then lead to eIF2α phosphorylation and reduced protein synthesis (249). However this was not shown, as the authors did not measure PKR activation or demonstrate that
PKR was required for eIF2α phosphorylation or inhibition of protein synthesis.
28 Subsequently we found that loss of both pIRS1 and pTRS1 during HCMV infection leads to activation of PKR and phosphorylation of eIF2α. We generated cells expressing an shRNA against IRS1 (HFF-shIRS1), and infected these cells with HCMV mutants lacking either the entire pTRS1 ORF (HCMVΔTRS1), or the pTRS1 PKR binding domain (HCMVΔPBD). The absence of pIRS1 in these cells allowed us to specifically test the role of pTRS1 in PKR inhibition during HCMV infection. Infection of HFF-shIRS1 with HCMVΔTRS1 caused robust PKR and eIF2α phosphorylation, reduced overall levels of protein synthesis, and decreased virus replication. Importantly each of these phenotypes could be rescued by depletion of PKR prior to infection. Identical results were found after infection with HCMVΔPBD, which importantly contains the pTRS1 dsRNA binding domain. These data suggest that the ability of pTRS1 to bind PKR, rather than dsRNA, is necessary for inhibition of PKR activation (77). Similar results were concurrently reported by another group (78); mutation of four amino acids in the pTRS1 PKR binding domain, that disrupt the interaction between pTRS1 and PKR, significantly attenuated HCMV replication in HFFs, even though the pTRS1 dsRNA remained intact. Further, mutant virus replication was rescued when PKR expression was ablated in using CRISPR/Cas9.
Interestingly, when our group infected IRS1 knockdown cells with HCMV
TRS1ΔPBD we found that stress granules accumulated during infection (77). Stress granules form when pre-initiation complexes stall on mRNAs in response to eIF2α phosphorylation (272). The stalled mRNAs are redistributed into protein-RNA aggregates called stress granules until cellular stress is resolved. Stress granules do not form during wild-type HCMV infection, suggesting that pTRS1, specifically through
29 an interaction in its carboxyl-terminus, prevents stress granule formation during HCMV infection. This is likely due to reduced eIF2α phosphorylation resulting from the inhibition of PKR. However, we also found that pTRS1 inhibits stress granule formation in response to an HRI agonist, arsenite, in PKR-deficient cells. This result demonstrates two things: (1) that pTRS1 inhibits eIF2α phosphorylation resulting from activation of multiple eIF2α kinases and (2) that it does so in a PKR-independent manner. Although the main function of pTRS1 during HCMV infection of HFFs appears to be PKR antagonism, the role for additional PKR-independent functions of pTRS1 in other cell types has yet to be determined.
One additional function for pTRS1 that we identified was the ability of pTRS1 to promote mRNA translation independent of PKR inhibition. We identified pTRS1 in a proteomics screen of m7G cap-binding proteins during HCMV infection. I found that pTRS1 directly interacts with m7G cap, independent of the presence of any other translation initiation factors or other viral proteins (273). pTRS1 also associates with
40S ribosomal subunits (273) and interacts with ribosomal subunit proteins (our unpublished data). Our group also showed that exogenous expression of pTRS1 in transfected cells increases overall levels of protein synthesis by 50%, and stimulates the translation of reporter mRNAs containing viral or cellular 5’UTRs. To determine if the ability of pTRS1 to stimulate translation was simply due to PKR inhibition, we repeated these experiments in cells where we disrupted PKR expression using CRISPR/Cas9 mutagenesis. We found that pTRS1 maintained the ability to increase protein synthesis and stimulate the translation of reporter mRNAs, although to lower levels that in wild type cells (273). These data show that while inhibition of PKR contributes to the
30 increase in translation, pTRS1 also has a PKR-independent mechanism to promote translation. These data suggest that pTRS1 may recruit ribosomal subunits directly to mRNAs to stimulate their translation. Additionally, pTRS1 may play a role in the regulation of viral mRNA translation during HCMV infection.
Outstanding questions in the field
The current data show that the primary function of pTRS1 during HCMV infection of HFFs, the cell type commonly used to study lytic HCMV replication, is to inhibit PKR activation. Inhibition of PKR by pTRS1/pIRS1 prevents phosphorylation of eIF2α and inhibition of viral protein synthesis. Additionally, the ability of pTRS1 to antagonize PKR is dependent on a direct interaction (274). However, the molecular mechanism by which pTRS1 prevents PKR activation remains unknown.
The PKR-independent role of pTRS1 in translation regulation during HCMV infection has not been examined. Our group has shown that translation of HCMV transcripts, but not cellular transcripts, becomes resistant to eIF4F disruption late in infection (113). This suggests that HCMV utilizes a non-canonical translation initiation complex to promote the translation of viral mRNAs late in infection. Could pTRS1 be responsible for this eIF4F-independent translation initiation? We have shown that pTRS1 preferentially stimulates the translation of reporter mRNAs containing viral
5’UTRs compared to reporter mRNAs containing cellular 5’UTRs (273). Thus it is possible that pTRS1 is promoting the recruitment of ribosomal subunits to viral mRNAs and ensuring their efficient translation under conditions of limiting translation initiation factors.
31 Another unresolved question is how does pTRS1 stimulate translation in a PKR- independent manner? In addition to binding PKR, pTRS1 also binds dsRNA. We have shown the pTRS1 RNA binding domain (amino acids 84-246) is dispensable for the inhibition of PKR activation, but it is necessary for the stimulation of translation of reporter mRNAs in transfected cells (273). The molecular mechanism(s) underlying this function remains unknown. pTRS1 may be functioning as a viral initiation factor that promotes the recruitment of ribosomes to target mRNAs. Alternatively pTRS1 may be antagonizing additional signaling pathways that inhibit translation. In any case, further work is needed to understand how pTRS1 stimulates translation in a PKR-independent manner.
In this thesis I explored the molecular mechanism by which pTRS1 antagonizes
PKR, and how pTRS1 stimulates translation in a PKR-independent manner. In Chapter
2 I showed that the ability of pTRS1 to bind dsRNA is dispensable for its ability to inhibit
PKR activation, in vitro. However, the ability to bind PKR is necessary. pTRS1 interacts with conserved residues within the PKR kinase domain that mediate the interaction of
PKR with eIF2α. pTRS1 binding to PKR mediates inhibition of PKR kinase activity, thus preventing PKR from undergoing activating autophosphorylation. In addition, pTRS1 interacts with HRI and limits the accumulation of eIF2α phosphorylation in response to arsenite treatment in PKR KO cells, suggesting that pTRS1 inhibits the activation of multiple eIF2α kinases.
In Chapter 3, I examined the ability of pTRS1 to stimulate cap-independent translation. I found that pTRS1 increases the activity of both the poliovirus IRES and
KSHV vFLIP IRES in a bicistronic reporter assay by promoting IRES-dependent
32 translation. To address technical caveats of bicistronic reporter assays, I also tested the ability of pTRS1 to stimulate translation of a circular mRNA reporter whose translation is driven by an IRES. This mRNA reporter contains neither a m7G cap nor a poly(A) tail, thus translation of the reporter mRNA is driven solely by the IRES element. I showed that pTRS1 stimulates cap-independent translation driven by both viral and cellular
IRES elements. The ability of pTRS1 to stimulate cap-independent translation is PKR- independent, but requires the pTRS1 dsRNA binding domain, suggesting that pTRS1 binds RNA secondary structures to promote the translation of specific mRNAs.
To understand how pTRS1 promotes translation independent of PKR inhibition, in Chapter 4 I identified ribosome-associated proteins that interact with pTRS1 using the
BioID approach. BioID works by generating a fusion protein consisting of your gene of interest fused to a promiscuous biotin ligase (BirA) domain. Proteins that interact with the BioID fusion protein are covalently biotinylated, and can be recovered with streptavidin affinity chromatography and identified by mass spectrometry. Using a pTRS1-BioID fusion protein, I identified proteins that specifically interact with pTRS1 and also associate with ribosomes. 114 binding partners were consistently identified in the screen, and multiple interactions were confirmed both in transfection and during
HCMV infection. I found that pTRS1 interacts with all isoforms of protein phosphatase 1
(PP1) catalytic subunits in transfected cells, and with the PP1β catalytic subunit during infection. pTRS1 immune complexes isolated from infected cells have associated phosphatase activity, and pTRS1 does not change the activity of PP1 catalytic subunits.
Additionally, we showed that pTRS1 changes the complement of proteins that interact with PP1. Thus pTRS1 does not inhibit PP1 activity, but likely targets PP1 to specific
33 substrates. As phosphorylated eIF2α is a PP1 target, this data suggests that pTRS1 may limit eIF2α phosphorylation through the recruitment of PP1 to eIF2α.
Lastly, in the Discussion (Chapter 5) I describe how my findings have advanced our understanding of the role of pTRS1 in regulating mRNA translation, and what remains to be done to understand how pTRS1 functions and how those functions affect
HCMV replication.
34 FIGURE
Figure 1. Schematic of human cytomegalovirus (HCMV) manipulation of translation initiation and elongation. (A) HCMV stimulates eIF4F formation and activity through multiple mechanisms. Infection increases levels of eIF4F components
(eIF4E, eIF4G, eIF4A) and poly(A) binding protein (PABP). HCMV also promotes eIF4F assembly by activating mTORC1, which phosphorylates and inhibits the eIF4F
35 antagonist 4EBP1. The HCMV UL38 protein prevents inactivation of mTORC1 by inhibiting the tuberous sclerosis complex (TSC). pUL38 also stimulates mTORC1 activity through a TSC2-independent mechanism. mTORC1 activates p70S6K, which phosphorylates eIF4B to increase eIF4A helicase activity. HCMV infection activates the
PI3K pathway to promote mTORC1 activation, and may also regulate translation through activation of MNK kinases. (B) Translation initiation and elongation are maintained during HCMV infection. Inhibition of eIF2α phosphorylation ensures the regeneration of ternary complexes and continued rounds of translation initiation. Levels of eEF2 increase during infection through a pUL38-dependent mechanism and may promote translation during HCMV infection.
36 REFERENCES
1. Davison AJ. 2007. Comparative analysis of the genomes. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
2. Davison AJ. 2007. Overview of classification. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
3. Liu F, Zhou ZH. 2007. Comparative virion structures of human herpesviruses. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
4. Wald A, Corey L. 2007. Persistence in the population: epidemiology, transmission. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
5. Boppana SB, Fowler KB. 2007. Persistence in the population: epidemiology and transmisson. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
6. Ambinder RF, Cesarman E. 2007. Clinical and pathological aspects of EBV and KSHV infection. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
7. Whitley R, Kimberlin DW, Prober CG. 2007. Pathogenesis and disease. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
8. Jarvis MA, Nelson JA. 2007. Molecular basis of persistence and latency. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R,
37 Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
9. Martin JN. 2007. The epidemiology of KSHV and its association with malignant disease. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
10. Pereira L, Maidji E, Fisher SJ, McDonagh S, Tabata T. 2007. HCMV persistence in the population: potential transplacental transmission. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
11. Kimberlin DW, Whitley RJ. 2007. Antiviral therapy of HSV-1 and -2. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
12. Cohen JI. 2007. VZV: molecular basis of persistence (latency and reactivation). In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
13. Raab-Traub N. 2007. EBV-induced oncogenesis. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
14. Britt W. 2007. Virus entry into host, establishment of infection, spread in host, mechanisms of tissue damage. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
15. Mori Y, Yamanishi K. 2007. HHV-6A, 6B, and 7: pathogenesis, host response, and clinical disease. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
38 16. Anonymous. Herpesviridae - dsDNA Viruses - dsDNA Viruses (2011) - International Committee on Taxonomy of Viruses (ICTV). Accessed
17. Mocarski Jr. ES. 2007. Comparative analysis of herpesvirus-common proteins. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
18. Chandran B, Hutt-Fletcher L. 2007. Gammaherpesviruses entry and early events during infection. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
19. Stinski MF, Meier JL. 2007. Immediate–early viral gene regulation and function. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
20. Kristie TM. 2007. Early events pre-initiation of alphaherpes viral gene expression. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
21. White EA, Spector DH. 2007. Early viral gene expression and function. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
22. Anders DG, Kerry JA, Pari GS. 2007. DNA synthesis and late viral gene expression. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
23. Kenney SC. 2007. Reactivation and lytic replication of EBV. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
24. Sandri-Goldin RM. 2007. Initiation of transcription and RNA synthesis, processing and transport in HSV and VZV infected cells. In Arvin A, Campadelli- Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed),
39 Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
25. Preston CM, Efstathiou S. 2007. Molecular basis of HSV latency and reactivation. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
26. O'Connor CM, Vanicek J, Murphy EA. 2014. Host MicroRNA Regulation of Human Cytomegalovirus Immediate Early Protein Translation Promotes Viral Latency. Journal of Virology 88:5524-5532.
27. Plaisance-Bonstaff K, Choi HS, Beals T, Krueger BJ, Boss IW, Gay LA, Haecker I, Hu J, Renne R. 2014. KSHV miRNAs Decrease Expression of Lytic Genes in Latently Infected PEL and Endothelial Cells by Targeting Host Transcription Factors. Viruses 6:4005-4023.
28. Liu W, Ding C. 2017. Roles of LncRNAs in Viral Infections. Frontiers in Cellular and Infection Microbiology 7.
29. Sinclair J. 2010. Chromatin structure regulates human cytomegalovirus gene expression during latency, reactivation and lytic infection. Biochimica Et Biophysica Acta 1799:286-295.
30. Nicoll MP, Proença JT, Efstathiou S. 2012. The molecular basis of herpes simplex virus latency. Fems Microbiology Reviews 36:684-705.
31. Karasneh GA, Shukla D. 2011. Herpes simplex virus infects most cell types in vitro: clues to its success. Virology Journal 8:481.
32. Zerboni L, Sen N, Oliver SL, Arvin AM. 2014. Molecular mechanisms of varicella zoster virus pathogenesis. Nature reviews Microbiology 12:197-210.
33. Moffat J, Ku C-C, Zerboni L, Sommer M, Arvin A. 2007. VZV: pathogenesis and the disease consequences of primary infection. In Arvin A, Campadelli- Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
40 34. Campadelli-Fiume G. 2007. The egress of alphaherpesviruses from the cell. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
35. Jarvis MA, Nelson JA. 2007. Human Cytomegalovirus Tropism for Endothelial Cells: Not All Endothelial Cells Are Created Equal. Journal of Virology 81:2095- 2101.
36. Sinzger C, Digel M, Jahn G. 2008. Cytomegalovirus Cell Tropism, p 63-83, Human Cytomegalovirus. Springer, Berlin, Heidelberg.
37. Kurata T, Iwasaki T, Sata T, Wakabayashi T, Yamaguchi K, Okuno T, Yamanishi K, Takei Y. 1990. Viral pathology of human herpesvirus 6 infection. Advances in Experimental Medicine and Biology 278:39-47.
38. Chakraborty S, Veettil MV, Chandran B. 2012. Kaposi’s Sarcoma Associated Herpesvirus Entry into Target Cells. Frontiers in Microbiology 3.
39. Boshoff C, Schulz TF, Kennedy MM, Graham AK, Fisher C, Thomas A, McGee JO, Weiss RA, O'Leary JJ. 1995. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nature Medicine 1:1274-1278.
40. Thorley-Lawson DA, Hawkins JB, Tracy SI, Shapiro M. 2013. The Pathogenesis of Epstein-Barr Virus Persistent Infection. Current opinion in virology 3:227-232.
41. Landolfo S, Gariglio M, Gribaudo G, Lembo D. 2003. The human cytomegalovirus. Pharmacology & Therapeutics 98:269-297.
42. Stagno S, Pass RF, Dworsky ME, Henderson RE, Moore EG, Walton PD, Alford CA. 1982. Congenital Cytomegalovirus Infection. New England Journal of Medicine 306:945-949.
43. Nogalski MT, Collins-McMillen D, Yurochko AD. 2014. Overview of Human Cytomegalovirus Pathogenesis, p 15-28, Human Cytomegaloviruses. Humana Press, Totowa, NJ.
41 44. Britt W. 2008. Manifestations of human cytomegalovirus infection: proposed mechanisms of acute and chronic disease. Current Topics in Microbiology and Immunology 325:417-470.
45. Navalpotro D, Gimeno C, Navarro D. 2006. PCR detection of viral DNA in serum as an ancillary analysis for the diagnosis of acute mononucleosis-like syndrome due to human cytomegalovirus (HCMV) in immunocompetent patients. Journal of Clinical Virology: The Official Publication of the Pan American Society for Clinical Virology 35:193-196.
46. Streblow DN, Dumortier J, Moses AV, Orloff SL, Nelson JA. 2008. Mechanisms of cytomegalovirus-accelerated vascular disease: induction of paracrine factors that promote angiogenesis and wound healing. Current Topics in Microbiology and Immunology 325:397-415.
47. Jacobi CA, Riessen R, Schumacher U, Autenrieth IB, Jahn G, Gregor M, Raible A, Hamprecht K. 2010. Life-threatening pneumonia caused by human cytomegalovirus and Mycoplasma pneumoniae coinfection in a young, immunocompetent patient. Journal of Medical Microbiology 59:980-983.
48. Hayes K, Danks DM, Gibas H, Jack I. 1972. Cytomegalovirus in human milk. The New England Journal of Medicine 287:177-178.
49. Reynolds DW, Stagno S, Hosty TS, Tiller M, Alford CA. 1973. Maternal cytomegalovirus excretion and perinatal infection. The New England Journal of Medicine 289:1-5.
50. Drew WL, Mintz L, Miner RC, Sands M, Ketterer B. 1981. Prevalence of cytomegalovirus infection in homosexual men. The Journal of Infectious Diseases 143:188-192.
51. Stagno S, Cloud GA. 1994. Working parents: the impact of day care and breast- feeding on cytomegalovirus infections in offspring. Proceedings of the National Academy of Sciences of the United States of America 91:2384-2389.
52. Walter EA, Greenberg PD, Gilbert MJ, Finch RJ, Watanabe KS, Thomas ED, Riddell SR. 1995. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. The New England Journal of Medicine 333:1038-1044.
42 53. Boeckh M, Geballe AP. 2011. Cytomegalovirus: pathogen, paradigm, and puzzle. The Journal of Clinical Investigation 121:1673-1680.
54. Goodrum FD, Jordan CT, High K, Shenk T. 2002. Human cytomegalovirus gene expression during infection of primary hematopoietic progenitor cells: a model for latency. Proceedings of the National Academy of Sciences of the United States of America 99:16255-16260.
55. Maciejewski JP, Bruening EE, Donahue RE, Mocarski ES, Young NS, St Jeor SC. 1992. Infection of hematopoietic progenitor cells by human cytomegalovirus. Blood 80:170-178.
56. Minton EJ, Tysoe C, Sinclair JH, Sissons JG. 1994. Human cytomegalovirus infection of the monocyte/macrophage lineage in bone marrow. Journal of Virology 68:4017-4021.
57. Söderberg-Nauclér C, Streblow DN, Fish KN, Allan-Yorke J, Smith PP, Nelson JA. 2001. Reactivation of Latent Human Cytomegalovirus in CD14+ Monocytes Is Differentiation Dependent. Journal of Virology 75:7543-7554.
58. Söderberg-Nauclér C, Fish KN, Nelson JA. 1997. Reactivation of latent human cytomegalovirus by allogeneic stimulation of blood cells from healthy donors. Cell 91:119-126.
59. Reeves M, Sinclair J. 2008. Aspects of human cytomegalovirus latency and reactivation. Current Topics in Microbiology and Immunology 325:297-313.
60. Peterson PK, Balfour HH, Marker SC, Fryd DS, Howard RJ, Simmons RL. 1980. Cytomegalovirus Disease in Renal Allograft Recipients: A Prospective Study of the Clinical Features, Risk Factors and Impact on Renal Transplantation. Medicine 59:283.
61. Pollard RB. 1988. Cytomegalovirus infections in renal, heart, heart-lung and liver transplantation. The Pediatric Infectious Disease Journal 7:S97-102.
62. Rubie H, Attal M, Campardou AM, Gayet-Mengelle C, Payen C, Sanguignol F, Calot JP, Charlet JP, Robert A, Huguet F. 1993. Risk factors for cytomegalovirus infection in BMT recipients transfused exclusively with seronegative blood products. Bone Marrow Transplantation 11:209-214.
43 63. Winston DJ, Ho WG, Bartoni K, Du Mond C, Ebeling DF, Buhles WC, Champlin RE. 1993. Ganciclovir prophylaxis of cytomegalovirus infection and disease in allogeneic bone marrow transplant recipients. Results of a placebo- controlled, double-blind trial. Annals of Internal Medicine 118:179-184.
64. Webster A, Phillips AN, Lee CA, Janossy G, Kernoff PB, Griffiths PD. 1992. Cytomegalovirus (CMV) infection, CD4+ lymphocyte counts and the development of AIDS in HIV-1-infected haemophiliac patients. Clinical and Experimental Immunology 88:6-9.
65. Vinters HV, Kwok MK, Ho HW, Anders KH, Tomiyasu U, Wolfson WL, Robert F. 1989. Cytomegalovirus in the nervous system of patients with the acquired immune deficiency syndrome. Brain: A Journal of Neurology 112 ( Pt 1):245-268.
66. Sison RF, Holland GN, MacArthur LJ, Wheeler NC, Gottlieb MS. 1991. Cytomegalovirus Retinopathy as the Initial Manifestation of the Acquired Immunodeficiency Syndrome. American Journal of Ophthalmology 112:243-249.
67. Hyde TB, Schmid DS, Cannon MJ. 2010. Cytomegalovirus seroconversion rates and risk factors: implications for congenital CMV. Reviews in Medical Virology 20:311-326.
68. Boppana SB, Ross SA, Shimamura M, Palmer AL, Ahmed A, Michaels MG, Sánchez PJ, Bernstein DI, Tolan RWJ, Novak Z, Chowdhury N, Britt WJ, Fowler KB. 2011. Saliva Polymerase-Chain-Reaction Assay for Cytomegalovirus Screening in Newborns. New England Journal of Medicine 364:2111-2118.
69. Gaytant MA, Steegers EAP, Semmekrot BA, Merkus HMMW, Galama JMD. 2002. Congenital Cytomegalovirus Infection: Review of the Epidemiology and Outcome. Obstetrical & Gynecological Survey 57:245–256.
70. Britt WJ. 2017. Congenital Human Cytomegalovirus Infection and the Enigma of Maternal Immunity. Journal of Virology 91:e02392-02316.
71. Komatsu TE, Pikis A, Naeger LK, Harrington PR. 2014. Resistance of human cytomegalovirus to ganciclovir/valganciclovir: a comprehensive review of putative resistance pathways. Antiviral Research 101:12-25.
44 72. Compton T, Feire A. 2007. Early events in human cytomegalovirus infection. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
73. Compton T. 2004. Receptors and immune sensors: the complex entry path of human cytomegalovirus. Trends in Cell Biology 14:5-8.
74. Compton T, Nepomuceno RR, Nowlin DM. 1992. Human cytomegalovirus penetrates host cells by pH-independent fusion at the cell surface. Virology 191:387-395.
75. Wang X, Huong S-M, Chiu ML, Raab-Traub N, Huang E-S. 2003. Epidermal growth factor receptor is a cellular receptor for human cytomegalovirus. Nature 424:456-461.
76. Mocarski Jr. ES. 2007. Betaherpes viral genes and their functions. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
77. Ziehr B, Vincent HA, Moorman NJ. 2016. Human cytomegalovirus pTRS1 and pIRS1 antagonize PKR to facilitate virus replication. Journal of Virology doi:10.1128/JVI.02714-15.
78. Braggin JE, Child SJ, Geballe AP. 2016. Essential role of protein kinase R antagonism by TRS1 in human cytomegalovirus replication. Virology 489:75-85.
79. Child SJ, Hakki M, De Niro KL, Geballe AP. 2004. Evasion of cellular antiviral responses by human cytomegalovirus TRS1 and IRS1. Journal of Virology 78:197-205.
80. Moorman NJ, Cristea IM, Terhune SS, Rout MP, Chait BT, Shenk T. 2008. Human cytomegalovirus protein UL38 inhibits host cell stress responses by antagonizing the tuberous sclerosis protein complex. Cell Host & Microbe 3:253- 262.
81. Grey F, Nelson J. 2008. Identification and Function of Human Cytomegalovirus microRNAs. Journal of clinical virology : the official publication of the Pan American Society for Clinical Virology 41:186-191.
45 82. Dunn W, Trang P, Zhong Q, Yang E, Van Belle C, Liu F. 2005. Human cytomegalovirus expresses novel microRNAs during productive viral infection. Cellular Microbiology 7:1684-1695.
83. Pfeffer S, Sewer A, Lagos-Quintana M, Sheridan R, Sander C, Grässer FA, van Dyk LF, Ho CK, Shuman S, Chien M, Russo JJ, Ju J, Randall G, Lindenbach BD, Rice CM, Simon V, Ho DD, Zavolan M, Tuschl T. 2005. Identification of microRNAs of the herpesvirus family. Nature Methods 2:269-276.
84. Bergamini G, Reschke M, Battista MC, Boccuni MC, Campanini F, Ripalti A, Landini MP. 1998. The major open reading frame of the beta2.7 transcript of human cytomegalovirus: in vitro expression of a protein posttranscriptionally regulated by the 5' region. Journal of Virology 72:8425-8429.
85. Kulesza CA, Shenk T. 2004. Human cytomegalovirus 5-kilobase immediate- early RNA is a stable intron. Journal of Virology 78:13182-13189.
86. Stern-Ginossar N, Weisburd B, Michalski A, Le VTK, Hein MY, Huang S-X, Ma M, Shen B, Qian S-B, Hengel H, Mann M, Ingolia NT, Weissman JS. 2012. Decoding human cytomegalovirus. Science (New York, NY) 338:1088-1093.
87. Gatherer D, Seirafian S, Cunningham C, Holton M, Dargan DJ, Baluchova K, Hector RD, Galbraith J, Herzyk P, Wilkinson GWG, Davison AJ. 2011. High- resolution human cytomegalovirus transcriptome. Proceedings of the National Academy of Sciences 108:19755-19760.
88. DeMarchi JM, Schmidt CA, Kaplan AS. 1980. Patterns of transcription of human cytomegalovirus in permissively infected cells. Journal of Virology 35:277-286.
89. Chua CC, Carter TH, St Jeor S. 1981. Transcription of the human cytomegalovirus genome in productively infected cells. The Journal of General Virology 56:1-11.
90. Greaves RF, Mocarski ES. 1998. Defective growth correlates with reduced accumulation of a viral DNA replication protein after low-multiplicity infection by a human cytomegalovirus ie1 mutant. Journal of Virology 72:366-379.
91. Marchini A, Liu H, Zhu H. 2001. Human cytomegalovirus with IE-2 (UL122) deleted fails to express early lytic genes. Journal of Virology 75:1870-1878.
46 92. Nevels M, Paulus C, Shenk T. 2004. Human cytomegalovirus immediate-early 1 protein facilitates viral replication by antagonizing histone deacetylation. Proceedings of the National Academy of Sciences of the United States of America 101:17234-17239.
93. Kamata T, Tanaka S, Watanabe Y. 1978. Human cytomegalovirus-induced chromatin factors responsible for changes in template activity and structure of infected cell chromatin. Virology 90:197-208.
94. Hayhurst GP, Bryant LA, Caswell RC, Walker SM, Sinclair JH. 1995. CCAAT box-dependent activation of the TATA-less human DNA polymerase alpha promoter by the human cytomegalovirus 72-kilodalton major immediate-early protein. Journal of Virology 69:182-188.
95. Wiebusch L, Hagemeier C. 2001. The human cytomegalovirus immediate early 2 protein dissociates cellular DNA synthesis from cyclin-dependent kinase activation. The EMBO journal 20:1086-1098.
96. Paulus C, Nevels M. 2009. The Human Cytomegalovirus Major Immediate-Early Proteins as Antagonists of Intrinsic and Innate Antiviral Host Responses. Viruses 1:760-779.
97. Cihlar T, Fuller MD, Mulato AS, Cherrington JM. 1998. A point mutation in the human cytomegalovirus DNA polymerase gene selected in vitro by cidofovir confers a slow replication phenotype in cell culture. Virology 248:382-393.
98. Hwang E-s, Kim J, Jong H-s, Park J-w, Park C-g, Cha C-y. 2000. Characteristics of DNA-Binding Activity of Human Cytomegalovirus ppUL44. Microbiology and Immunology 44:827-832.
99. Weiland KL, Oien NL, Homa F, Wathen MW. 1994. Functional analysis of human cytomegalovirus polymerase accessory protein. Virus Research 34:191- 206.
100. Kiehl A, Huang L, Franchi D, Anders DG. 2003. Multiple 5' ends of human cytomegalovirus UL57 transcripts identify a complex, cycloheximide-resistant promoter region that activates oriLyt. Virology 314:410-422.
101. McCormick AL, Meiering CD, Smith GB, Mocarski ES. 2005. Mitochondrial cell death suppressors carried by human and murine cytomegalovirus confer
47 resistance to proteasome inhibitor-induced apoptosis. Journal of Virology 79:12205-12217.
102. Patterson CE, Shenk T. 1999. Human cytomegalovirus UL36 protein is dispensable for viral replication in cultured cells. Journal of Virology 73:7126- 7131.
103. Fehr AR, Yu D. 2011. Human cytomegalovirus early protein pUL21a promotes efficient viral DNA synthesis and the late accumulation of immediate-early transcripts. Journal of Virology 85:663-674.
104. Margulies BJ, Browne H, Gibson W. 1996. Identification of the human cytomegalovirus G protein-coupled receptor homologue encoded by UL33 in infected cells and enveloped virus particles. Virology 225:111-125.
105. Lares AP, Tu CC, Spencer JV. 2013. The human cytomegalovirus US27 gene product enhances cell proliferation and alters cellular gene expression. Virus Research 176:312-320.
106. Isomura H, Stinski MF. 2013. Coordination of late gene transcription of human cytomegalovirus with viral DNA synthesis: recombinant viruses as potential therapeutic vaccine candidates. Expert Opinion on Therapeutic Targets 17:157- 166.
107. Borst EM, Wagner K, Binz A, Sodeik B, Messerle M. 2008. The essential human cytomegalovirus gene UL52 is required for cleavage-packaging of the viral genome. Journal of Virology 82:2065-2078.
108. Dai X, Yu X, Gong H, Jiang X, Abenes G, Liu H, Shivakoti S, Britt WJ, Zhu H, Liu F, Zhou ZH. 2013. The smallest capsid protein mediates binding of the essential tegument protein pp150 to stabilize DNA-containing capsids in human cytomegalovirus. PLoS pathogens 9:e1003525.
109. Tirosh O, Cohen Y, Shitrit A, Shani O, Le-Trilling VTK, Trilling M, Friedlander G, Tanenbaum M, Stern-Ginossar N. 2015. The Transcription and Translation Landscapes during Human Cytomegalovirus Infection Reveal Novel Host-Pathogen Interactions. PLOS Pathogens 11:e1005288.
110. Stinski MF. 1977. Synthesis of proteins and glycoproteins in cells infected with human cytomegalovirus. Journal of Virology 23:751-767.
48 111. Tanaka S, Furukawa T, Plotkin SA. 1975. Human cytomegalovirus stimulates host cell RNA synthesis. Journal of Virology 15:297-304.
112. Walsh D, Perez C, Notary J, Mohr I. 2005. Regulation of the translation initiation factor eIF4F by multiple mechanisms in human cytomegalovirus- infected cells. Journal of Virology 79:8057-8064.
113. Lenarcic EM, Ziehr B, De Leon G, Mitchell D, Moorman NJ. 2014. Differential role for host translation factors in host and viral protein synthesis during human cytomegalovirus infection. Journal of Virology 88:1473-1483.
114. Walsh D, Perez C, Notary J, Mohr I. 2005. Regulation of the translation initiation factor eIF4F by multiple mechanisms in human cytomegalovirus- infected cells. Journal of Virology 79:8057-8064.
115. Hinnebusch AG, Lorsch JR. 2012. The mechanism of eukaryotic translation initiation: new insights and challenges. Cold Spring Harbor Perspectives in Biology 4.
116. Jan E, Mohr I, Walsh D. 2016. A Cap-to-Tail Guide to mRNA Translation Strategies in Virus-Infected Cells. Annual Review of Virology 3:283-307.
117. Ziehr B, Vincent HA, Moorman NJ. 2016. Human Cytomegalovirus pTRS1 and pIRS1 Antagonize Protein Kinase R To Facilitate Virus Replication. J Virol 90:3839-3848.
118. McArdle J, Moorman NJ, Munger J. 2012. HCMV targets the metabolic stress response through activation of AMPK whose activity is important for viral replication. PLoS Pathog 8:e1002502.
119. Moorman NJ, Shenk T. 2010. Rapamycin-resistant mTORC1 kinase activity is required for herpesvirus replication. Journal of Virology 84:5260-5269.
120. Gingras A-C, Raught B, Sonenberg N. 2001. Regulation of translation initiation by FRAP/mTOR. Genes & Development 15:807-826.
121. Gingras AC, Raught B, Sonenberg N. 1999. eIF4 initiation factors: effectors of mRNA recruitment to ribosomes and regulators of translation. Annual Review of Biochemistry 68:913-963.
49 122. Tahara SM, Morgan MA, Shatkin AJ. 1981. Two forms of purified m7G-cap binding protein with different effects on capped mRNA translation in extracts of uninfected and poliovirus-infected HeLa cells. The Journal of Biological Chemistry 256:7691-7694.
123. Grifo JA, Tahara SM, Morgan MA, Shatkin AJ, Merrick WC. 1983. New initiation factor activity required for globin mRNA translation. Journal of Biological Chemistry 258:5804-5810.
124. Sonenberg N, Shatkin AJ. 1978. Nonspecific effect of m7GMP on protein-RNA interactions. The Journal of Biological Chemistry 253:6630-6632.
125. Thomas A, Spaan W, van Steeg H, Voorma HO, Benne R. 1980. Mode of action of protein synthesis initiation factor eIF-1 from rabbit reticulocytes. FEBS letters 116:67-71.
126. Carlberg U, Nilsson A, Nygard O. 1990. Functional properties of phosphorylated elongation factor 2. Eur J Biochem 191:639-645.
127. Gao YG, Selmer M, Dunham CM, Weixlbaumer A, Kelley AC, Ramakrishnan V. 2009. The structure of the ribosome with elongation factor G trapped in the posttranslocational state. Science 326:694-699.
128. Kimata Y, Kohno K. 1994. Elongation factor 2 mutants deficient in diphthamide formation show temperature-sensitive cell growth. J Biol Chem 269:13497- 13501.
129. Stansfield I, Kushnirov VV, Jones KM, Tuite MF. 1997. A conditional-lethal translation termination defect in a sup45 mutant of the yeast Saccharomyces cerevisiae. European journal of biochemistry / FEBS 245:557-563.
130. Zhouravleva G, Frolova L, Le Goff X, Le Guellec R, Inge-Vechtomov S, Kisselev L, Philippe M. 1995. Termination of translation in eukaryotes is governed by two interacting polypeptide chain release factors, eRF1 and eRF3. The EMBO Journal 14:4065-4072.
131. De Benedetti A, Rhoads RE. 1990. Overexpression of eukaryotic protein synthesis initiation factor 4E in HeLa cells results in aberrant growth and morphology. Proceedings of the National Academy of Sciences of the United States of America 87:8212-8216.
50 132. Fan H, Penman S. 1970. Regulation of protein synthesis in mammalian cells. Journal of Molecular Biology 50:655-670.
133. Singer RH, Penman S. 1972. Biological Sciences: Stability of HeLa Cell mRNA in Actinomycin. Nature 240:100-102.
134. Jackson RJ, Hellen CU, Pestova TV. 2010. The mechanism of eukaryotic translation initiation and principles of its regulation. Nat Rev Mol Cell Biol 11:113- 127.
135. Imataka H, Sonenberg N. 1997. Human eukaryotic translation initiation factor 4G (eIF4G) possesses two separate and independent binding sites for eIF4A. Molecular and Cellular Biology 17:6940-6947.
136. Sonenberg N, Trachsel H, Hecht S, Shatkin AJ. 1980. Differential stimulation of capped mRNA translation in vitro by cap binding protein. Nature 285:331-333.
137. Marcotrigiano J, Gingras AC, Sonenberg N, Burley SK. 1999. Cap-dependent translation initiation in eukaryotes is regulated by a molecular mimic of eIF4G. Molecular Cell 3:707-716.
138. Slepenkov SV, Korneeva NL, Rhoads RE. 2008. Kinetic mechanism for assembly of the m7GpppG.eIF4E.eIF4G complex. The Journal of Biological Chemistry 283:25227-25237.
139. Hinton TM, Coldwell MJ, Carpenter GA, Morley SJ, Pain VM. 2007. Functional analysis of individual binding activities of the scaffold protein eIF4G. The Journal of Biological Chemistry 282:1695-1708.
140. Shin B-S, Kim J-R, Walker SE, Dong J, Lorsch JR, Dever TE. 2011. Initiation Factor eIF2γ Promotes eIF2–GTP–Met-tRNAiMet Ternary Complex Binding to the 40S Ribosome. Nature structural & molecular biology 18:1227-1234.
141. Methot N, Song MS, Sonenberg N. 1996. A region rich in aspartic acid, arginine, tyrosine, and glycine (DRYG) mediates eukaryotic initiation factor 4B (eIF4B) self-association and interaction with eIF3. Mol Cell Biol 16:5328-5334.
142. Asano K, Clayton J, Shalev A, Hinnebusch AG. 2000. A multifactor complex of eukaryotic initiation factors, eIF1, eIF2, eIF3, eIF5, and initiator tRNA(Met) is an
51 important translation initiation intermediate in vivo. Genes & Development 14:2534-2546.
143. Valásek L, Nielsen KH, Zhang F, Fekete CA, Hinnebusch AG. 2004. Interactions of eukaryotic translation initiation factor 3 (eIF3) subunit NIP1/c with eIF1 and eIF5 promote preinitiation complex assembly and regulate start codon selection. Molecular and Cellular Biology 24:9437-9455.
144. Rozovsky N, Butterworth AC, Moore MJ. 2008. Interactions between eIF4AI and its accessory factors eIF4B and eIF4H. RNA 14:2136-2148.
145. LeFebvre AK, Korneeva NL, Trutschl M, Cvek U, Duzan RD, Bradley CA, Hershey JWB, Rhoads RE. 2006. Translation initiation factor eIF4G-1 binds to eIF3 through the eIF3e subunit. The Journal of Biological Chemistry 281:22917- 22932.
146. Kozak M. 1980. Influence of mRNA secondary structure on binding and migration of 40S ribosomal subunits. Cell 19:79-90.
147. Parsyan A, Svitkin Y, Shahbazian D, Gkogkas C, Lasko P, Merrick WC, Sonenberg N. 2011. mRNA helicases: the tacticians of translational control. Nat Rev Mol Cell Biol 12:235-245.
148. Rozovsky N, Butterworth AC, Moore MJ. 2008. Interactions between eIF4AI and its accessory factors eIF4B and eIF4H. RNA 14:2136-2148.
149. Berthelot K, Muldoon M, Rajkowitsch L, Hughes J, McCarthy JEG. 2004. Dynamics and processivity of 40S ribosome scanning on mRNA in yeast. Molecular Microbiology 51:987-1001.
150. Parsyan A, Shahbazian D, Martineau Y, Petroulakis E, Alain T, Larsson O, Mathonnet G, Tettweiler G, Hellen CU, Pestova TV, Svitkin YV, Sonenberg N. 2009. The helicase protein DHX29 promotes translation initiation, cell proliferation, and tumorigenesis. Proc Natl Acad Sci U S A 106:22217-22222.
151. Lee C-S, Dias AP, Jedrychowski M, Patel AH, Hsu JL, Reed R. 2008. Human DDX3 functions in translation and interacts with the translation initiation factor eIF3. Nucleic Acids Research 36:4708-4718.
52 152. Hartman TR, Qian S, Bolinger C, Fernandez S, Schoenberg DR, Boris- Lawrie K. 2006. RNA helicase A is necessary for translation of selected messenger RNAs. Nature Structural & Molecular Biology 13:509-516.
153. Pisareva VP, Pisarev AV, Komar AA, Hellen CUT, Pestova TV. 2008. Translation initiation on mammalian mRNAs with structured 5'UTRs requires DExH-box protein DHX29. Cell 135:1237-1250.
154. Kozak M. 1987. An analysis of 5'-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Research 15:8125-8148.
155. Pestova TV, Lomakin IB, Lee JH, Choi SK, Dever TE, Hellen CU. 2000. The joining of ribosomal subunits in eukaryotes requires eIF5B. Nature 403:332-335.
156. Lin TA, Kong X, Haystead TA, Pause A, Belsham G, Sonenberg N, Lawrence JC. 1994. PHAS-I as a link between mitogen-activated protein kinase and translation initiation. Science (New York, NY) 266:653-656.
157. Jefferies HB, Fumagalli S, Dennis PB, Reinhard C, Pearson RB, Thomas G. 1997. Rapamycin suppresses 5'TOP mRNA translation through inhibition of p70s6k. The EMBO journal 16:3693-3704.
158. Loewith R, Jacinto E, Wullschleger S, Lorberg A, Crespo JL, Bonenfant D, Oppliger W, Jenoe P, Hall MN. 2002. Two TOR complexes, only one of which is rapamycin sensitive, have distinct roles in cell growth control. Molecular Cell 10:457-468.
159. Helliwell SB, Wagner P, Kunz J, Deuter-Reinhard M, Henriquez R, Hall MN. 1994. TOR1 and TOR2 are structurally and functionally similar but not identical phosphatidylinositol kinase homologues in yeast. Molecular Biology of the Cell 5:105-118.
160. Kim D-H, Sarbassov DD, Ali SM, King JE, Latek RR, Erdjument-Bromage H, Tempst P, Sabatini DM. 2002. mTOR interacts with raptor to form a nutrient- sensitive complex that signals to the cell growth machinery. Cell 110:163-175.
161. Sarbassov DD, Ali SM, Kim D-H, Guertin DA, Latek RR, Erdjument-Bromage H, Tempst P, Sabatini DM. 2004. Rictor, a novel binding partner of mTOR, defines a rapamycin-insensitive and raptor-independent pathway that regulates the cytoskeleton. Current biology: CB 14:1296-1302.
53 162. Thoreen CC. 2013. Many roads from mTOR to eIF4F. Biochemical Society Transactions 41:913-916.
163. Pause A, Belsham GJ, Gingras AC, Donzé O, Lin TA, Lawrence JC, Sonenberg N. 1994. Insulin-dependent stimulation of protein synthesis by phosphorylation of a regulator of 5'-cap function. Nature 371:762-767.
164. Fadden P, Haystead TA, Lawrence JC. 1997. Identification of phosphorylation sites in the translational regulator, PHAS-I, that are controlled by insulin and rapamycin in rat adipocytes. The Journal of Biological Chemistry 272:10240- 10247.
165. Shahbazian D, Roux PP, Mieulet V, Cohen MS, Raught B, Taunton J, Hershey JWB, Blenis J, Pende M, Sonenberg N. 2006. The mTOR/PI3K and MAPK pathways converge on eIF4B to control its phosphorylation and activity. The EMBO journal 25:2781-2791.
166. Faller WJ, Jackson TJ, Knight JRP, Ridgway RA, Jamieson T, Karim SA, Jones C, Radulescu S, Huels DJ, Myant KB, Dudek KM, Casey HA, Scopelliti A, Cordero JB, Vidal M, Pende M, Ryazanov AG, Sonenberg N, Meyuhas O, Hall MN, Bushell M, Willis AE, Sansom OJ. 2015. mTORC1- mediated translational elongation limits intestinal tumour initiation and growth. Nature 517:497-500.
167. Ozes AR, Feoktistova K, Avanzino BC, Fraser CS. 2011. Duplex unwinding and ATPase activities of the DEAD-box helicase eIF4A are coupled by eIF4G and eIF4B. J Mol Biol 412:674-687.
168. Wang X, Li W, Williams M, Terada N, Alessi DR, Proud CG. 2001. Regulation of elongation factor 2 kinase by p90RSK1 and p70 S6 kinase. The EMBO Journal 20:4370-4379.
169. Wang L, Wang X, Proud CG. 2000. Activation of mRNA translation in rat cardiac myocytes by insulin involves multiple rapamycin-sensitive steps. American Journal of Physiology Heart and Circulatory Physiology 278:H1056-1068.
170. Redpath NT, Foulstone EJ, Proud CG. 1996. Regulation of translation elongation factor-2 by insulin via a rapamycin-sensitive signalling pathway. The EMBO journal 15:2291-2297.
54 171. Gao X, Pan D. 2001. TSC1 and TSC2 tumor suppressors antagonize insulin signaling in cell growth. Genes & Development 15:1383-1392.
172. Manning BD, Cantley LC. 2003. Rheb fills a GAP between TSC and TOR. Trends in Biochemical Sciences 28:573-576.
173. Saucedo LJ, Gao X, Chiarelli DA, Li L, Pan D, Edgar BA. 2003. Rheb promotes cell growth as a component of the insulin/TOR signalling network. Nature Cell Biology 5:566-571.
174. Inoki K, Li Y, Xu T, Guan K-L. 2003. Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling. Genes & Development 17:1829- 1834.
175. Hardie DG, Schaffer BE, Brunet A. 2016. AMPK: An Energy-Sensing Pathway with Multiple Inputs and Outputs. Trends Cell Biol 26:190-201.
176. Inoki K, Zhu T, Guan K-L. 2003. TSC2 mediates cellular energy response to control cell growth and survival. Cell 115:577-590.
177. Manning BD, Tee AR, Logsdon MN, Blenis J, Cantley LC. 2002. Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3-kinase/akt pathway. Molecular Cell 10:151-162.
178. Jang SK, Kräusslich HG, Nicklin MJ, Duke GM, Palmenberg AC, Wimmer E. 1988. A segment of the 5' nontranslated region of encephalomyocarditis virus RNA directs internal entry of ribosomes during in vitro translation. Journal of Virology 62:2636-2643.
179. Pelletier J, Sonenberg N. 1988. Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature 334:320- 325.
180. Jackson RJ. 2005. Alternative mechanisms of initiating translation of mammalian mRNAs. Biochemical Society Transactions 33:1231-1241.
181. Sweeney TR, Abaeva IS, Pestova TV, Hellen CUT. 2014. The mechanism of translation initiation on Type 1 picornavirus IRESs. The EMBO journal 33:76-92.
55 182. Schüler M, Connell SR, Lescoute A, Giesebrecht J, Dabrowski M, Schroeer B, Mielke T, Penczek PA, Westhof E, Spahn CMT. 2006. Structure of the ribosome-bound cricket paralysis virus IRES RNA. Nature Structural & Molecular Biology 13:1092-1096.
183. Spahn CMT, Jan E, Mulder A, Grassucci RA, Sarnow P, Frank J. 2004. Cryo- EM visualization of a viral internal ribosome entry site bound to human ribosomes: the IRES functions as an RNA-based translation factor. Cell 118:465- 475.
184. Kieft JS. 2008. Viral IRES RNA structures and ribosome interactions. Trends Biochem Sci 33:274-283.
185. Komar AA, Hatzoglou M. 2011. Cellular IRES-mediated translation: the war of ITAFs in pathophysiological states. Cell Cycle 10:229-240.
186. Gil A, Sharp PA, Jamison SF, Garcia-Blanco MA. 1991. Characterization of cDNAs encoding the polypyrimidine tract-binding protein. Genes & Development 5:1224-1236.
187. Mitchell SA, Spriggs KA, Coldwell MJ, Jackson RJ, Willis AE. 2003. The Apaf-1 Internal Ribosome Entry Segment Attains the Correct Structural Conformation for Function via Interactions with PTB and unr. Molecular Cell 11:757-771.
188. Izumi RE, Das S, Barat B, Raychaudhuri S, Dasgupta A. 2004. A peptide from autoantigen La blocks poliovirus and hepatitis C virus cap-independent translation and reveals a single tyrosine critical for La RNA binding and translation stimulation. Journal of Virology 78:3763-3776.
189. McBratney S, Sarnow P. 1996. Evidence for involvement of trans-acting factors in selection of the AUG start codon during eukaryotic translational initiation. Molecular and Cellular Biology 16:3523-3534.
190. Kaminski A, Hunt SL, Patton JG, Jackson RJ. 1995. Direct evidence that polypyrimidine tract binding protein (PTB) is essential for internal initiation of translation of encephalomyocarditis virus RNA. RNA (New York, NY) 1:924-938.
56 191. Costa-Mattioli M, Svitkin Y, Sonenberg N. 2004. La autoantigen is necessary for optimal function of the poliovirus and hepatitis C virus internal ribosome entry site in vivo and in vitro. Molecular and Cellular Biology 24:6861-6870.
192. Fitzgerald KD, Semler BL. 2009. Bridging IRES elements in mRNAs to the eukaryotic translation apparatus. Biochimica Et Biophysica Acta 1789:518-528.
193. Svitkin YV, Meerovitch K, Lee HS, Dholakia JN, Kenan DJ, Agol VI, Sonenberg N. 1994. Internal translation initiation on poliovirus RNA: further characterization of La function in poliovirus translation in vitro. Journal of Virology 68:1544-1550.
194. Hunt SL, Jackson RJ. 1999. Polypyrimidine-tract binding protein (PTB) is necessary, but not sufficient, for efficient internal initiation of translation of human rhinovirus-2 RNA. RNA (New York, NY) 5:344-359.
195. Kozak M. 1991. A short leader sequence impairs the fidelity of initiation by eukaryotic ribosomes. Gene Expression 1:111-115.
196. Kozak M. 1991. Structural features in eukaryotic mRNAs that modulate the initiation of translation. The Journal of Biological Chemistry 266:19867-19870.
197. Pestova TV, Kolupaeva VG. 2002. The roles of individual eukaryotic translation initiation factors in ribosomal scanning and initiation codon selection. Genes & Development 16:2906-2922.
198. Kozak M. 1991. Effects of long 5' leader sequences on initiation by eukaryotic ribosomes in vitro. Gene Expression 1:117-125.
199. Abaeva IS, Marintchev A, Pisareva VP, Hellen CUT, Pestova TV. 2011. Bypassing of stems versus linear base-by-base inspection of mammalian mRNAs during ribosomal scanning. The EMBO journal 30:115-129.
200. Soto-Rifo R, Rubilar PS, Limousin T, de Breyne S, Décimo D, Ohlmann T. 2012. DEAD-box protein DDX3 associates with eIF4F to promote translation of selected mRNAs. The EMBO journal 31:3745-3756.
201. Svitkin YV, Pause A, Haghighat A, Pyronnet S, Witherell G, Belsham GJ, Sonenberg N. 2001. The requirement for eukaryotic initiation factor 4A (eIF4A)
57 in translation is in direct proportion to the degree of mRNA 5 secondary structure. RNA 7:382-394.
202. Rubio CA, Weisburd B, Holderfield M, Arias C, Fang E, DeRisi JL, Fanidi A. 2014. Transcriptome-wide characterization of the eIF4A signature highlights plasticity in translation regulation. Genome Biology 15:476.
203. Arribere JA, Gilbert WV. 2013. Roles for transcript leaders in translation and mRNA decay revealed by transcript leader sequencing. Genome Research 23:977-987.
204. Hinnebusch AG, Ivanov IP, Sonenberg N. 2016. Translational control by 5 - untranslated regions of eukaryotic mRNAs. Science 352:1413-1416.
205. Ingolia NT, Ghaemmaghami S, Newman JRS, Weissman JS. 2009. Genome- Wide Analysis in Vivo of Translation with Nucleotide Resolution Using Ribosome Profiling. Science 324:218-223.
206. Johnstone TG, Bazzini AA, Giraldez AJ. 2016. Upstream ORFs are prevalent translational repressors in vertebrates. The EMBO journal 35:706-723.
207. Sonenberg N, Hinnebusch AG. 2009. Regulation of Translation Initiation in Eukaryotes: Mechanisms and Biological Targets. Cell 136:731-745.
208. Algire MA, Maag D, Lorsch JR. 2005. Pi release from eIF2, not GTP hydrolysis, is the step controlled by start-site selection during eukaryotic translation initiation. Molecular Cell 20:251-262.
209. Flynn A, Oldfield S, Proud CG. 1993. The role of the beta-subunit of initiation factor eIF-2 in initiation complex formation. Biochimica Et Biophysica Acta 1174:117-121.
210. Pavitt GD, Ramaiah KV, Kimball SR, Hinnebusch AG. 1998. eIF2 independently binds two distinct eIF2B subcomplexes that catalyze and regulate guanine-nucleotide exchange. Genes & Development 12:514-526.
211. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. 2001. Tight binding of the phosphorylated alpha subunit of initiation factor 2 (eIF2alpha) to the regulatory subunits of guanine nucleotide exchange factor eIF2B is
58 required for inhibition of translation initiation. Molecular and Cellular Biology 21:5018-5030.
212. Singh CR, Udagawa T, Lee B, Wassink S, He H, Yamamoto Y, Anderson JT, Pavitt GD, Asano K. 2007. Change in nutritional status modulates the abundance of critical pre-initiation intermediate complexes during translation initiation in vivo. J Mol Biol 370:315-330.
213. Wek RC, Jiang HY, Anthony TG. 2006. Coping with stress: eIF2 kinases and translational control. Biochemical Society Transactions 34:7-11.
214. Lu L, Han AP, Chen JJ. 2001. Translation initiation control by heme-regulated eukaryotic initiation factor 2alpha kinase in erythroid cells under cytoplasmic stresses. Molecular and Cellular Biology 21:7971-7980.
215. Walter P, Ron D. 2011. The unfolded protein response: from stress pathway to homeostatic regulation. Science 334:1081-1086.
216. Dever TE, Feng L, Wek RC, Cigan AM, Donahue TF, Hinnebusch AG. 1992. Phosphorylation of initiation factor 2 alpha by protein kinase GCN2 mediates gene-specific translational control of GCN4 in yeast. Cell 68:585-596.
217. Dever TE, Hinnebusch AG. 2005. GCN2 whets the appetite for amino acids. Mol Cell 18:141-142.
218. Dar AC, Dever TE, Sicheri F. 2005. Higher-order substrate recognition of eIF2alpha by the RNA-dependent protein kinase PKR. Cell 122:887-900.
219. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition. Cell 122:901-913.
220. Romano PR, Zhang F, Tan SL, Garcia-Barrio MT, Katze MG, Dever TE, Hinnebusch AG. 1998. Inhibition of double-stranded RNA-dependent protein kinase PKR by vaccinia virus E3: role of complex formation and the E3 N- terminal domain. Mol Cell Biol 18:7304-7316.
221. Zhang F, Romano PR, Nagamura-Inoue T, Tian B, Dever TE, Mathews MB, Ozato K, Hinnebusch AG. 2001. Binding of double-stranded RNA to protein
59 kinase PKR is required for dimerization and promotes critical autophosphorylation events in the activation loop. J Biol Chem 276:24946- 24958.
222. Novoa I, Zeng H, Harding HP, Ron D. 2001. Feedback Inhibition of the Unfolded Protein Response by GADD34-Mediated Dephosphorylation of eIF2α. The Journal of Cell Biology 153:1011-1022.
223. Choy MS, Yusoff P, Lee IC, Newton JC, Goh CW, Page R, Shenolikar S, Peti W. 2015. Structural and functional analysis of the GADD34:PP1 eIF2α phosphatase. Cell reports 11:1885-1891.
224. Pestova TV, Kolupaeva VG, Lomakin IB, Pilipenko EV, Shatsky IN, Agol VI, Hellen CU. 2001. Molecular mechanisms of translation initiation in eukaryotes. Proc Natl Acad Sci U S A 98:7029-7036.
225. McKinney C, Perez C, Mohr I. 2012. Poly(A) binding protein abundance regulates eukaryotic translation initiation factor 4F assembly in human cytomegalovirus-infected cells. Proceedings of the National Academy of Sciences of the United States of America 109:5627-5632.
226. Kudchodkar SB, Yu Y, Maguire TG, Alwine JC. 2004. Human cytomegalovirus infection induces rapamycin-insensitive phosphorylation of downstream effectors of mTOR kinase. Journal of Virology 78:11030-11039.
227. McKinney C, Zavadil J, Bianco C, Shiflett L, Brown S, Mohr I. 2014. Global reprogramming of the cellular translational landscape facilitates cytomegalovirus replication. Cell Rep 6:9-17.
228. Kudchodkar SB, Yu Y, Maguire TG, Alwine JC. 2006. Human cytomegalovirus infection alters the substrate specificities and rapamycin sensitivities of raptor- and rictor-containing complexes. Proceedings of the National Academy of Sciences of the United States of America 103:14182-14187.
229. Clippinger AJ, Maguire TG, Alwine JC. 2011. Human cytomegalovirus infection maintains mTOR activity and its perinuclear localization during amino acid deprivation. J Virol 85:9369-9376.
60 230. Kudchodkar SB, Del Prete GQ, Maguire TG, Alwine JC. 2007. AMPK- mediated inhibition of mTOR kinase is circumvented during immediate-early times of human cytomegalovirus infection. Journal of Virology 81:3649-3651.
231. Tilton C, Clippinger AJ, Maguire T, Alwine JC. 2011. Human cytomegalovirus induces multiple means to combat reactive oxygen species. J Virol 85:12585- 12593.
232. Bai Y, Xuan B, Liu H, Zhong J, Yu D, Qian Z. 2015. Tuberous Sclerosis Complex Protein 2-Independent Activation of mTORC1 by Human Cytomegalovirus pUL38. Journal of Virology 89:7625-7635.
233. Johnson RA, Wang X, Ma X-L, Huong S-M, Huang E-S. 2001. Human Cytomegalovirus Up-Regulates the Phosphatidylinositol 3-Kinase (PI3-K) Pathway: Inhibition of PI3-K Activity Inhibits Viral Replication and Virus-Induced Signaling. Journal of Virology 75:6022-6032.
234. Buchkovich NJ, Yu Y, Zampieri CA, Alwine JC. 2008. The TORrid affairs of viruses: effects of mammalian DNA viruses on the PI3K-Akt-mTOR signalling pathway. Nat Rev Microbiol 6:266-275.
235. Clippinger AJ, Maguire TG, Alwine JC. 2011. The Changing Role of mTOR Kinase in the Maintenance of Protein Synthesis during Human Cytomegalovirus Infection. Journal of Virology 85:3930-3939.
236. Spencer CM, Schafer XL, Moorman NJ, Munger J. 2011. Human cytomegalovirus induces the activity and expression of acetyl-coenzyme A carboxylase, a fatty acid biosynthetic enzyme whose inhibition attenuates viral replication. J Virol 85:5814-5824.
237. Grainger L, Cicchini L, Rak M, Petrucelli A, Fitzgerald KD, Semler BL, Goodrum F. 2010. Stress-inducible alternative translation initiation of human cytomegalovirus latency protein pUL138. Journal of Virology 84:9472-9486.
238. Petrucelli A, Rak M, Grainger L, Goodrum F. 2009. Characterization of a novel Golgi apparatus-localized latency determinant encoded by human cytomegalovirus. J Virol 83:5615-5629.
239. Weingarten-Gabbay S, Elias-Kirma S, Nir R, Gritsenko AA, Stern-Ginossar N, Yakhini Z, Weinberger A, Segal E. 2016. Systematic discovery of cap-
61 independent translation sequences in human and viral genomes. Science 351:aad4939.
240. Isler JA, Skalet AH, Alwine JC. 2005. Human cytomegalovirus infection activates and regulates the unfolded protein response. Journal of Virology 79:6890-6899.
241. Yu Y, Pierciey FJ, Maguire TG, Alwine JC. 2013. PKR-like endoplasmic reticulum kinase is necessary for lipogenic activation during HCMV infection. PLoS pathogens 9:e1003266.
242. Buchkovich NJ, Maguire TG, Yu Y, Paton AW, Paton JC, Alwine JC. 2008. Human cytomegalovirus specifically controls the levels of the endoplasmic reticulum chaperone BiP/GRP78, which is required for virion assembly. J Virol 82:31-39.
243. Buchkovich NJ, Yu Y, Pierciey FJ, Alwine JC. 2010. Human cytomegalovirus induces the endoplasmic reticulum chaperone BiP through increased transcription and activation of translation by using the BiP internal ribosome entry site. Journal of Virology 84:11479-11486.
244. Kim YK, Back SH, Rho J, Lee SH, Jang SK. 2001. La autoantigen enhances translation of BiP mRNA. Nucleic Acids Research 29:5009-5016.
245. Isler JA, Skalet AH, Alwine JC. 2005. Human cytomegalovirus infection activates and regulates the unfolded protein response. J Virol 79:6890-6899.
246. Bierle CJ, Semmens KM, Geballe AP. 2013. Double-stranded RNA binding by the human cytomegalovirus PKR antagonist TRS1. Virology 442:28-37.
247. Vincent HA, Ziehr B, Moorman NJ. 2017. Mechanism of Protein Kinase R Inhibition by Human Cytomegalovirus pTRS1. J Virol 91.
248. Yu Y, Pierciey FJ, Jr., Maguire TG, Alwine JC. 2013. PKR-like endoplasmic reticulum kinase is necessary for lipogenic activation during HCMV infection. PLoS Pathog 9:e1003266.
62 249. Marshall EE, Bierle CJ, Brune W, Geballe AP. 2009. Essential role for either TRS1 or IRS1 in human cytomegalovirus replication. Journal of Virology 83:4112-4120.
250. Marshall EE, Bierle CJ, Brune W, Geballe AP. 2009. Essential role for either TRS1 or IRS1 in human cytomegalovirus replication. J Virol 83:4112-4120.
251. Kitajewski J, Schneider RJ, Safer B, Munemitsu SM, Samuel CE, Thimmappaya B, Shenk T. 1986. Adenovirus VAI RNA antagonizes the antiviral action of interferon by preventing activation of the interferon-induced eIF-2 alpha kinase. Cell 45:195-200.
252. Sharp TV, Schwemmle M, Jeffrey I, Laing K, Mellor H, Proud CG, Hilse K, Clemens MJ. 1993. Comparative analysis of the regulation of the interferon- inducible protein kinase PKR by Epstein-Barr virus RNAs EBER-1 and EBER-2 and adenovirus VAI RNA. Nucleic Acids Res 21:4483-4490.
253. Cassady KA, Gross M, Roizman B. 1998. The herpes simplex virus US11 protein effectively compensates for the gamma1(34.5) gene if present before activation of protein kinase R by precluding its phosphorylation and that of the alpha subunit of eukaryotic translation initiation factor 2. J Virol 72:8620-8626.
254. He B, Gross M, Roizman B. 1997. The gamma(1)34.5 protein of herpes simplex virus 1 complexes with protein phosphatase 1alpha to dephosphorylate the alpha subunit of the eukaryotic translation initiation factor 2 and preclude the shutoff of protein synthesis by double-stranded RNA-activated protein kinase. Proc Natl Acad Sci U S A 94:843-848.
255. Xuan B, Qian Z, Torigoi E, Yu D. 2009. Human cytomegalovirus protein pUL38 induces ATF4 expression, inhibits persistent JNK phosphorylation, and suppresses endoplasmic reticulum stress-induced cell death. J Virol 83:3463- 3474.
256. Bertolotti A, Zhang Y, Hendershot LM, Harding HP, Ron D. 2000. Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat Cell Biol 2:326-332.
257. Harding HP, Novoa I, Bertolotti A, Zeng H, Zhang Y, Urano F, Jousse C, Ron D. 2001. Translational regulation in the cellular response to biosynthetic load on the endoplasmic reticulum. Cold Spring Harb Symp Quant Biol 66:499-508.
63 258. Yu Y, Pierciey FJ, Maguire TG, Alwine JC. 2013. PKR-like endoplasmic reticulum kinase is necessary for lipogenic activation during HCMV infection. PLoS pathogens 9:e1003266.
259. Donnelly N, Gorman AM, Gupta S, Samali A. 2013. The eIF2alpha kinases: their structures and functions. Cell Mol Life Sci 70:3493-3511.
260. Won S, Eidenschenk C, Arnold CN, Siggs OM, Sun L, Brandl K, Mullen TM, Nemerow GR, Moresco EM, Beutler B. 2012. Increased susceptibility to DNA virus infection in mice with a GCN2 mutation. J Virol 86:1802-1808.
261. Isler JA, Skalet AH, Alwine JC. 2005. Human cytomegalovirus infection activates and regulates the unfolded protein response. Journal of Virology 79:6890-6899.
262. Hakki M, Geballe AP. 2008. Cellular serine/threonine phosphatase activity during human cytomegalovirus infection. Virology 380:255-263.
263. Hakki M, Geballe AP. 2008. Cellular serine/threonine phosphatase activity during human cytomegalovirus infection. Virology 380:255-263.
264. Stasiak PC, Mocarski ES. 1992. Transactivation of the cytomegalovirus ICP36 gene promoter requires the alpha gene product TRS1 in addition to IE1 and IE2. Journal of Virology 66:1050-1058.
265. Pari GS, Kacica MA, Anders DG. 1993. Open reading frames UL44, IRS1/TRS1, and UL36-38 are required for transient complementation of human cytomegalovirus oriLyt-dependent DNA synthesis. Journal of Virology 67:2575- 2582.
266. Iskenderian AC, Huang L, Reilly A, Stenberg RM, Anders DG. 1996. Four of eleven loci required for transient complementation of human cytomegalovirus DNA replication cooperate to activate expression of replication genes. Journal of Virology 70:383-392.
267. Blankenship CA, Shenk T. 2002. Mutant human cytomegalovirus lacking the immediate-early TRS1 coding region exhibits a late defect. Journal of Virology 76:12290-12299.
64 268. Adamo JE, Schröer J, Shenk T. 2004. Human cytomegalovirus TRS1 protein is required for efficient assembly of DNA-containing capsids. Journal of Virology 78:10221-10229.
269. Chang H-W, Jacobs BL. 1993. Identification of a Conserved Motif That Is Necessary for Binding of the Vaccinia Virus E3L Gene Products to Double- Stranded RNA. Virology 194:537-547.
270. Hakki M, Geballe AP. 2005. Double-stranded RNA binding by human cytomegalovirus pTRS1. Journal of Virology 79:7311-7318.
271. Marshall EE, Bierle CJ, Brune W, Geballe AP. 2009. Essential role for either TRS1 or IRS1 in human cytomegalovirus replication. Journal of Virology 83:4112-4120.
272. White JP, Lloyd RE. 2012. Regulation of stress granules in virus systems. Trends in Microbiology 20:175-183.
273. Ziehr B, Lenarcic E, Vincent HA, Cecil C, Garcia B, Shenk T, Moorman NJ. 2015. Human cytomegalovirus TRS1 protein associates with the 7- methylguanosine mRNA cap and facilitates translation. Proteomics 15:1983- 1994.
274. Hakki M, Marshall EE, De Niro KL, Geballe AP. 2006. Binding and nuclear relocalization of protein kinase R by human cytomegalovirus TRS1. Journal of Virology 80:11817-11826.
65 CHAPTER 2: MECHANISM OF PKR INHIBITION BY HCMV PTRS11
OVERVIEW
Double stranded RNAs (dsRNA) produced during human cytomegalovirus
(HCMV) infection activate the antiviral kinase PKR, which potently inhibits virus replication. The HCMV pTRS1 and pIRS1 proteins antagonize PKR to promote HCMV protein synthesis and replication, however the mechanism by which pTRS1 inhibits PKR is unclear. PKR activation occurs in a three-step cascade. First, binding to dsRNA triggers PKR homodimerizaton. PKR dimers then autophosporylate, leading to a conformational shift that exposes the binding site for the PKR substrate eIF2α.
Consistent with previous in vitro studies, we found that pTRS1 bound and inhibited
PKR. pTRS1 binding to PKR was not mediated by an RNA intermediate, and mutations in the pTRS1 RNA binding domain did not affect PKR binding or inhibition. Rather, mutations that disrupted the pTRS1 interaction with PKR ablated the ability of pTRS1 to antagonize PKR activation by dsRNA. pTRS1 did not block PKR dimerization, and could bind and inhibit a constitutively dimerized PKR kinase domain. In addition, pTRS1 binding to PKR inhibited PKR kinase activity. Single amino acid point mutations in the conserved eIF2α binding domain of PKR disrupted pTRS1 binding, and rendered PKR resistant to inhibition by pTRS1. Consistent with a critical role for the conserved eIF2α
1 This chapter previously appeared as an article in the Journal of Virology. The original citation is as follows: Vincent HA, Ziehr B, Moorman NJ. “Mechanism of PKR inhibition by HCMV pTRS1.” J Virol. 2017 Feb 14;91(5).
66 contact site in PKR binding, pTRS1 bound an additional eIF2α kinase, HRI, and inhibited eIF2α phosphorylation in response to an HRI agonist. Together our data suggest that pTRS1 inhibits PKR by binding to conserved amino acids in the PKR eIF2α binding site and blocking PKR kinase activity.
INTRODUCTION
Protein kinase R (PKR) plays a critical role in the cellular intrinsic antiviral response (1). Double stranded RNAs (dsRNAs) produced during viral infection activate
PKR, leading to phosphorylation of its only known substrate, the alpha subunit of the eIF2 translation initiation factor (eIF2α (2-4). Phosphorylation of eIF2α reduces formation of the ternary complex, which is composed of eIF2α, GTP and a charged methionyl tRNA (tRNAMet) (5). Since the ternary complex is required for translation initiation, activation of PKR limits the synthesis of both cellular and viral proteins and inhibits virus replication (6, 7). Active PKR further limits virus replication by increasing type I interferon expression, stimulating apoptosis, inducing autophagy and possibly activating inflammasomes (8-14). Together these functions make PKR a potent antiviral effector capable of inhibiting the replication of many viruses.
Both RNA and DNA viruses generate dsRNA ligands that activate PKR during infection. Structured regions of RNA virus genomes and dsRNA replicative intermediates are the most likely triggers of PKR activation during RNA virus infection
(15-17). dsRNAs produced by overlapping transcriptional units during DNA virus replication likely drive PKR activation (18, 19). While PKR exists as an inactive monomer in uninfected cells, binding to dsRNA, mediated by two amino terminal dsRNA
67 binding domains (dsRBDs), results in a conformational shift that allows for PKR dimerization. Dimerization results in PKR autophosphorylation on multiple serine and threonine residues, however only phosphorylation on threonine 446 (Thr446) and 451
(Thr451) in the PKR kinase domain (KD) activation loop are necessary for PKR activation (16, 20-23). Following autophosphorylation, conserved residues within the αG helix in the C-lobe of the PKR kinase domain mediate its interaction with eIF2α (3, 4).
PKR binding to dsRNA ligands thus leads to a catalytically active PKR capable of binding and phosphorylating eIF2α, resulting in inhibition of protein synthesis and decreased viral replication.
Almost every virus encodes a PKR inhibitor, underscoring the critical role of PKR in the antiviral response (1). In several cases a single virus employs multiple proteins to limit PKR activation or mitigate the effects of eIF2α phosphorylation. For example, herpes simplex virus (HSV) encodes three proteins that limit PKR activation and eIF2α phosphorylation to facilitate virus replication. The vhs protein degrades viral dsRNAs that otherwise activate PKR, while the US11 protein binds PKR and prevents PKR- dependent eIF2α phosphorylation (12, 24). In addition, the ICP34.5 protein recruits the cellular protein phosphatase 1 (PP1) to dephosphorylate eIF2α and ensure continued synthesis of viral proteins (25). Similarly, the murine cytomegalovirus (MCMV) m142 and m143 proteins and the vaccinia virus (VV) K3L and E3L proteins each function to limit PKR activation (15, 26-33). Other viruses including orthomyxoviruses, flaviviruses and retroviruses each encode at least one PKR antagonist, highlighting the important role of PKR antagonism for successful viral replication (15, 17, 34).
68 Like the above viruses, human cytomegalovirus (HCMV) encodes PKR antagonists that are necessary for virus replication. The HCMV TRS1 and IRS1 proteins
(pTRS1 and pIRS1, respectively) each inhibit PKR activation (7, 35-37). The amino terminal two thirds of both pTRS1 and pIRS1 are conserved and contain a noncanonical
RNA binding domain that competitively inhibits PKR binding to dsRNA ligands in vitro
(38, 39). While the carboxyl-terminal third of pTRS1 and pIRS1 differ in sequence, both proteins contain a C-terminal PKR binding domain and both can inhibit PKR activation
(6, 7, 37). Expression of either pTRS1 or pIRS1 is sufficient to prevent PKR activation during HCMV infection of human fibroblasts, and viral protein synthesis and HCMV replication are dramatically reduced in the absence of both proteins (6, 7, 18). pTRS1 and pIRS1 have additional functions, including promoting mRNA translation and limiting autophagy (40-49). However, the viral replication defect observed in the absence of both pTRS1 and pIRS1 is largely reversed through inhibition of PKR expression (6, 7).
Though the other functions of pTRS1 and pIRS1 may play important roles in other cell types, inhibiting PKR activation appears to be the predominant role for pTRS1 and pIRS1 during HCMV infection of primary human fibroblasts.
While the role of pTRS1 and pIRS1 as PKR antagonists is well established, questions remain concerning the mechanism by which they inhibit PKR. pTRS1 binds both dsRNA and PKR, but the relative importance of these two functions for PKR antagonism is unknown. In this study we performed a series of molecular biology and biochemical experiments to further elucidate the molecular mechanism(s) by which pTRS1 prevents PKR activation. We found that pTRS1 binding to both dsRNA and
PKR contribute to PKR antagonism, although pTRS1 binding to PKR is the predominant
69 mechanism that mediates PKR inhibition by the full-length TRS1 protein. Our results suggest that pTRS1 does not block PKR dimerization, but rather prevents activating autophosphorylation of PKR dimers. pTRS1 binding to PKR is mediated through interactions with conserved amino acids in the PKR eIF2α contact site, and results in inhibition of PKR kinase activity. We also found that pTRS1 binds the HRI kinase and inhibits eIF2α phosphorylation in response to the HRI agonist arsenite in PKR-deficient cells. Together our results provide further insight into the mechanism of PKR antagonism by pTRS1.
MATERIALS AND METHODS
Cells and viruses. HEK293T cells and MRC-5 fibroblasts were cultured in
Dulbecco’s Modified Eagle Medium (Sigma) supplemented with 10% fetal bovine serum
(Sigma) and 100 U/ml penicillin-streptomycin (Sigma) at 37°C and 5% CO2. PKR- deficient HeLa cells (PKR KO), HEK293T cells containing an inducible TRS1 expression cassette (293T-pTRS1i), and PKR-deficient cells containing an inducible TRS1 expression cassette (PKR KO-pTRS1i) were maintained in 1 µg/ml puromycin in normal growth media. PKR-deficient HeLa cells have been previously described (35). The
HCMV AD169inGFP strain was used for all infections. This strain is derived from a bacterial artificial chromosome (BAC) and contains a green fluorescent protein (GFP) reporter driven by the SV40 promoter.
Generation of recombinant plasmids. Vectors expressing full-length TRS1 or
TRS1 truncation mutants fused to a carboxyl-terminal 6XHis epitope were a kind gift from Dr. Adam Geballe (University of Washington;(39)). To generate full-length TRS1
70 containing arginine to alanine mutations that disrupt dsRNA binding
(R121,124,125A;(38)) the full-length TRS1 expression plasmid was amplified with primers Triple F and Triple R (Table 1), and recircularized using Gibson cloning (NEB).
The R121A,R124A,R125A point mutations were introduced into the 1-550 TRS1 expression vector, as above, using the pcDNATRS1 1-550 plasmid as a template. Point mutations within the pTRS1 PKR binding domain that disrupt PKR binding (R658A,
D660,661A, E662A;(7)) were incorporated into full-length TRS1 and TRS1triple using primers Mut1 F and Mut1 R (Table 1) followed by recircularization using a Gibson reaction (New England BioLabs), in which a mixture of DNA exonuclease, DNA polymerase and DNA ligase anneal overlapping complementary DNA ends into a contiguous double stranded DNA molecule. The portion of each clone containing the
TRS1 open reading frame (ORF) was sequenced to ensure the absence of unintended mutations.
The human protein kinase R (PKR) ORF containing homology arms to the pcDNA3.1 V5 His expression vector was ordered as a gBlock Gene Fragment (IDT;
Table 1). The pcDNA3.1 vector was digested with BamHI and XhoI, and the full-length
PKR ORF was inserted using a Gibson cloning reaction according to the manufacturer’s directions. The kinase-dead PKR (K296R) ORF containing homology arms to the pcDNA3.1 expression vector was also ordered as a gBlock Gene Fragment (Table 1) and introduced into pcDNA3.1 by Gibson cloning as above. A flag epitope was fused to the N-terminus of PKR K296R by amplification of the PKR ORF with primers PKRflag F and PKR R (Table 1) followed by Gibson cloning into the pcDNA3.1 vector digested with
BamHI and XhoI as above. An N-terminal myc tag was added to both wild type PKR
71 and PKR K296R as above using the primers PKRmyc F and PKR R. The following point mutations were introduced into PKR: E375V, T446A, A473T, D486V, T487A, E490A and F495R using the following primer pairs: PKR_E375V F and PKR_E375V R,
PKR_A446T F and PKR_ A446T R, PKR_T473A F and PKR_T473A R, PKR_D486V F and PKR_ D486V R, PKR_T487A F and PKR_ T487A R, PKR_E490A F and PKR_
E490A R, PKR_F495R F and PKR_ F495R R, respectively. All primer sequences are listed in Table 1. The full-length PKR expression plasmid was amplified from both myc- tagged wild type and myc-tagged PKR K296R, in pcDNA3.1 backgrounds, with the above primer pairs, gel extracted and circularized using Gibson cloning. For each PKR mutant, the portion of each vector containing the PKR ORF was sequenced to confirm the absence of additional mutations.
The PKR kinase domain (KD) fused to glutathione-S-transferase (GST) was created by first amplifying GST from pET-41 b(+) using primers GST-pCW F and GST R
(Table 1). The PKR KD (amino acids 259-551) was then amplified from the full-length
PKR expression vector using primers KDGST F and KDGSTpCW R (Table 1). The plasmid pCW-Cas9 (Addgene #50661) was digested with NheI and BamHI to remove the Cas9 open reading frame. The GST ORF and PKR KD were then cloned into the digested pCW vector using a three part Gibson reaction. Full-length TRS1 was introduced into the pCW plasmid by amplifying the TRS1 ORF using primers TRS1ind F and TRS1ind R (Table 1), followed by Gibson assembly into pCW-Cas9 previously digested with NheI and BamHI. Sanger sequencing was used to confirm the absence of unintended mutations.
72 The heme-regulated inhibitor (HRI) ORF was amplified from the pJP1563 vector
(DNASU #38827) using primers HRImyc F and HRImyc R (Tale 1). The HRI ORF was then cloned into the pcDNA3.1 vector digested with BamHI and XhoI via Gibson reaction. The complete HRI ORF was sequenced to ensure no unintended mutations.
Production of lentivirus stocks and pTRS1-inducible cell lines. Lentivirus stocks were generated by transfecting HEK293T cells with the pCW TRS1 expression plasmid (described above) and Lentivirus Packaging Mix (Sigma) using polyethylenimine (PEI; Polysciences). The media was collected at 48 hours and 72 hours after transfection and clarified by passage through a 0.22 µm filter. The virus was then stored at -80oC until use. 293T-pTRS1i and PKR KO-pTRS1i cells were generated by transduction with the pCW TRS1 lentivirus in the presence of 8 µg/ml polybrene.
After bulk selection in 1 µg/ml puromycin for 48 hours, clonal cell lines were isolated by limiting dilution in media containing puromycin. Clonal cells lines were assayed for minimal basal expression of pTRS1 and maximal induction of pTRS1 expression after treatment with 1 µg/ml doxycycline by Western blot.
PKR and HRI activation assays. The ability of pTRS1 or pTRS1 mutants to inhibit PKR was measured in HEK293T cells. Cells were transfected with 1 µg of either a GFP or TRS1 expression plasmid using PEI and harvested at 36 hours after transfection. Cells were washed with PBS, pelleted and lysed in 120 µl of PKR lysis buffer (50 mM Tris-HCl, pH 7.5; 150 mM NaCl; 0.5% NP-40; 10% glycerol; 1 mM EDTA, with protease and phosphatase inhibitors (Roche)) on ice for 15 min. Insoluble debris was removed by centrifugation at 21,000xg for 10 min at 4oC. Unless otherwise indicated, 1 mg/ml of poly(I:C) (LMW, Invivogen) was added to 100 µl of supernatant for
73 30 min at 30°C to activate PKR. To examine PKR KD activation, HEK293T cells were transfected with 500 ng of GST-KD expression plasmid together with 500 ng of either
GFP, TRS1 or TRS1mut1 expression plasmid for 36 hours. Doxycycline (1 µg/ml) was added to cells overnight prior to harvest to induce GST-KD expression. Cells were washed once with PBS, pelleted and stored at -80oC until processed. To measure activation of PKR mutants in PKR KO cells, PKR KO-pTRS1i, cells were co-transfected with 500 ng of PKR expression plasmid and 500 ng of either GFP or TRS1 expression plasmid. Twenty hours after DNA transfection, cells were transfected with 20 µg of poly(I:C) using Lipofectamine2000 (Thermo Fisher) for four hours. Cells were washed once in PBS, pelleted and stored at -80oC until processed.
PKR KO-pTRS1i cells were used to measure the effect of pTRS1 on arsenite- induced HRI activation. Cells were treated overnight with doxycycline (1 µg/ml) to induce pTRS1 expression. The next day, cells were treated with 0.05, 0.25 or 0.5 mM sodium arsenite (Sigma) for two hours to activate HRI (50). Cells were then washed once with PBS, pelleted and stored at -80oC until processed.
Analysis of protein expression. Unless otherwise stated, cells were lysed in
RIPA buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 1% NP-40; 0.25% deoxycholate; 1 mM EDTA containing protease and phosphatase inhibitors (Roche). Protein concentration was determined using a Bradford assay. Equal amounts of protein were resolved on 10% SDS-PAGE gels, transferred to nitrocellulose membranes (Amersham) and blocked in 5% non-fat milk in TBS-T (50 mM Tris, pH 7.4; 150 mM NaCl; 0.1%
Tween-20) for 1 hour at room temperature (RT) or overnight at 4oC. For mouse monoclonal antibodies, membranes were incubated with primary antibodies for either 1
74 hour at RT or overnight at 4oC in 1% BSA in TBS-T. For rabbit polyclonal antibodies, membranes were blocked as above and incubated at 4oC overnight in 5% BSA in TBS-
T, with the exception of the aHis antibody which was diluted in 1% BSA in TBS-T.
Membranes were incubated in appropriate secondary antibodies (KPL) at a 1:10,000 dilution in 1% BSA in TBS-T for 1 hour at RT. A modified protocol was used for total and phosphorylated eIF2α blots. Membranes probed for total eIF2α were blocked and incubated in primary antibody as above, and then incubated in secondary antibody
(1:10,000) in 5% milk in TBS-T for 1 hour at RT before being developed. Membranes probed for phospho-eIF2α (Ser51) were blocked in 5% BSA in TBS-T overnight at 4oC prior to an overnight incubation in primary antibody at 4oC. Membranes were then incubated in secondary antibody (1:10,000) in 1% BSA in TBS-T for 1 hour at RT. The following antibodies were used: pTRS1 (1:100; (43)), His (1:1,000; CST #2365S), PKR
Total (1:1,000; Santa Cruz sc-707), phospho-PKR (Thr446) (1:1,000; Abcam ab32036),
Myc (1:1,000; CST #2276S), Flag (1:1,000; Sigma F3165), eIF2α (1:1,000; CST
#9722S), phospho-eIF2α (Ser51) (1:1,000; CST #3398S), HRI (1:1,000; Abcam ab28530), Tubulin (1:20,000; Sigma T6199).
Immunoprecipitation. HEK293T cells were transfected with 5 µg of the indicated plasmid for a total of 36 hours. Where indicated, doxycycline (1 µg/ml) was added 18 hours prior to harvest to induce pTRS1 expression. Cells were lysed in 1 ml of
PKR lysis buffer for 15 min on ice. Insoluble debris was removed by centrifugation for
10 min at 21,000xg at 4oC. 100 µl of protein A/G sepharose beads (Santa Cruz) in PKR lysis buffer (20 µl packed bead volume) was added to each sample and nutated at 4oC for 30 minutes to pre-clear the lysate. Where indicated, 2,000 units of microccocal
75 nuclease (NEB) and 5 µl of 0.5M CaCl2 were added to each sample prior to the pre- clear step, and incubated at 37oC for 30 min to digest nucleic acid species. The beads were then removed and the lysate was transferred to a new tube containing 1 µg of the indicated antibody, or 1 µg of either mouse IgG or tubulin antibody as a negative control. Samples were nutated with antibody at 4oC for 4 hours whereupon 100 µl of protein A/G beads in PKR lysis buffer (20 µl packed bead volume) was added to each sample. For the αTRS1 immunoprecipitation, antibodies were pre-conjugated to protein
A/G beads overnight before addition to sample lysates. Samples were nutated for 90 minutes after the addition of the beads. The beads were collected by brief centrifugation and washed three times with 1 ml of PKR lysis buffer before addition of 30 µl of 1X sample buffer (0.1 M Tris-HCl, pH 6.8; 6% glycerol; 2% SDS; 0.1 M DTT; 0.002% bromophenol blue). The samples were boiled at 95oC for 10 minutes, briefly centrifuged at 21,000xg and stored at -20oC prior to analysis by Western blot.
For the glutathione sepharose pull-down, HEK293T cells were co-transfected with 2.5 µg of inducible GST-PKR kinase domain (GST-KD) expression plasmid and 2.5
µg of either TRS1 or TRS1mut1 expression plasmid for 36 hours. GST-KD expression was induced by addition of 1 µg/ml doxycycline 18 hours before harvest. Upon harvest, cells were washed once with PBS, pelleted and lysed in 500 µl Wash buffer (20 mM
Hepes, pH 7.0; 100 mM NaCl; 1 mM DTT; 1% glycerol) for 15 min on ice. Insoluble debris was removed by centrifugation at 21,000xg for 10 min at 4oC. 100 µl of washed glutathione sepharose 4B (GE) (15 µl packed sepharose volume) was added to each sample. Samples were nutated at RT for 1 hour, and then washed three times with 1 ml
76 of Wash buffer. 30 µl of 1X sample buffer was added to the beads, whereupon samples were boiled, briefly centrifuged and stored at -20oC until analysis by Western blot.
PKR Kinase assay. HEK293T cells were transfected with 5 µg of the indicated plasmids for 36 hours. For co-transfections, 2.5 µg of each plasmid was transfected.
Immunoprecipitation using αMyc or mouse IgG was performed as described above.
After washing in PKR lysis buffer, beads were re-suspended in 25 µl of 1X kinase reaction buffer (40 mM Tris, pH 7.5; 20 mM MgCl2; 0.1 mg/ml BSA; 500 µM ATP
(Promega)) containing 7.5 µg of histone H2A peptide (NEB) as a substrate. The kinase reaction was run for 30 min at RT. The ADP-GloTM kinase assay (Promega) was then used to determine PKR kinase activity. Briefly, 25 µl of ADP-Glo reagent was added to each kinase reaction and incubated at RT for 40 min. 50 µl of Kinase Detection Reagent was added to each sample and incubated at RT for an additional 40 minutes, whereupon the luminescence of each sample was read using a luminometer (Molecular
Devices). Fold change in luminescence compared to the IgG control immunoprecipitation was calculated and statistical significance was determined using an unpaired students’ t-test.
RESULTS
pTRS1 inhibits PKR activation by dsRNA. We began by examining the ability of pTRS1 to inhibit PKR activation in transfected cells. Increasing amounts of a synthetic dsRNA (poly(I:C)) were added to lysates of cells transfected with either a GFP or pTRS1 expression plasmid, and PKR activation was measured by Western blot. PKR is activated by autophosphorylation on Thr446 (T446) and Thr451 (T451), although
77 phosphorylation of T451 is dependent upon T446 phosphorylation (23, 51). We therefore used PKR T446 phosphorylation as a marker of PKR activation. Low levels of phosphorylated PKR were present in the absence of poly(I:C) in cells expressing GFP, likely due to stress from transfection. Poly(I:C) activated PKR in cells expressing GFP
(Fig 1A). In contrast, the presence of pTRS1 completely suppressed PKR activation, even at the highest doses of poly(I:C). Thus pTRS1 is sufficient to prevent PKR activation by dsRNA in cell culture.
We next examined the regions of pTRS1 necessary for PKR antagonism. In these experiments we used a series of plasmids expressing amino or carboxyl terminal truncations of pTRS1, as they have previously been characterized in vitro (37, 39)(Fig
1B, Table 2) and measured PKR T446 phosphorylation after poly(I:C) treatment (Fig
1C). It is important to note that truncation mutants may unintentionally affect protein folding, and thus negative results using these mutants should be interpreted with caution. As before, full-length pTRS1 blocked activation of PKR by poly(I:C) treatment.
However, pTRS1 lacking its amino terminal 240 or 390 amino acids (240-795 and 390-
795, respectively), which contain the dsRNA binding domain, did not inhibit PKR activation. Similarly, a pTRS1 mutant lacking the final 116 amino acids (1-679), containing the PKR binding domain, also failed to antagonize PKR. Curiously, the removal of an additional 180 amino acids from the C-terminus (1-550) restored PKR inhibition, but further deletion (1-370) again led to an inability to inhibit PKR activation.
Purified pTRS1 and PKR directly interact in vitro, and deletion or mutation of the
PKR binding domain ablates the ability of pTRS1 to inhibit PKR during HCMV infection
(6, 7, 37). These results suggest that pTRS1 interaction with PKR is critical for PKR
78 antagonism. We therefore tested the ability of each pTRS1 mutant to bind PKR in transfected cells (Fig 1D). Cells were transfected with vectors expressing pTRS1 mutants together with kinase-dead PKR (K296R), and the presence of pTRS1 in PKR- specific immune complexes was determined by Western blot. Only full-length pTRS1 bound to PKR. None of the pTRS1 truncation mutants co-purified with PKR, even pTRS1 240-795 and pTRS1 390-795, which both contain the previously described PKR binding domain. Interestingly, pTRS1 1-550 inhibited PKR activation (Fig 1C), but did not bind PKR. These data suggest that pTRS1 binding to PKR is sufficient, but not necessary, to prevent PKR activation by dsRNA. However as noted above, negative results in this assay should be interpreted cautiously due to the potential effect of the truncations on protein folding.
Relative role of dsRNA and PKR binding by pTRS1 in PKR antagonism.
Amino acids 84-246 of pTRS1 contain a noncanonical double stranded RNA binding domain (RBD) that allows purified pTRS1 to competitively inhibit the binding of purified
PKR to dsRNA (39). This function suggests that in addition to inhibiting PKR via a direct interaction, pTRS1 may also inhibit PKR by preventing recognition of its dsRNA ligands
(38, 39). Consistent with this idea, the pTRS1 1-550 mutant, which lacks the PKR binding domain, did not co-purify with PKR but could still prevent PKR activation in response to dsRNA treatment. To determine if pTRS1 1-550 inhibits PKR by binding to dsRNA, we introduced three amino acid changes in the dsRBD of pTRS1 1-550 (1-
550triple) that prevent binding of purified pTRS1 to dsRNA (38). We then measured the ability of this mutant to prevent poly(I:C)-induced PKR activation (Fig 2A). pTRS1 1-
550triple could not prevent PKR activation in response to poly(I:C) treatment. Thus in
79 the context of the truncated pTRS1 protein (1-550), RNA binding is necessary for PKR antagonism.
In contrast we found that pTRS1 could still inhibit PKR activation when the same dsRBD mutations were introduced into full-length pTRS1 (pTRS1triple) (Fig 2A). We reasoned this could be due to pTRS1 binding to PKR through the previously defined
PKR binding domain (PBD). We therefore mutated amino acids in pTRS1 necessary for binding to PKR (7) either alone (pTRS1mut1) or in combination with the dsRBD mutations (pTRS1mut1.triple). Mutation of the pTRS1 PBD ablated both binding to PKR
(Fig 2C) and the ability to inhibit poly(I:C)-induced PKR activation, even when the dsRBD was intact (Fig 2B), suggesting that the primary mechanism of PKR antagonism by full-length pTRS1 is mediated through interactions with PKR. To further explore this hypothesis, we tested if pTRS1 associated with PKR via an RNA intermediate (Fig 2D).
Micrococcal nuclease treatment had no effect on pTRS1 binding to PKR. We also demonstrated that pTRS1 interacts with PKR in HCMV infected cells, which had not been previously demonstrated (Fig 2E). Together these data suggest that pTRS1 has two distinct mechanisms to inhibit PKR activation. In some circumstances pTRS1 may prevent PKR activation by sequestering dsRNA from PKR, independent of binding to
PKR. However, at least in this system, the primary mechanism of PKR inhibition by full- length pTRS1 requires an RNA-independent interaction with PKR.
pTRS1 blocks PKR activation at a step after PKR dimerization. Binding to dsRNA induces PKR dimerization, which is necessary for PKR activation (3, 52). We thus hypothesized that pTRS1 might block PKR activation by preventing PKR dimerization. To test this hypothesis, cells expressing pTRS1 were co-transfected with
80 plasmids encoding myc- and flag-tagged PKR K296R. Catalytically inactive PKR was used in these experiments to limit toxicity. Poly(I:C) was added to cell lysates to induce
PKR dimerization, and the ability of flag-tagged PKR to co-purify with myc-tagged PKR was measured by immunoprecipitation followed by Western blot (Fig 3). As expected, flag-tagged PKR co-purified with myc-tagged PKR in the absence of pTRS1. Similar amounts of flag-PKR associated with myc-PKR in the presence of pTRS1, suggesting that pTRS1 does not affect PKR homodimer formation. In addition, we detected pTRS1 in the myc-PKR immune complexes suggesting that pTRS1 might bind to PKR dimers.
These data show that pTRS1 does not inhibit PKR activation by preventing PKR dimerization and suggest that pTRS1 may in fact bind to PKR dimers.
To further test if pTRS1 prevents PKR activation at a step after dimerization, we determined if pTRS1 could bind and inhibit a constitutively dimerized, active PKR. The glutathione-S-transferase (GST) protein constitutively forms dimers; replacing the amino terminal dsRBD of PKR with GST results in a constitutively dimerized, and thus constitutively active, PKR kinase domain (GST-KD) (53). We asked if pTRS1 co-purified with GST-KD to determine if pTRS1 could bind the dimerized PKR kinase domain (Fig
4A). As long-term expression of GST-KD is toxic, a doxycycline-inducible promoter was used to regulate GST-KD expression. Wild type pTRS1 co-purified with glutathione resin from cells expressing GST-KD, while the PKR binding domain mutant
(pTRS1mut1) did not. We next determined if pTRS1 could inhibit the activity of a constitutively dimerized PKR KD by measuring the effect of pTRS1 expression on GST-
KD autophosphorylation (Fig 4B). pTRS1 expression decreased GST-KD autophosphorylation as compared to GFP and pTRS1mut1 controls. Together these
81 results suggest that pTRS1 binds and inhibits dimerized PKR, and that pTRS1 binds
PKR through an interaction with the PKR KD.
Amino acids in the PKR eIF2α contact site are important for pTRS1 binding.
Previous studies identified key amino acids in the αG helix of the PKR kinase domain that are necessary for binding of PKR to its substrate eIF2α (Fig 5A)(3, 4). To determine if pTRS1 interacted with the PKR eIF2α binding site, we examined the ability of pTRS1 to bind and inhibit previously characterized PKR mutants with single amino acids changes in the αG helix (Table 2). Mutation of PKR residues Thr487 and Phe495 to alanine and arginine (T487A and F495R, respectively) ablate PKR phosphorylation of eIF2α, but do not affect PKR catalytic activity or autophosphorylation (3). Two amino acids within the PKR αG helix, T487 and Glu490 (E490), are conserved in all eIF2α kinases (4). Therefore, we generated plasmids expressing PKR T487A, F495R or
E490A, and measured their binding to pTRS1 (Fig 5B). While pTRS1 efficiently co- purified with wild type PKR, very low levels of pTRS1 were recovered with the T487A and E490A PKR mutants, which were only visible after overexposure of the blot. pTRS1 co-purified with the PKR F495R mutant, but at reduced levels. pTRS1 also maintained the ability to bind a PKR mutant in which the activating phosphorylation site had been mutated to alanine (T446A) (Fig 5B). These data suggest that amino acids in the PKR eIF2α contact site are important for pTRS1 binding and that PKR activation is not necessary for its interaction with pTRS1.
We next tested the ability of pTRS1 to prevent activation of each of the above
PKR mutants. pTRS1 and the PKR mutants were expressed in PKR-deficient cells, and
PKR autophosphorylation was measured after poly(I:C) transfection (Fig 5C).
82 Consistent with previous results, poly(I:C) induced the autophosphorylation of each
PKR mutant in control cells expressing GFP, demonstrating that each PKR mutant was catalytically active (4, 27, 54). Though PKR F495 was activated to lower levels than wild type PKR in control cells, pTRS1 prevented PKR F495R autophosphorylation. In contrast the PKR T487A and E490A mutants were resistant to inhibition by pTRS1, consistent with the inability of pTRS1 to bind these mutants. From these data, we conclude that binding of pTRS1 to the PKR αG helix is necessary for inhibition of PKR activation.
The vaccinia virus PKR antagonist, K3L, also binds PKR at an eIF2α contact residue within the αG helix, Asp486 (D486). Recently, PKR mutants resistant to K3L binding and inhibition were identified (Fig 5A) (27). To determine if pTRS1 bound PKR in a structurally similar manner as K3L, we tested the ability of pTRS1 to bind and inhibit three PKR mutants (E375V, A473T and D486V) that escape inhibition by K3L (Fig 5D).
We found that pTRS1 efficiently bound both PKR E375V and A473T, but showed decreased binding to PKR D486V. pTRS1 prevented autophosphorylation of PKR
E375V and A473T and reduced autophosphorylation of PKR D486V (Fig 5E). These data suggest that while pTRS1 and K3L interact with the same region of PKR, these interactions occur in a structurally distinct manner.
pTRS1 inhibits PKR kinase activity. The inhibition of PKR autophosphorylation by pTRS1 at a step after PKR dimerization suggested that pTRS1 might inhibit PKR kinase activity. To test this hypothesis we measured PKR kinase activity in the presence of pTRS1 (Fig 6). Cells were transfected with a myc-PKR expression plasmid together with plasmids expressing the control protein GFP, wild type pTRS1, or pTRS1mut1,
83 which cannot bind PKR. We then measured PKR kinase activity in myc-specific immune complexes using a peptide from histone H2A as a substrate. The histone H2A peptide has previously been shown to be a substrate for PKR in vitro (3). PKR kinase activity was significantly lower in pTRS1 expressing cells as compared to control cells expressing GFP or cells expressing kinase-dead PKR K296R. The pTRS1mut1 protein did not inhibit PKR kinase activity, consistent with its inability to bind PKR and prevent
PKR autophosphorylation. In fact in some experiments slightly elevated PKR kinase activity was found in cells expressing pTRS1mut1, though the difference was not statistically significant. Together with the above data, these results suggest that pTRS1 binding to the PKR kinase domain inhibits PKR kinase activity, preventing activating autophosphorylation of PKR and subsequent phosphorylation of its substrate eIF2α.
pTRS1 binds and inhibits an additional eIF2α kinase. Our results show that pTRS1 binds and inhibits PKR through interactions with amino acids in the PKR eIF2α contact site. Interestingly, two of the amino acids in the PKR αG helix important for pTRS1 binding are conserved in other eIF2α kinases. To further test the hypothesis that interaction with the PKR eIF2α contact site is critical for pTRS1 binding, we determined if pTRS1 could also bind an additional eIF2α kinase, HRI. We found that
HRI co-purified with pTRS1 in transfected 293T cells (Fig 7A) and in HCMV infected fibroblasts (Fig 7B). Treating cells with arsenite leads to an accumulation of reactive oxygen species (ROS), which activates the HRI kinase, resulting in eIF2α phosphorylation (50). To determine if pTRS1 could prevent HRI-dependent eIF2α phosphorylation, we measured eIF2α phosphorylation after arsenite treatment in PKR- deficient cells (Fig 7C). Arsenite induced a dose-dependent increase in eIF2α
84 phosphorylation, which was decreased in cells expressing pTRS1. This result is consistent with our previous study showing that pTRS1 inhibits stress granule formation in PKR-deficient cells treated with arsenite (6). As HRI activation does not require an interaction with dsRNA, these data further confirm that pTRS1 can inhibit eIF2α kinases independent of its ability to bind RNA. In addition these data further support the conclusion that residues in the eIF2α contact site are critical for pTRS1 interaction with
PKR, and may mediate interactions with additional eIF2α kinases.
DISCUSSION
In this study we sought to examine the mechanism by which HCMV pTRS1 inhibits the antiviral kinase PKR. Our results suggest that pTRS1 employs two mechanisms to block PKR activation. In isolation, the amino terminal 550 amino acids of pTRS1, which contain the RNA binding domain, can prevent PKR activation independent of binding to PKR (Fig 1). However in the context of full-length pTRS1, mutations that disrupt pTRS1 RNA binding do not affect its ability to prevent PKR activation (Fig 2). Rather, binding of pTRS1 to PKR appears to be essential for PKR antagonism. pTRS1 did not block PKR dimerization (Fig 3), but instead could bind and inhibit a constitutively dimerized PKR kinase domain (Fig 4). pTRS1 binding to PKR inhibits PKR kinase activity (Fig 6) and is mediated through interactions with conserved amino acids in the PKR eIF2α contact site. Mutation of these residues rendered PKR resistant to pTRS1 antagonism (Fig 5). Consistent with this finding, pTRS1 interacted with an additional eIF2α kinase, HRI, and limited HRI activation in PKR-deficient cells
(Fig 7). Together our data suggest that inhibition of PKR kinase activity, mediated by
85 pTRS1 binding to conserved residues within the PKR eIF2α contact site, is the dominant mechanism of PKR antagonism.
Both pTRS1 and pIRS1 contain a noncanonical RNA binding domain in their conserved N-terminus. Although the affinity of pTRS1 for dsRNA is lower than that of
PKR, the higher abundance of pTRS1 and pIRS1 in relation to PKR during infection may allow for competition for dsRNA binding (38, 39). Our data show that in some contexts, dsRNA binding by pTRS1 contributes to PKR antagonism. pTRS1 1-550 did not bind PKR (Fig 1D) but maintained the ability to inhibit PKR activation in response to poly(I:C) treatment (Fig 1C). Mutations within pTRS1 1-550 that disrupt dsRNA binding
(1-550triple) ablated the ability of pTRS1 1-550 to inhibit PKR (Fig 2A) suggesting that in the context of the truncated protein, dsRNA binding is necessary for PKR inhibition.
Interestingly, this truncation mutant comprises the conserved region between pTRS1 and pIRS1 suggesting that dsRNA binding by both proteins may contribute to PKR antagonism during infection, although the function of dsRNA binding by pTRS1/pIRS1 during infection has not been examined. It is also possible that pTRS1 1-550 has a higher binding affinity for dsRNA than full-length pTRS1 allowing for efficient binding and sequestration of poly(I:C) in vitro. Nevertheless, these data suggest that in some contexts dsRNA binding by pTRS1 contributes to PKR antagonism.
While dsRNA binding was necessary for the pTRS1 1-550 mutant to inhibit PKR, the ability of full-length pTRS1 to bind dsRNA was dispensable for PKR antagonism (Fig
2B). Consistent with this result, pTRS1 binding to PKR was not dependent on an RNA intermediate (Fig 2D) and pTRS1 inhibited the activation of a constitutively active PKR kinase domain lacking its dsRNA binding domains (Fig 4B). Instead, we found that an
86 interaction between PKR and full-length pTRS1 was necessary for PKR antagonism
(Fig 2). Thus, at least in this system, pTRS1 binding to PKR, rather than competition for dsRNA ligands, is the primary mechanism by which pTRS1 prevents PKR activation.
This is consistent with previous studies showing that the pTRS1 PKR binding domain is necessary for PKR inhibition during HCMV infection in primary human fibroblasts (7,
35). However, as discussed above, our results suggest that dsRNA binding by pTRS1 does play a role in PKR inhibition. Previous studies have shown that a pTRS1 mutant incapable of binding dsRNA (pTRS1triple) did not rescue replication of a vaccinia virus lacking its PKR antagonist, even though pTRS1triple maintained the ability to bind PKR
(38). While dsRNA accumulates in HCMV infected cells (18), the signals driving PKR activation during HCMV infection are unknown. Perhaps these two functions of pTRS1, binding to dsRNA and PKR, are most important for PKR inhibition in different settings, or in response to different PKR activating signals.
We also identified the domain of PKR necessary for its association with pTRS1.
Our finding that pTRS1 could bind and inhibit a constitutively active PKR kinase domain
(GST-KD) suggested that an interaction between pTRS1 and the PKR kinase domain is necessary for PKR antagonism (Fig 4A,B). Using a series of previously described PKR point mutants that retain PKR catalytic activity (3, 4, 27), we found pTRS1 interacts with specific amino acids in the αG helix of the PKR kinase domain that are necessary for binding to its substrate eIF2α (Fig 5). pTRS1 did not co-purify with PKR mutants containing αG helix mutations (PKR T487A or E490A), and weakly interacted with an additional αG helix mutant, PKR D486V (Fig 5B, D). The loss of binding correlated with a decrease in the ability of pTRS1 to inhibit activation of these PKR mutants (Fig 5C,E).
87 PKR T487A and E490A were completely resistant to inhibition by pTRS1, while D486V was partially resistant. In contrast, mutations in PKR residues more distal to the eIF2α contact site (E375V and A473T) had a minimal effect on pTRS1 binding, and pTRS1 prevented activation of both mutants in response to poly(I:C) treatment (Fig 5D,E). Thus our data suggest that inhibition of PKR by pTRS1 requires an interaction with the eIF2α contact site in the PKR kinase domain.
These results were reminiscent of PKR inhibition by the vaccinia virus K3L protein. K3L binds the PKR αG helix and functions as a PKR pseudosubstrate (26, 27,
55). Mutation of PKR E375 or A473 to valine and threonine, respectively, confers resistance of PKR to K3L inhibition (27). We found that pTRS1 bound both the E375 and A473 PKR mutants (Fig 5D), and prevented their autophosphorylation in response to poly(I:C) treatment (Fig 5E). These results are again consistent with a model in which pTRS1 binding mediates PKR antagonism. In addition, these data suggest that pTRS1 inhibition of PKR activation occurs in a manner structurally distinct from that of other viral PKR antagonists.
In addition to a distinct binding modality, we found that pTRS1 also differs from
K3L in its mode of PKR inhibition. K3L does not inhibit PKR catalytic activity, but rather prevents phosphorylation of eIF2α by acting as a PKR pseudosubstrate (26, 55). Our data show that pTRS1 binds PKR and inhibits PKR kinase activity (Fig 6). Importantly, the kinase assay used measures the conversion of ATP to AMP as a result of kinase activity. In the presence of pTRS1, AMP accumulation is reduced, indicating a loss of
PKR catalytic activity. Thus, unlike K3L, pTRS1 does not act as a PKR pseudosubstrate, but rather inhibits PKR kinase activity towards the histone H2A
88 substrate. Together these results suggest that pTRS1 binds the PKR kinase domain and inhibits its catalytic activity.
When considered in sum, our data suggests that pTRS1 inhibits PKR activation and subsequent eIF2α phosphorylation at multiple levels. In some scenarios pTRS1 may sequester dsRNA ligands to prevent PKR dimerization and activation. In addition, we find that pTRS1 inhibits PKR at a step after dimerization through an interaction with the αG helix of the PKR kinase domain. pTRS1 binding to the kinase domain inhibits
PKR catalytic activity, thus preventing autophosphorylation of dimerized PKR and subsequent phosphorylation of eIF2α. While not addressed herein, the fact that pTRS1 binds to PKR residues necessary for its interaction with eIF2α suggests that pTRS1 could also prevent eIF2α recognition by PKR. Thus inhibition of PKR kinase activity by pTRS1 ensures that eIF2α is not phosphorylated, even when PKR has bound activating dsRNA ligands.
As the two residues in PKR recognized by pTRS1 are conserved across all eIF2α kinases, we reasoned that pTRS1 might bind and inhibit additional eIF2α kinases. We previously found that pTRS1 can prevent stress granule formation in response to arsenite treatment in PKR-deficient cells, suggesting that pTRS1 has the ability to block HRI-dependent eIF2α phosphorylation (50). Here we found that pTRS1 binds HRI (Fig 7A and B) and limits eIF2α phosphorylation in response to arsenite treatment in PKR-deficient cells (Fig 7C). These data further support the conclusion that residues in the PKR αG helix necessary for eIF2α binding are critical for interaction of pTRS1 with PKR. In addition these data suggest that pTRS1 could have a more general role in suppressing cellular stress responses during HCMV infection, allowing for
89 efficient cellular and viral protein synthesis in the face of robust cellular stress induced by infection.
CONCLUSIONS
We wish to thank members of the Moorman, Heise, de Silva, and de Paris labs as well as the members of the UNC Virology community for helpful discussions and input. We also thank Westefson Sanders for continued support and Anne Beall for her valuable feedback. This work was supported by NIH grant AI03311 to N.J.M. Additional support was provided by the North Carolina University Cancer Research Fund, and an award from the UNC Virology Training grant (T32 AI07419 to B.Z.).
90 FIGURES
Figure 2.1. Domains of pTRS1 necessary for PKR binding and inhibition. (A)
Lysates from HEK293T cells transfected with either a GFP or pTRS1 expression vector were treated with poly(I:C) to induce PKR activation. Levels of phosphorylated PKR
(T446) were measured by Western blot. (B) Schematic showing location of pTRS1 functional domains and pTRS1 truncation mutants (RBD = dsRNA binding domain; PBD
= PKR binding domain) Areas shaded grey are conserved between pTRS1 and pIRS1.
(C) HEK293T cells were transfected with GFP, full-length TRS1 or the indicated TRS1 mutants. Lysates were treated with poly(I:C) and PKR phosphorylation (T446) was measured by Western blot. (D) HEK293T cells were co-transfected with myc-tagged kinase-dead PKR (K296R) and either full-length TRS1 or the indicated TRS1 mutants.
91 Cell lysates were immunoprecipitated with αMyc (PKR) or IgG (specificity control) antibodies. The presence of pTRS1 in the immune complexes was determined by
Western blot. Representative results from one of at least three independent experiments are shown for each panel.
Figure 2.2. PKR binding by full-length pTRS1 is necessary for inhibition of PKR activation. (A) Lysates from HEK293T cells transfected with full-length TRS1 (WT) or the indicated TRS1 mutants. pTRS1 1-550 (1-550) expresses only the first 550 amino acids of pTRS1. pTRS1triple (Triple) contains mutations that disrupt dsRNA binding. pTRS1.1-550.triple (1-550 Triple) expresses the 1-550 truncation mutant containing mutations in the dsRNA binding domain. Lysates were incubated with poly(I:C) to activate PKR. PKR phosphorylation (T446) was examined by Western blot. (B) Same as in A, except with different TRS1 mutants. pTRS1mut1 (Mut1) contains mutations that disrupt PKR binding. (C) HEK293T cells were transfected with myc-tagged kinase dead
92 PKR (K296R) together with full-length TRS1 or the indicated TRS1 mutants. pTRS1mut1.triple (Mut1-triple) contains point mutations in both the PKR and dsRNA binding domainsCell lysates were immunoprecipitated with αMyc (PKR) or IgG
(specificity control) antibodies. The presence of pTRS1 in the immune complexes was determined by Western blot. (D) As in C, except lysates were treated with micrococcal nuclease prior to immunoprecipitation. (E) MRC-5 fibroblasts were infected with HCMV at an MOI of 3 for 24 hours. The presence of PKR in pTRS1 or IE1 specific immune complexes was determined by Western blot. Representative results from one of at least three independent experiments are shown for each panel.
Figure 2.3. pTRS1 does not prevent PKR dimerization. HEK293T cells expressing pTRS1 from an inducible promoter (293T-pTRS1i) were co-transfected with flag- and myc-tagged PKR K296R expression vectors. pTRS1 expression was induced with doxycycline treatment. Cell lysates were treated with poly(I:C) and PKR was immunoprecipitated with αMyc antibody. The interaction of flag- and myc-tagged PKR was examined by Western blot. Representative results from one of at least three independent experiments are shown.
93
Figure 2.4. pTRS1 binds and inhibits a constitutively active PKR kinase domain.
(A) HEK293T cells were co-transfected with a doxycycline-inducible GST-tagged PKR kinase domain (GST-KD) expression vector and an expression vector encoding either pTRS1 or pTRS1 Mut1. GST-KD expression was induced with doxycycline treatment.
Glutathione resin was used to capture the GST-KD, and the co-purification of pTRS1 or pTRS1mut1 (Mut1) was examined by Western blot. (B) Cells were transfected as in (A) with the addition of cells co-transfected with a GFP expression vector and the GST-KD expression vector. GST-KD expression was induced with doxycycline treatment, and levels of GST-KD autophosphorylation (T446) were determined by Western blot analysis. Representative results from one of at least three independent experiments are shown for each panel.
94
Figure 2.5. pTRS1 interacts with the PKR kinase domain to prevent PKR autophosphorylation. (A) PyMOL image of the PKR kinase domain and eIF2α (PDB:
2A1A). The PKR αG helix is represented in cyan. Amino acids required for eIF2a binding are shown in magenta, while mutations that confer resistance to inhibition by
K3L are shown in blue. (B) 293T-pTRS1i cells were transfected with myc-tagged PKR
K296R or the indicated myc-tagged PKR mutants. pTRS1 expression was induced with doxycycline treatment before lysates were immunoprecipitated with αMyc (PKR) or IgG
(specificity control) antibodies. The presence of pTRS1 in the immune complexes was determined by Western blot. (C) PKR-deficient HeLa cells (PKR KO) were transfected with the indicated PKR mutants together with vectors expressing either GFP or TRS1.
Cells were transfected with poly(I:C) to activate PKR and PKR autophosphorylation
(T446) was measured by Western blot. (D and E) As in B and C, except cells were
95 transfected with the indicated PKR mutants. Representative results from one of at least three independent experiments are shown for each panel.
Figure 2.6. pTRS1 inhibits PKR kinase activity. HEK293T cells were transfected with the indicated expression vectors (PKR K296R = kinase dead PKR; Mut1 = pTRS1 containing PKR binding domain mutations). PKR was immunoprecipitated with αMyc antibody and re-suspended in kinase buffer. The kinase reaction was performed “on bead” as described in Materials and Methods, using histone H2A peptide as a substrate. PKR kinase activity was measured using the ADP-Glo™ kinase assay. The amount of kinase activity is normalized to the background activity associated with non- specific (IgG) immunoprecipitates, which was set to one. Statistical significance was determined using an unpaired Student’s t-test (n=3; * = p<0.05).
96
Figure 2.7. pTRS1 binds HRI and limits its activation. (A) 293T-pTRS1i cells were transfected with myc-tagged HRI, and pTRS1 expression was induced with doxycycline.
Lysates were immunoprecipitated with αMyc (PKR) or control antibody (αTubulin), and the presence of HRI the immune complexes was determined by Western blot. (B) MRC-
5 fibroblasts were infected with HCMV at an MOI of 3 for 72 hours. The amount of HRI co-purifying with pTRS1 specific immune complexes or with beads alone (control) was determined by Western blot. (C) PKR KO-pTRS1i cells were treated with doxycycline to induce pTRS1 expression and subsequently treated with increasing concentrations of arsenite (ARS) to activate HRI. Levels of eIF2α phosphorylation were measured by
Western blot. The ratio of phosphorylated eF2α to total eIF2a levels (P/total) was calculated by densitometry. Representative results from one of at least three independent experiments are shown for each panel.
97 TABLES
Table 2.1. Primer and gBlock Gene Fragment Sequences
Primer Name Sequence (5' à 3') Triple F GCGGCCATGGCCGCCTGGTCGCAGCGCGACGCGGGCAC GACCAGGCGGCCATGGCCGCGCCCGTGCTGTTGGCCAGA Triple R A CTTTGGCCCGGGCCGCCGCCGATTGGAAACCGCCACGTCT Mut1 F CCCTGGGGAAG CAATCGGCGGCGGCCCGGGCCAAAGCACGTCCCAAACTG Mut1 R GCTTGGGGAGTC TAAGCTTGGTACCGAGCTCGCCACCATGGACTACAAGGAC PKRflag F GACGACGACAAGGCTGGTGATCTTTCAGCAGG CGCGGGCCCTCTAGACTCGATCAACATGTGTGTCGTTCATT PKR R TT TAAGCTTGGTACCGAGCTCGCCACCATGGAACAAAAACTCA PKRmyc F TCTCAGAAGAGGATCTGGCTGGTGATCTTTCAGCAGG AGGGACCTTGGTACAATGGATTGAAAAAAGAAGAGGCGAG PKR_E375V F AAAC CAATCCATTGTACCAAGGTCCCTTTATCACAGAATTCCATTT PKR_E375V R GG GTGGACCTCTACACTTTGGGGCTAATTCTTGCTGAACTTCTT PKR_A473T F C PKR_A473T R CCCAAAGTGTAGAGGTCCACTTCCTTTCCATAGTCTTGCG PKR_T446A F GATGGAAAGCGAGCAAGGAGTAAGGGAACTTTGCGATACA PKR_T446A R CTCCTTGCTCGCTTTCCATCATTTTTCAGAGATGTTACAAG PKR_D486V F CTTCATGTATGTGTCACTGCTTTTGAAACATCAAAGTT PKR_D486V R AGCAGTGACACATACATGAAGAAGTTCAGCAAGAATTAGCC GTATGTGACGCTGCTTTTGAAACATCAAAGTTTTTCACAGAC PKR_T487A F CTACGG GTTTCAAAAGCAGCGTCACATACATGAAGAAGTTCAGCAAG PKR_T487A R AATTAGCC GACACTGCTTTTGCAACATCAAAGTTTTTCACAGACCTACGG PKR_E490A F GATG CTTTGATGTTGCAAAAGCAGTGTCACATACATGAAGAAGTTC PKR_E490A R AG PKR_F495R F CAAAGTTTCGCACAGACCTACGGGATGGCATCATCTCAG
98 PKR_F495R R TAGGTCTGTGCGAAACTTTGATGTTTCAAAAGCAGTGTCAC CAGATCGCCTGGAGAATTGGCTAGCgccaccATGTCCCCTAT GST_pCW F ACTAGGTTATTGG GST R TGGAGGATGGTCGCCACCAC KD_GST F GTGGTGGCGACCATCCTCCAACTGTGGACAAGAGGTTTGG KD_GST_pCW CGCAACCCCAACCCCGGATCCTCAACATGTGTGTCGTTCAT R TTTTC CAGATCGCCTGGAGAATTGGCTAGCgccaccATGGCCCAGCG TRS1ind F CAACGGCAT TRS1ind R CGCAACCCCAACCCCGGATCCTTATTGAGCATTGTAATGGT TAAGCTTGGTACCGAGCTCGgccgccaccATGGAACAAAAACT HRImyc F CATCTCAGAAGAGGATCTGCAGGGGGGCAACTCCGGGGTC CGCGGGCCCTCTAGACTCGATCACAATCCCACGCCCCCAT HRImyc R CCTTTC TAAGCTTGGTACCGAGCTCGCCACCATGGCTGGTGATCTTT CAGCAGGTTTCTTCATGGAGGAACTTAATACATACCGTCAG AAGCAGGGAGTAGTACTTAAATATCAAGAACTGCCTAATTC AGGACCTCCACATGATAGGAGGTTTACATTTCAAGTTATAAT AGATGGAAGAGAATTTCCAGAAGGTGAAGGTAGATCAAAGA AGGAAGCAAAAAATGCCGCAGCCAAATTAGCTGTTGAGATA CTTAATAAGGAAAAGAAGGCAGTTAGTCCTTTATTATTGACA ACAACGAATTCTTCAGAAGGATTATCCATGGGGAATTACATA GGCCTTATCAATAGAATTGCCCAGAAGAAAAGACTAACTGT AAATTATGAACAGTGTGCATCGGGGGTGCATGGGCCAGAA GGATTTCATTATAAATGCAAAATGGGACAGAAAGAATATAGT ATTGGTACAGGTTCTACTAAACAGGAAGCAAAACAATTGGC CGCTAAACTTGCATATCTTCAGATATTATCAGAAGAAACCTC AGTGAAATCTGACTACCTGTCCTCTGGTTCTTTTGCTACTAC PKR gBlock GTGTGAGTCCCAAAGCAACTCTTTAGTGACCAGCACACTCG CTTCTGAATCATCATCTGAAGGTGACTTCTCAGCAGATACAT CAGAGATAAATTCTAACAGTGACAGTTTAAACAGTTCTTCGT TGCTTATGAATGGTCTCAGAAATAATCAAAGGAAGGCAAAA AGATCTTTGGCACCCAGATTTGACCTTCCTGACATGAAAGA AACAAAGTATACTGTGGACAAGAGGTTTGGCATGGATTTTA AAGAAATAGAATTAATTGGCTCAGGTGGATTTGGCCAAGTTT TCAAAGCAAAACACAGAATTGACGGAAAGACTTACGTTATTA AACGTGTTAAATATAATAACGAGAAGGCGGAGCGTGAAGTA AAAGCATTGGCAAAACTTGATCATGTAAATATTGTTCACTAC AATGGCTGTTGGGATGGATTTGATTATGATCCTGAGACCAG TGATGATTCTCTTGAGAGCAGTGATTATGATCCTGAGAACA GCAAAAATAGTTCAAGGTCAAAGACTAAGTGCCTTTTCATCC AAATGGAATTCTGTGATAAAGGGACCTTGGAACAATGGATT GAAAAAAGAAGAGGCGAGAAACTAGACAAAGTTTTGGCTTT
99 GGAACTCTTTGAACAAATAACAAAAGGGGTGGATTATATAC ATTCAAAAAAATTAATTCATAGAGATCTTAAGCCAAGTAATAT ATTCTTAGTAGATACAAAACAAGTAAAGATTGGAGACTTTGG ACTTGTAACATCTCTGAAAAATGATGGAAAGCGAACAAGGA GTAAGGGAACTTTGCGATACATGAGCCCAGAACAGATTTCT TCGCAAGACTATGGAAAGGAAGTGGACCTCTACGCTTTGGG GCTAATTCTTGCTGAACTTCTTCATGTATGTGACACTGCTTT TGAAACATCAAAGTTTTTCACAGACCTACGGGATGGCATCA TCTCAGATATATTTGATAAAAAAGAAAAAACTCTTCTACAGA AATTACTCTCAAAGAAACCTGAGGATCGACCTAACACATCT GAAATACTAAGGACCTTGACTGTGTGGAAGAAAAGCCCAGA GAAAAATGAACGACACACATGTTCGAGTCTAGAGGGCCCG CG TAAGCTTGGTACCGAGCTCGCCACCATGGCTGGTGATCTTT CAGCAGGTTTCTTCATGGAGGAACTTAATACATACCGTCAG AAGCAGGGAGTAGTACTTAAATATCAAGAACTGCCTAATTC AGGACCTCCACATGATAGGAGGTTTACATTTCAAGTTATAAT AGATGGAAGAGAATTTCCAGAAGGTGAAGGTAGATCAAAGA AGGAAGCAAAAAATGCCGCAGCCAAATTAGCTGTTGAGATA CTTAATAAGGAAAAGAAGGCAGTTAGTCCTTTATTATTGACA ACAACGAATTCTTCAGAAGGATTATCCATGGGGAATTACATA GGCCTTATCAATAGAATTGCCCAGAAGAAAAGACTAACTGT AAATTATGAACAGTGTGCATCGGGGGTGCATGGGCCAGAA GGATTTCATTATAAATGCAAAATGGGACAGAAAGAATATAGT ATTGGTACAGGTTCTACTAAACAGGAAGCAAAACAATTGGC CGCTAAACTTGCATATCTTCAGATATTATCAGAAGAAACCTC AGTGAAATCTGACTACCTGTCCTCTGGTTCTTTTGCTACTAC GTGTGAGTCCCAAAGCAACTCTTTAGTGACCAGCACACTCG PKR K296R CTTCTGAATCATCATCTGAAGGTGACTTCTCAGCAGATACAT gBlock CAGAGATAAATTCTAACAGTGACAGTTTAAACAGTTCTTCGT TGCTTATGAATGGTCTCAGAAATAATCAAAGGAAGGCAAAA AGATCTTTGGCACCCAGATTTGACCTTCCTGACATGAAAGA AACAAAGTATACTGTGGACAAGAGGTTTGGCATGGATTTTA AAGAAATAGAATTAATTGGCTCAGGTGGATTTGGCCAAGTTT TCAAAGCAAAACACAGAATTGACGGAAAGACTTACGTTATT CGGCGTGTTAAATATAATAACGAGAAGGCGGAGCGTGAAG TAAAAGCATTGGCAAAACTTGATCATGTAAATATTGTTCACT ACAATGGCTGTTGGGATGGATTTGATTATGATCCTGAGACC AGTGATGATTCTCTTGAGAGCAGTGATTATGATCCTGAGAA CAGCAAAAATAGTTCAAGGTCAAAGACTAAGTGCCTTTTCAT CCAAATGGAATTCTGTGATAAAGGGACCTTGGAACAATGGA TTGAAAAAAGAAGAGGCGAGAAACTAGACAAAGTTTTGGCT TTGGAACTCTTTGAACAAATAACAAAAGGGGTGGATTATATA CATTCAAAAAAATTAATTCATAGAGATCTTAAGCCAAGTAAT ATATTCTTAGTAGATACAAAACAAGTAAAGATTGGAGACTTT GGACTTGTAACATCTCTGAAAAATGATGGAAAGCGAACAAG
100 GAGTAAGGGAACTTTGCGATACATGAGCCCAGAACAGATTT CTTCGCAAGACTATGGAAAGGAAGTGGACCTCTACGCTTTG GGGCTAATTCTTGCTGAACTTCTTCATGTATGTGACACTGCT TTTGAAACATCAAAGTTTTTCACAGACCTACGGGATGGCATC ATCTCAGATATATTTGATAAAAAAGAAAAAACTCTTCTACAG AAATTACTCTCAAAGAAACCTGAGGATCGACCTAACACATCT GAAATACTAAGGACCTTGACTGTGTGGAAGAAAAGCCCAGA GAAAAATGAACGACACACATGTTCGAGTCTAGAGGGCCCG CG
Table 2.2. pTRS1 and PKR Mutations and Functional Descriptions
PKR expression vectors Plasmid Mutation Description Ref Name Mutation in activating PKR T446A T446A (23) phosphorylation site PKR E375V E375V Confers resistance to PKR A473T A473T (27) inhibition by VV K3L PKR D486V D486V PKR T487A T487A Disrupts PKR (3) PKR E490A F495R phosphorylation of eIF2α Mutation in conserved PKR F495R E490A (4) eIF2α αG helix residue PKR K296R Kinase dead PKR with K296R FLAG N-terminal flag tag PKR K296R Kinase dead PKR with K296R myc N-terminal myc tag (3) Doxycycline-inducible PKR GST- fusion of GST to the NA KD PKR kinase domain (aa 259-551) HRI expression vector, HRI-myc NA NA C-terminal myc tag pTRS1 expression vectors Plasmid Mutation Description Ref Name Binds both dsRNA and WT Full length pTRS1 (37, 39) PKR
101 C-terminal Contains RNA binding 1-370 truncation mutant domain. Lacks PKR lacking final 425 aa binding domain C-terminal Contains RNA binding 1-550 truncation mutant domain. Lacks PKR lacking final 245 aa binding domain C-terminal Contains RNA binding 1-679 truncation mutant domain. Lacks PKR lacking final 116 aa binding domain N-terminal Lacks RNA binding truncation mutant 240-795 domain. Contains PKR lacking the first 239 binding domain aa N-terminal Lacks RNA binding truncation mutant 390-795 domain. Contains PKR lacking the first 299 binding domain aa Full length pTRS1 containing point mutations in the Lacks ability to bind Triple (38) RNA binding dsRNA domain(R121A,R12 4A,R125A) Full length pTRS1 contatining point mutations in the Mut1 PKR binding Lacks ability to bind PKR (7) domain (R658A, D660,661A, E662A) Full length pTRS1 containing point mutations in the RNA (R121A,R124A,R12 Lacks ability to bind both Triple/Mut1 (7, 38) 5A) and PKR dsRNA and PKR (R658A, D660,661A, E662A) binding domains pTRS1 deletion mutant lacking final Lacks ability to bind both 1-550.Triple 245 aa and (38, 39) dsRNA and PKR containing point mutations in RNA
102 binding domain (R658A, D660,661A, E662A) pCW- Doxycycline-inducible NA NA pTRS1 full length pTRS1
103 REFERENCES
1. García MA, Gil J, Ventoso I, Guerra S, Domingo E, Rivas C, Esteban M. 2006. Impact of protein kinase PKR in cell biology: from antiviral to antiproliferative action. Microbiology and molecular biology reviews: MMBR 70:1032-1060.
2. Barber GN, Wambach M, Wong ML, Dever TE, Hinnebusch AG, Katze MG. 1993. Translational regulation by the interferon-induced double-stranded-RNA- activated 68-kDa protein kinase. Proceedings of the National Academy of Sciences of the United States of America 90:4621-4625.
3. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic Link between PKR Dimerization, Autophosphorylation, and eIF2α Substrate Recognition. Cell 122:901-913.
4. Dar AC, Dever TE, Sicheri F. 2005. Higher-order substrate recognition of eIF2alpha by the RNA-dependent protein kinase PKR. Cell 122:887-900.
5. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. 2001. Tight binding of the phosphorylated alpha subunit of initiation factor 2 (eIF2alpha) to the regulatory subunits of guanine nucleotide exchange factor eIF2B is required for inhibition of translation initiation. Molecular and Cellular Biology 21:5018-5030.
6. Ziehr B, Vincent HA, Moorman NJ. 2016. Human Cytomegalovirus pTRS1 and pIRS1 Antagonize Protein Kinase R To Facilitate Virus Replication. Journal of Virology 90:3839-3848.
7. Braggin JE, Child SJ, Geballe AP. 2016. Essential role of protein kinase R antagonism by TRS1 in human cytomegalovirus replication. Virology 489:75-85.
8. Balachandran S, Kim CN, Yeh WC, Mak TW, Bhalla K, Barber GN. 1998. Activation of the dsRNA-dependent protein kinase, PKR, induces apoptosis through FADD-mediated death signaling. The EMBO Journal 17:6888-6902.
9. Der SD, Yang Y-L, Weissmann C, Williams BRG. 1997. A double-stranded RNA-activated protein kinase-dependent pathway mediating stress-induced
104 apoptosis. Proceedings of the National Academy of Sciences of the United States of America 94:3279-3283.
10. Perry AK, Chen G, Zheng D, Tang H, Cheng G. 2005. The host type I interferon response to viral and bacterial infections. Cell Research 15:407-422.
11. Tallóczy Z, Jiang W, Virgin HW, Leib DA, Scheuner D, Kaufman RJ, Eskelinen E-L, Levine B. 2002. Regulation of starvation- and virus-induced autophagy by the eIF2α kinase signaling pathway. Proceedings of the National Academy of Sciences 99:190-195.
12. Lussignol M, Queval C, Bernet-Camard M-F, Cotte-Laffitte J, Beau I, Codogno P, Esclatine A. 2013. The herpes simplex virus 1 Us11 protein inhibits autophagy through its interaction with the protein kinase PKR. Journal of Virology 87:859-871.
13. Lu B, Nakamura T, Inouye K, Li J, Tang Y, Lundbäck P, Valdes-Ferrer SI, Olofsson PS, Kalb T, Roth J, Zou Y, Erlandsson-Harris H, Yang H, Ting JP- Y, Wang H, Andersson U, Antoine DJ, Chavan SS, Hotamisligil GS, Tracey KJ. 2012. Novel role of PKR in inflammasome activation and HMGB1 release. Nature 488:670-674.
14. Yim HC, Wang D, Yu L, White CL, Faber PW, Williams BR, Sadler AJ. 2016. The kinase activity of PKR represses inflammasome activity. Cell Research 26:367-379.
15. Tu Y-C, Yu C-Y, Liang J-J, Lin E, Liao C-L, Lin Y-L. 2012. Blocking Double- Stranded RNA-Activated Protein Kinase PKR by Japanese Encephalitis Virus Nonstructural Protein 2A. Journal of Virology 86:10347-10358.
16. Heinicke LA, Wong CJ, Lary J, Nallagatla SR, Diegelman-Parente A, Zheng X, Cole JL, Bevilacqua PC. 2009. RNA dimerization promotes PKR dimerization and activation. Journal of Molecular Biology 390:319-338.
17. Brand SR, Kobayashi R, Mathews MB. 1997. The Tat protein of human immunodeficiency virus type 1 is a substrate and inhibitor of the interferon- induced, virally activated protein kinase, PKR. The Journal of Biological Chemistry 272:8388-8395.
105 18. Marshall EE, Bierle CJ, Brune W, Geballe AP. 2009. Essential role for either TRS1 or IRS1 in human cytomegalovirus replication. Journal of Virology 83:4112-4120.
19. Jacquemont B, Roizman B. 1975. RNA synthesis in cells infected with herpes simplex virus. X. Properties of viral symmetric transcripts and of double-stranded RNA prepared from them. Journal of Virology 15:707-713.
20. Anderson E, Quartararo C, Brown RS, Shi Y, Yao X, Cole JL. 2010. Analysis of monomeric and dimeric phosphorylated forms of protein kinase R. Biochemistry 49:1217-1225.
21. Husain B, Hesler S, Cole JL. 2015. Regulation of PKR by RNA: formation of active and inactive dimers. Biochemistry 54:6663-6672.
22. Lemaire PA, Anderson E, Lary J, Cole JL. 2008. Mechanism of PKR Activation by dsRNA. Journal of Molecular Biology 381:351-360.
23. Romano PR, Garcia-Barrio MT, Zhang X, Wang Q, Taylor DR, Zhang F, Herring C, Mathews MB, Qin J, Hinnebusch AG. 1998. Autophosphorylation in the Activation Loop Is Required for Full Kinase Activity In Vivo of Human and Yeast Eukaryotic Initiation Factor 2α Kinases PKR and GCN2. Molecular and Cellular Biology 18:2282-2297.
24. Sciortino MT, Parisi T, Siracusano G, Mastino A, Taddeo B, Roizman B. 2013. The virion host shutoff RNase plays a key role in blocking the activation of protein kinase R in cells infected with herpes simplex virus 1. Journal of Virology 87:3271-3276.
25. Li Y, Zhang C, Chen X, Yu J, Wang Y, Yang Y, Du M, Jin H, Ma Y, He B, Cao Y. 2011. ICP34.5 protein of herpes simplex virus facilitates the initiation of protein translation by bridging eukaryotic initiation factor 2alpha (eIF2alpha) and protein phosphatase 1. The Journal of Biological Chemistry 286:24785-24792.
26. Kawagishi-Kobayashi M, Silverman JB, Ung TL, Dever TE. 1997. Regulation of the protein kinase PKR by the vaccinia virus pseudosubstrate inhibitor K3L is dependent on residues conserved between the K3L protein and the PKR substrate eIF2alpha. Molecular and Cellular Biology 17:4146-4158.
106 27. Seo EJ, Liu F, Kawagishi-Kobayashi M, Ung TL, Cao C, Dar AC, Sicheri F, Dever TE. 2008. Protein kinase PKR mutants resistant to the poxvirus pseudosubstrate K3L protein. Proceedings of the National Academy of Sciences 105:16894-16899.
28. Romano PR, Zhang F, Tan SL, Garcia-Barrio MT, Katze MG, Dever TE, Hinnebusch AG. 1998. Inhibition of double-stranded RNA-dependent protein kinase PKR by vaccinia virus E3: role of complex formation and the E3 N- terminal domain. Molecular and Cellular Biology 18:7304-7316.
29. Thakur M, Seo EJ, Dever TE. 2014. Variola virus E3L Zα domain, but not its Z- DNA binding activity, is required for PKR inhibition. RNA (New York, NY) 20:214- 227.
30. Budt M, Niederstadt L, Valchanova RS, Jonjić S, Brune W. 2009. Specific inhibition of the PKR-mediated antiviral response by the murine cytomegalovirus proteins m142 and m143. Journal of Virology 83:1260-1270.
31. Child SJ, Geballe AP. 2009. Binding and Relocalization of Protein Kinase R by Murine Cytomegalovirus. Journal of Virology 83:1790-1799.
32. Ostermann E, Warnecke G, Waibler Z, Brune W. 2015. Knockout of the host resistance gene Pkr fully restores replication of murine cytomegalovirus (MCMV) m142 and m143 mutants in vivo. Journal of Virology doi:10.1128/JVI.02003- 15:JVI.02003-02015.
33. Valchanova RS, Picard-Maureau M, Budt M, Brune W. 2006. Murine cytomegalovirus m142 and m143 are both required to block protein kinase R- mediated shutdown of protein synthesis. Journal of Virology 80:10181-10190.
34. Li S, Min J-Y, Krug RM, Sen GC. 2006. Binding of the influenza A virus NS1 protein to PKR mediates the inhibition of its activation by either PACT or double- stranded RNA. Virology 349:13-21.
35. Ziehr B, Vincent HA, Moorman NJ. 2016. Human cytomegalovirus pTRS1 and pIRS1 antagonize PKR to facilitate virus replication. Journal of Virology doi:10.1128/JVI.02714-15.
107 36. Child SJ, Hakki M, De Niro KL, Geballe AP. 2004. Evasion of cellular antiviral responses by human cytomegalovirus TRS1 and IRS1. Journal of Virology 78:197-205.
37. Hakki M, Marshall EE, De Niro KL, Geballe AP. 2006. Binding and nuclear relocalization of protein kinase R by human cytomegalovirus TRS1. Journal of Virology 80:11817-11826.
38. Bierle CJ, Semmens KM, Geballe AP. 2013. Double-stranded RNA binding by the human cytomegalovirus PKR antagonist TRS1. Virology 442:28-37.
39. Hakki M, Geballe AP. 2005. Double-stranded RNA binding by human cytomegalovirus pTRS1. Journal of Virology 79:7311-7318.
40. Ziehr B, Lenarcic E, Vincent HA, Cecil C, Garcia B, Shenk T, Moorman NJ. 2015. Human cytomegalovirus TRS1 protein associates with the 7- methylguanosine mRNA cap and facilitates translation. Proteomics 15:1983- 1994.
41. Chaumorcel M, Lussignol M, Mouna L, Cavignac Y, Fahie K, Cotte-Laffitte J, Geballe A, Brune W, Beau I, Codogno P, Esclatine A. 2012. The Human Cytomegalovirus Protein TRS1 Inhibits Autophagy via Its Interaction with Beclin 1. Journal of Virology 86:2571-2584.
42. Mouna L, Hernandez E, Bonte D, Brost R, Amazit L, Delgui LR, Brune W, Geballe AP, Beau I, Esclatine A. 2016. Analysis of the role of autophagy inhibition by two complementary human cytomegalovirus BECN1/Beclin 1- binding proteins. Autophagy 12:327-342.
43. Romanowski MJ, Shenk T. 1997. Characterization of the human cytomegalovirus irs1 and trs1 genes: a second immediate-early transcription unit within irs1 whose product antagonizes transcriptional activation. Journal of Virology 71:1485-1496.
44. Adamo JE, Schröer J, Shenk T. 2004. Human cytomegalovirus TRS1 protein is required for efficient assembly of DNA-containing capsids. Journal of Virology 78:10221-10229.
45. Iskenderian AC, Huang L, Reilly A, Stenberg RM, Anders DG. 1996. Four of eleven loci required for transient complementation of human cytomegalovirus
108 DNA replication cooperate to activate expression of replication genes. Journal of Virology 70:383-392.
46. Pari GS, Anders DG. 1993. Eleven loci encoding trans-acting factors are required for transient complementation of human cytomegalovirus oriLyt- dependent DNA replication. Journal of Virology 67:6979-6988.
47. Pari GS, Kacica MA, Anders DG. 1993. Open reading frames UL44, IRS1/TRS1, and UL36-38 are required for transient complementation of human cytomegalovirus oriLyt-dependent DNA synthesis. Journal of Virology 67:2575- 2582.
48. Strang BL, Geballe AP, Coen DM. 2010. Association of human cytomegalovirus proteins IRS1 and TRS1 with the viral DNA polymerase accessory subunit UL44. Journal of General Virology 91:2167-2175.
49. Strang BL, Bender BJ, Sharma M, Pesola JM, Sanders RL, Spector DH, Coen DM. 2012. A Mutation Deleting Sequences Encoding the Amino Terminus of Human Cytomegalovirus UL84 Impairs Interaction with UL44 and Capsid Localization. Journal of Virology 86:11066-11077.
50. McEwen E, Kedersha N, Song B, Scheuner D, Gilks N, Han A, Chen J-J, Anderson P, Kaufman RJ. 2005. Heme-regulated inhibitor kinase-mediated phosphorylation of eukaryotic translation initiation factor 2 inhibits translation, induces stress granule formation, and mediates survival upon arsenite exposure. The Journal of Biological Chemistry 280:16925-16933.
51. Zhang F, Romano PR, Nagamura-Inoue T, Tian B, Dever TE, Mathews MB, Ozato K, Hinnebusch AG. 2001. Binding of double-stranded RNA to protein kinase PKR is required for dimerization and promotes critical autophosphorylation events in the activation loop. The Journal of Biological Chemistry 276:24946-24958.
52. Dey M, Mann BR, Anshu A, Mannan MA-u. 2014. Activation of protein kinase PKR requires dimerization-induced cis-phosphorylation within the activation loop. The Journal of Biological Chemistry 289:5747-5757.
53. Ung TL, Cao C, Lu J, Ozato K, Dever TE. 2001. Heterologous dimerization domains functionally substitute for the double-stranded RNA binding domains of the kinase PKR. The EMBO journal 20:3728-3737.
109 54. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition. Cell 122:901-913.
55. Dar AC, Sicheri F. 2002. X-Ray Crystal Structure and Functional Analysis of Vaccinia Virus K3L Reveals Molecular Determinants for PKR Subversion and Substrate Recognition. Molecular Cell 10:295-305.
110 CHAPTER 3: HUMAN CYTOMEGALOVIRUS PTRS1 STIMULATES CAP- INDEPENDENT TRANSLATION
OVERVIEW
Human cytomegalovirus (HCMV) significantly manipulates multiple cellular processes, including the control of mRNA translation of host and viral mRNAs. While translation of host and viral mRNAs is thought to occur in a cap-dependent manner,
HCMV mRNAs are efficiently translated when cap-dependent translation is inhibited late in HCMV infection. This suggests that viral mRNAs use a non-canonical translation initiation complex to ensure viral protein synthesis. The viral protein TRS1 (pTRS1) has been shown to associate with the mRNA cap, 40S ribosomal subunits, and double- stranded RNA (dsRNA), and to inhibit the antiviral kinase PKR. Exogenous expression of pTRS1 increases overall levels of protein synthesis and promotes cap-dependent mRNA translation. However, the effect of pTRS1 on cap-independent translation has not been examined. Here we show that pTRS1 enhances cap-independent translation. pTRS1 increases internal ribosome entry site (IRES) activity from both a bicistronic reporter and also a circular RNA reporter, which lacks an mRNA cap and poly(A) tail. pTRS1 expression also stimulates the activity of cellular IRESs that control the expression of proteins needed for HCMV replication. While its effect on PKR contributes to the enhancement of cap-independent translation, we find that pTRS1 also promotes cap-independent translation in PKR deficient cells. Stimulation of cap-independent translation is dependent on the ability of pTRS1 to bind dsRNA. Together these data
111 show that pTRS1 stimulates cap-independent translation, and suggest that pTRS1 may play a role in translation regulation during HCMV infection.
INTRODUCTION
Control of protein synthesis is key to the regulation of multiple cellular processes including metabolism, cell cycle progression, and antiviral responses (1-3). Translation initiation is the rate-limiting step of protein synthesis and is highly regulated (4, 5).
Factors such as the availability of nutrients, the presence of unfolded proteins, or virus infection affect how efficiently translation initiation occurs. In fact, activation of many cellular stress response pathways leads to inhibition of translation initiation. Therefore, by regulating the rate of translation initiation, overall levels of protein synthesis can be tuned to match the cellular environment (6).
Translation initiation on the majority of cellular mRNAs occurs in a cap- dependent manner. The 7-methylguanosine mRNA cap (m7G) is bound by the eIF4F complex, which consists of the cap-binding protein eIF4E, the RNA helicase eIF4A, and the scaffold protein eIF4G (7-9). Nucleation of the eIF4F complex on an mRNA leads to the recruitment of the 43S preinitation complex (PIC), consisting of eIF3, eIF2-GTP, an initiator methionyl tRNA (Met-tRNAi) and the 40S ribosomal subunit, to form the 48S complex (10-12). The 48S complex scans the mRNA 5’ untranslated region (UTR) until the translation start codon is recognized, at which time the 60S ribosomal subunit is recruited and translation elongation begins (13).
While most translation requires the mRNA cap for initiation, translation initiation can also occur through cap-independent mechanisms. Internal ribosome entry sites, or
112 IRESs, are RNA structures that directly recruit ribosomal subunits and/or translation initiation factors to allow for translation initiation independent of the m7G cap (5, 14-16).
This allows for efficient translation of IRES-containing mRNAs under conditions of cell stress where eIF4F abundance is limited and cap-dependent translation is inhibited (17,
18). In fact, most mRNAs containing IRES elements encode proteins that are involved in stress response and cell death pathways (18, 19). Thus IRES-mediated translation provides an alternative to cap-dependent translation to allow for ongoing protein synthesis during periods of cellular stress.
Cap-independent translation was first described for viral mRNAs (14), as viruses commonly inhibit cap-dependent translation initiation. IRESs present on viral mRNAs allows for ongoing viral protein synthesis despite inhibition of cap-dependent translation.
The IRES controlling expression of the poliovirus polyprotein provides an excellent example (20-22). Poliovirus encodes a viral protease that cleaves eIF4G during infection, inhibiting cap-dependent translation. However, stem loop structures within the poliovirus IRES (located in the 5’UTR) directly bind the carboxyl-terminus of cleaved eIF4G, which then recruits eIF4A and the 43S PIC. The 48S complex then scans through an additional stem loop within the poliovirus 5’UTR and initiates translation at the proper AUG (23-27). eIF4E is not necessary for the translation of poliovirus RNA, as the presence of a cap is not necessary for eIF4G recruitment. Rhinovirus, picornavirus, flavivirus, and pestivirus RNAs contain IRESs to ensure the synthesis of viral proteins during infection (28-31). Thus viral IRESs allow for the continued expression of viral proteins under conditions with limited cap-dependent translation initiation.
113 While cap-independent translation has been well characterized in the context of
RNA virus infections, its role in the translation of DNA viruses mRNAs has been studied in less detail. Human cytomegalovirus (HCMV), a double-stranded DNA virus of the β- herpesvirus family, transcribes its capped mRNAs in the nucleus, which are then transported to the cytoplasm and translated by host ribosomes (32, 33). Unlike most viruses, HCMV infection does not induce translational shut-off, and both cellular and viral mRNAs are translated throughout infection (34, 35). The ongoing translation of host mRNAs encoding proteins needed for HCMV replication is due in part to an increase in eIF4F levels in infected cells (36, 37). However, the translation of HCMV mRNAs becomes increasingly resistant to eIF4F disruption as infection progresses (38).
While only a single HCMV IRES has been described in detail, ~100 HCMV 5’UTRs have been shown to promote cap-independent translation in a bicistronic mRNA reporter screen (39, 40). These data suggest that HCMV uses an eIF4F-independent mechanism to initiate translation of viral mRNAs, but the mechanism(s) controlling translation initiation on HCMV mRNAs is currently unknown.
We previously found that the HCMV TRS1 protein (pTRS1) increases overall levels of protein synthesis and stimulates cap-dependent translation. pTRS1 increases the translation of monocistronic mRNAs containing both cellular and HCMV 5’UTRs, though pTRS1 preferentially enhances the translation of mRNAs containing viral
5’UTRs (41). pTRS1 binds double-stranded (dsRNA) and the m7G cap, and inhibits activation of the antiviral kinase PKR, which dramatically inhibits translation initiation during infection (41-46). However, pTRS1 also stimulates mRNA translation in cells lacking PKR (41). Thus pTRS1 enhances translation independent of PKR inhibition.
114 While pTRS1 has been shown to stimulate cap-dependent translation, its effect on cap- independent translation has not been examined.
Here we show that HCMV pTRS1 enhances cap-independent translation. pTRS1 increased the activity of both the poliovirus and KSHV vFLIP IRESs in a bicistronic reporter assay. To avoid technical issues associated with bicistronic reporters, we developed a new assay to measure IRES activity from circular RNAs. pTRS1 enhanced translation of circular mRNAs containing the poliovirus or KSHV IRES, as well as the cellular BiP and c-myc IRES elements. pTRS1 also stimulated cap-independent translation of bicistronic and circular mRNAs in cells lacking PKR, showing that the ability of pTRS1 to stimulate cap-independent translation was separable from its ability to antagonize PKR. A pTRS1 mutant that cannot bind dsRNA did not enhance cap- independent translation, suggesting that pTRS1 directly binds RNAs to promote ribosome recruitment. Together these data show that pTRS1 stimulates cap- independent translation, and may drive the eIF4F-independent translation of viral transcripts during HCMV infection.
MATERIALS AND METHODS
Cells and viruses. HeLa cells and HEK293T cells were maintained in
Dulbecco’s Modified Eagle Medium (Sigma) supplemented with 10% fetal bovine serum
(Gibco) and 1% penicillin/streptomycin solution (Sigma). HeLa cells lacking PKR expression (PKR KO, (45)) were maintained in media as above supplemented with 1 ug/ml puromycin.
115 Generation of recombinant plasmids. The poliovirus and KSHV vFLIP bicistronic reporters were a kind gift from Dr. Felicia Goodrum at the University of
Arizona (39). The circular GFP reporter (TR-circGFP) was a kind gift from Dr. Aravind
Asokan at the University of North Carolina at Chapel Hill (47). To clone IRES elements into the circular RNA reporter, the circular GFP reporter plasmid was amplified using circGFP F and circGFP R (Table 1). The poliovirus IRES element was amplified from the bicistronic reporter above using poliocirc F and poliocirc R (Table 1). The KSHV vFLIP IRES was amplified from the bicistronic reporter above using KSHVcirc F and
KSHVcirc R (Table 1). The IRES elements were ligated into the amplified circular GFP reporter using Gibson cloning (NEB). The BiP and c-myc IRES elements were ordered as gBlock gene fragments (IDT; Table 1) and ligated to the amplified circular GFP reporter using Gibson cloning. The sequences of all circular GFP constructs were confirmed by Sanger sequencing. The pTRS1 mutants have been described previously
(46).
Luciferase assays. HeLa cells were seeded into 12-well plates (150,000 cells/well) and transfected with the indicated plasmids using polyethylenimine (PEI,
Polysciences). Twenty-four hours after transfection cells were washed in 1 ml 1X PBS
(Sigma) and lysed in 150 µl 1X Passive Lysis Buffer (Promega) for 10 min with rocking at room temperature. After lysis supernatants were cleared of debris by centrifugation at
10,000 x g for 1 min. For firefly luciferase, 40 µl of luciferase reagent (Promega) was added to 8 µl of sample, and luciferase activity was measured using a luminometer
(Molecular Devices). For experiments using bicistronic reporters, 40 µl of Stop and Glo reagent was added to each sample after measuring firefly luciferase levels. 40 µl of
116 renilla luciferase reagent (Promega) was then added, and luciferase activity was again measured. The amount of luciferase activity was normalized to the amount of protein present in each sample as determined by Bradford assay (Amresco).
Western blot analysis. Cells were washed once in 1X PBS before scraping, and cell pellets were stored at -80˚C until analysis. Cell pellets were lysed in RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP40, 1% sodium deoxycholate) containing protease and phosphatase inhibitors (Roche) for 10 min on ice. Cell debris was removed by centrifugation at 10,000 x g for 10 min. The protein concentration of the lysate was then determined by Bradford assay, and equal amounts of protein loaded onto SDS-PAGE gels. The resolved proteins were transferred to 0.45 µm nitrocellulose membranes (Amersham) and blocked in 5% non-fat milk in TBS-T (20 mM Tris-HCl, pH
7.6, 140 mM NaCl, 0.1% Tween 20) for 1 hour at room temperature. For mouse primary antibodies, membranes were incubated with the appropriate antibody diluted in 1% BSA in TBS-T for one hour at room temperature. For rabbit primary antibodies, membranes were incubated in the appropriate antibody diluted in 5% BSA in TBS-T overnight.
Secondary antibodies were diluted 1:10,000 in 1% BSA in TBS-T and incubated with the membrane for 1 hour at room temperature. Proteins were visualized by chemiluminesence using WesternBlot ECL (Advansta) and an imager (ChemiDoc;
BioRad). The following antibodies were used: anti-GFP (1:10,000; Sigma G1544), anti- tubulin (1:20,000; ThermoScientific A11126), anti-TRS1 (1:100; (48)), anti-His (1:1,000;
CST #2365).
Quantitative real-time PCR (qRT-PCR). RNA was extracted from cells using
Trizol essentially as described previously (41). Briefly, 100 µl of cell lysate was mixed
117 with 1 ml of Trizol and incubated at room temperature for 10 min. 200 µl of chloroform was then added, and samples were shaken for 60 sec before centrifugation at 10,000 x g for 10 min. The aqueous layer was removed and an equal volume of isopropanol was added. Samples were incubated at -20oC overnight, and the RNA was pelleted by centrifugation at 10,000 x g for 30 min at 4oC. RNA pellets were washed once in 70% ethanol, resuspended in 1X DNase buffer containing 20U DNase (Turbo DNase free kit;
Ambion), and incubated at 37oC for 30 min. DNase was inactivated according to the manufacturer’s directions. RNA concentrations were determined using a spectrophotometer (NanoDrop) and equal amounts of RNA were reverse transcribed to cDNA using the High Capacity cDNA Reverse Transcription Kit (ThermoFisher). 2 µl of cDNA was added to 1X SYBR Green Master Mix with the appropriate primers, and real- time PCR was performed using the following conditions: 95oC 5 minutes, followed by 40 cycles of 95oC 10 for seconds, 55oC for 30 seconds, 72oC for 90 seconds. Changes in
RNA levels were determined by the ∆∆Ct method as before using GAPDH as the reference gene (41). The following primer pairs were used (sequences listed in Table
1): GAPDH (GAPDH F and GAPDH R), renilla luciferase (RLUC F and RLUC R), firefly luciferase (FLUC F and FLUC R), circular GFP (qcircGFP F and qcircGFP R). All experiments were performed at least in triplicate and statistical significance was determined using a Student’s paired t-test.
RESULTS
pTRS1 enhances the activity of the poliovirus IRES. We previously found that pTRS1 increases the translation of monocistronic luciferase reporters regulated by cap-
118 dependent translation (41). To determine if pTRS1 could also enhance cap-independent translation, we measured the effect of pTRS1 expression on the translation of a bicistronic mRNA reporter containing the well-characterized poliovirus IRES (20, 22,
49). Translation of the first cistron of the bicistronic mRNA (renilla luciferase) is cap- dependent, while translation of the downstream cistron (firefly luciferase) is driven by an
IRES element in a cap-independent manner (Fig 1A). The effect of pTRS1 on cap- dependent and cap-independent translation was determined by comparing the effect of pTRS1 expression on renilla and firefly luciferase activity, respectively, from a bicistronic reporter containing no IRES element (control) or the poliovirus IRES (polio)
(Fig 1B). The bicistronic reporter containing the poliovirus IRES showed a 10-fold increase in firefly luciferase levels compared to the control when co-transfected with
GFP, consistent with previous results (39). pTRS1 expression increased both renilla and firefly luciferase levels from the control vector as compared to the GFP control (4- and 5.6-fold, respectively). While polio renilla levels were increased 2.2 fold in the presence of pTRS1, firefly luciferase levels were increased by 28.2-fold. Using the same data we also analyzed the effect of pTRS1 on IRES activity, defined as the ratio of firely to renilla luciferase activity (Fig 1A). pTRS1 expression did not affect IRES activity from the control reporter (Fig 1C), as levels of renilla and firefly luciferase were similar (Fig 1B). However, pTRS1 increased poliovirus IRES activity by 13-fold compared to the control. In addition, pTRS1 increased poliovirus IRES activity in a dose-dependent manner (Fig 1D), consistent with a specific role for pTRS1 in enhancing cap-independent translation.
119 KSHV vFLIP IRES activity is enhanced in the presence of pTRS1. To determine if the effect of pTRS1 on cap-independent translation was specific to the poliovirus IRES, we examined the effect of pTRS1 on the activity of an additional IRES from the Kaposi’s sarcoma-associated herpersvirus (KSHV) vFLIP mRNA (50). pTRS1 increased the activity of the KSHV vFLIP IRES by 29-fold as compared to the control
(Fig 2A). Analyzing the effect of pTRS1 expression on renilla and firefly luciferase levels separately, we found that pTRS1 expression had little to no effect on renilla luciferase activity (Fig 2C), but significantly increased firefly luciferase activity (Fig 2D).
Importantly, pTRS1 expression did not result in a significant change to firefly or renilla
RNA levels (Fig 2B). Thus the increase in poliovirus and KSHV vFLIP IRES activity in the presence of pTRS1 was not due to cryptic promoter activity or cryptic RNA splicing within the reporter plasmid or mRNA. These data suggest that pTRS1 can enhance cap-independent translation driven by multiple IRESs.
pTRS1 promotes translation of a circular mRNA. Translation of the firefly open reading frame (ORF) in the bicistronic reporter requires an IRES element.
However the bicistronic RNA encoding the firefly ORF contains a 7-methygluanosine
(m7G) mRNA cap, which pTRS1 is known to bind (41). Thus pTRS1 could promote translation of the firefly ORF by bridging an interaction with cap-associated proteins and the IRES element. This would mean that the ability of pTRS1 to stimulate IRES activity is not truly independent of the m7G mRNA cap. To assess this possibility, we determined the effect of pTRS1 on the translation of a circular reporter RNA. The circular RNA reporter consists of a single exon minigene containing split GFP in the reverse order (Fig 3A). Backsplicing of the primary transcript produces a circular RNA
120 (circRNA), which lacks both a m7G RNA cap and poly(A) tail, and contains the GFP coding sequence in the correct order. The inclusion of an IRES upstream of the GFP start codon allows the circRNA to be efficiently translated (47, 51). As the circRNA GFP reporter lacks an m7G mRNA cap and poly(A) tail, and translation is solely driven by the
IRES element, this system is ideal for assessing IRES activity.
We first cloned the poliovirus and KSHV vFLIP IRES into the circular GFP reporter (polio circGFP and KSHV circGFP, respectively). We then measured GFP expression from the circular RNA reporters. GFP was expressed in cells transfected with polio circGFP or KSHV circGFP, showing that both the poliovirus and KSHV vFLIP
IRESs promote circGFP translation (Fig 3B). We next examined the effect of pTRS1 expression on polio circGFP and KSHV circGFP (Fig 3B). GFP expression increased when the circGFP reporters were co-transfected with pTRS1 and these changes mirrored the increase in IRES activity in the presence of pTRS1 from the bicistronic reporters. In addition, pTRS1 did not change the levels of circGFP RNA (Fig 3C). These data suggest that pTRS1 promotes mRNA translation independent of the m7G mRNA cap or factors associated with a poly(A) tail.
We also tested the ability of pTRS1 to stimulate the translation of circRNA reporters containing cellular IRES elements. We examined the effect of pTRS1 on translation driven by the binding immunoglobulin protein (BiP) and c-myc IRES elements (16, 17, 19) (Fig 4). BiP IRES activity and protein levels, and c-myc RNA levels increase during HCMV infection (52-57). pTRS1 enhanced BiP IRES activity, as shown by increased GFP protein expression from the BiP circRNA reporter, though the effect was less pronounced than observed with the poliovirus and vFLIP IRESs. We
121 saw undetectable levels of GFP from the c-myc circRNA in the absence of pTRS1, suggesting that the c-myc IRES sequence normally has very low IRES activity. However as with the other IRESs, pTRS1 expression increased GFP protein expression driven by the c-myc IRES. Thus, in addition to increasing the activity of viral IRESs, pTRS1 also enhances translation driven by cellular IRES elements.
pTRS1 stimulates IRES activity independent of its ability to antagonize
PKR. pTRS1 enhances translation in part by preventing PKR activation. However, pTRS1 also promotes cap-dependent translation, independent of PKR antagonism (41,
45). To determine if the increase in cap-independent translation in the presence of pTRS1 was due to inhibition of PKR activation, we measured the effect of pTRS1 on
IRES activity in cells where CRISPR/Cas9 mutagenesis was used to ablate PKR expression (PKR KO cells, (45)). Using bicistronic reporters, we found that the poliovirus and KSHV vFLIP IRESs had increased activity in PKR KO cells, demonstrating that PKR inhibits poliovirus and KSHV vFLIP IRES activity. In addition pTRS1 increased the activity of both IRESs in PKR KO cells compared to cells expressing GFP (Fig 5A & 5B). We confirmed these results using the circGFP reporter containing the KSHV vFLIP IRES. pTRS1 expression increased GFP protein levels in
PKR KO cells, showing that pTRS1 stimulates KSHV vFLIP IRES activity in a PKR- independent manner (Fig 5C). Thus pTRS1 stimulates cap-independent translation independent of its ability to inhibit PKR activation.
The ability of pTRS1 to bind RNA is necessary to stimulate cap- independent translation. pTRS1 has a non-canonical RNA binding domain that preferentially binds double-stranded RNA (dsRNA) (43, 44). While this function is
122 dispensable for pTRS1 to antagonize PKR in transfected cells, we hypothesized it may be important to enhance cap-independent translation (46). To determine if pTRS1 binding to dsRNA was necessary to enhance cap-independent translation, we compared the ability of wild type pTRS1 and a TRS1 mutant that cannot bind dsRNA
(Triple) to increase GFP protein expression from the KSHV vFLIP circRNA (Fig 6). Wild type pTRS1 increased GFP expression compared to the control. However, no GFP expression was observed in cells transfected with pTRS1 triple. These results show that the ability of pTRS1 to bind dsRNA is necessary for its ability to stimulate cap- independent translation.
DISCUSSION
In this study we found that HCMV pTRS1 enhances cap-independent translation. pTRS1 stimulates the activity of multiple IRESs in both a bicistronic reporter assay, and in a novel circular RNA assay of IRES activity. pTRS1 stimulates cap-independent translation independent of its ability to inhibit PKR, as it increases the activity of multiple
IRESs in PKR deficient cells. The ability of pTRS1 to bind dsRNA is necessary to stimulate cap-independent translation, suggesting that pTRS1 may bind specific mRNA sequences or structures to facilitate ribosome recruitment.
Our results raise the question of how pTRS1 enhances cap-independent translation. While pTRS1 binds the m7G mRNA cap (41), this function is dispensable to stimulate IRES activity as pTRS1 enhances translation of circRNA reporters lacking an mRNA cap and a poly(A) tail. Our group and others have shown that pTRS1 binds RNA, with a preference for binding to double-stranded RNA (dsRNA) (43, 58). The
123 requirement for the pTRS1 dsRNA binding domain to stimulate cap-independent translation suggests that pTRS1 directly interacts with the IRES to recruit ribosomes.
The fact that pTRS1 stimulates the activity of multiple IRESs with minimal sequence homology suggests that pTRS1 likely binds common IRES RNA structures. As pTRS1 also associates with 40S ribosomal subunits (41), these data suggest a model wherein pTRS1 binds to double-stranded RNA structures in IRESs and facilitates the recruitment of ribosomal subunits, resulting in enhanced cap-independent translation.
pTRS1 potently antagonizes the antiviral kinase PKR, which inhibits both cap- dependent and -independent translation by limiting the availability of active ternary complexes (42, 45, 46, 59, 60). Our results show that the inhibition of PKR by pTRS1 contributes to increased cap-independent translation. However pTRS1 also enhances cap-independent translation in PKR deficient cells, possibly reflecting a role for pTRS1 in recruiting ribosomes as discussed above. Multiple host RNA binding proteins recognize viral RNA elements and activate restriction pathways that limit viral protein synthesis (3). Therefore it is possible that pTRS1 also antagonizes additional restriction factors that recognize and inhibit the translation of viral RNA. If so, the requirement for the pTRS1 RNA binding domain would suggest that pTRS1 competes with antiviral sensors to limit their detection of dsRNA structures.
How might the enhancement of cap-independent translation by pTRS1 impact
HCMV replication? HCMV infection increases eIF4F abundance, and the eIF4F complex is required for the synthesis of several cellular proteins needed for virus replication (36-38). Although, inhibiting the activity or formation of the eIF4F complex does not limit the association of HCMV mRNAs with polysomes, especially during the
124 later stages of infection (38). Our results showing that pTRS1 enhances the translation of a circRNA reporter suggests that pTRS1 may help recruit ribosomes to HCMV mRNAs independent of the eIF4F complex. The increase in cap-independent translation caused by pTRS1 may also promote the expression of cellular proteins necessary for
HCMV replication. BiP IRES activity and protein levels increase during HCMV infection, and BiP expression is necessary for efficient HCMV replication (52-55). Perhaps the stimulation of cap-independent translation by pTRS1 allows for the efficient translation of BiP and other pro-viral cellular factors during infection to allow for successful HCMV replication
Our results add to the growing list of functions ascribed to pTRS1 (41, 44-46, 58,
61, 62). While antagonism of PKR is an important role of pTRS1 during infection of primary fibroblasts, the enhancement of cap-independent translation may be more important in other cell types relevant to HCMV disease, such as CD34+ progenitor cells, epithelial cells or placental cells (63, 64). Further studies to define the mechanism behind the ability of pTRS1 to stimulate translation will therefore be critical to fully understand the role of pTRS1 in HCMV pathogenesis.
CONCLUSIONS
We would like to thank the Asokan and Marzluff labs for providing the circular
GFP reporters and valuable conversations. We would also like to thank the Laederach,
Heise, Baric, Moody, and de Silva labs for their helpful comments and suggestions.
Special thanks are given to Westchester Sanders and Anne Beall for their continued support. This work was supported by NIH grant AI03311 to N.J.M. Additional support
125 was provided by the North Carolina University Cancer Research Fund, an award from the UNC Virology Training grant (T32 AI07419 to B.Z.).
FIGURES
Figure 3.1. pTRS1 stimulates poliovirus IRES activity. (A) Schematic of bicistronic mRNA reporter. IRES activity is determined by comparing the level of firefly luciferase
(FLuc) to the level of renilla luciferase (RLuc). (B) HeLa cells were transfected with a bicistronic reporter containing no IRES element (EV) or a reporter containing the poliovirus IRES (Polio) in the presence of GFP or TRS1. Fold change of renilla luciferase (black bars) and firefly luciferase (white bars) compared to EV co-transfected with GFP are shown. (C) As in (B), IRES activity is represented as fold change compared to EV or Polio reporters co-transfected with GFP. (D) HeLa cells were
126 transfected with the Polio reporter and increasing amounts of TRS1 DNA. Fold change in IRES activity compared to reporter in the absence of TRS1 is shown.
Figure 3.2. pTRS1 increases the activity of multiple viral IRESs. (A) HeLa cells were co-transfected with a bicistronic reporter containing no IRES element (EV), the
KSHV vFLIP IRES (KSHV) or the poliovirus IRES (Polio) and either GFP or TRS1. The fold change in IRES activity in reporters co-transfected with TRS1 compared to co- transfection with GFP is shown. (B) RNA was extracted from cells transfected as in (A).
Renilla and fireflly luciferase RNA levels were determined by qRT-PCR. The ratio of firefly luciferase (FLuc) to renilla luciferase (RLuc) RNA in cells transfected with TRS1 compared to cells transfected with GFP is shown for each bicistronic reporter. (C) As in
(A), the fold change in renilla luciferase levels in cells transfected with TRS1 compared to cells transfected with GFP is shown for each reporter. (D) As in (C), except levels of firefly luciferase are shown.
127
Figure 3.3. pTRS1 stimulates the translation of a circular GFP mRNA. (A) Cartoon of the circular GFP reporter and the backsplicing event that results in the formation of a circular GFP mRNA. (B) HeLa cells were transfected with circular RNA reporters containing either the poliovirus IRES (Polio) or the KSHV vFLIP IRES (KSHV). GFP protein levels were determined by Western blot 48 hours post transfection. (C) As in (B), except following transfection, RNA was extracted and circular GFP mRNA levels were determined by qRT-PCR. The fold change in GFP RNA levels in cells transfected with
TRS1 compared to cells transfected with EV are shown for each reporter.
128
Figure 3.4. pTRS1 stimulates cellular IRES activity. (A) HeLa cells were co- transfected with a circular GFP reporter containing either the BiP IRES (BiP) or the c- myc IRES (c-myc) with either a vector control (EV) or TRS1. GFP levels were determined by Western blot 72 hours post transfection.
Figure 3.5. Stimulation of IRES activity by pTRS1 is not dependent on PKR antagonism. (A) WT HeLa cells were co-transfected with a bicistronic reporter containing the KSHV vFLIP IRES (KSHV) or the poliovirus IRES (Polio) with either GFP or TRS1. IRES activity of each reporter (FLuc/RLuc) is shown. (B) As in (A), but in PKR
KO cells. (C) WT and PKR KO HeLa cells were co-transfected with a circular GFP reporter containing the KSHV vFLIP IRES and either a vector control (EV) or TRS1.
Levels of GFP expression were determined by Western blot.
129
Figure 3.6. pTRS1 dsRNA binding is necessary to promote cap-independent translation. (A) HeLa cells were co-transfected with a circular GFP reporter containing the KSHV vFLIP IRES (KSHV circGFP) and either a vector control (EV) or the indicated
TRS1 construct. TRS1 triple (Triple) lacks the ability to bind dsRNA. Levels of GFP expression were determined by Western blot.
130 TABLES
Table 3.1. Primer and gBlock Gene Fragment Sequences
Primer Name Sequence (5' à 3') circGFP F ACCATGGTGAGCAAGGGCGA circGFP R GTCGACTGCAGAATTCAGAT poliocirc F ATCTGAATTCTGCAGTCGACTTAAAACAGCTCTGGGGTTGTTCC poliocirc R TCGCCCTTGCTCACCATGGTATATCTTAACAATGAGGTAATTCC KSHVcirc F ATCTGAATTCTGCAGTCGACTTGGACAGACTCCTACTT
KSHVcirc R TCGCCCTTGCTCACCATGGTTTGTATATGTGAAGGCACCG
ATCTGAATTCTGCAGTCGACAGGTCGACGCCGGCCAAGACAGCA CAGACAGATTGACCTATTGGGGTGTTTCGCGAGTGTGAGAGGGA BiP IRES AGCGCCGCGGCCTGTATTTCTAGACCTGCCCTTCGCCTGGTTCG gBlock TGGCGCCTTGTGACCCCGGGCCCCTGCCGCCTGCAAGTCGAAAT TGCGCTGTGCTCCTGTGCTACGGCCTGTGGCTGGACTGCCTGCT GCTGCCCAACTGGCTGGCAAGACCATGGTGAGCAAGGGCGA
ATCTGAATTCTGCAGTCGACTAATCCAGAACTGGATCGGGGTAAA GTGACTTGTCAAGATGGGAGAGGAGAAGGCAGAGGGAAAACGG GAATGGTTTTTAAGACTACCCTTTCGAGATTTCTGCCTTATGAATA TATTCACGCTGACTCCCGGCCGGTCGGACATTCCTGCTTTATTGT c-myc IRES GTTAATTGCTCTCTGGGTTTTGGGGGGCTGGGGGTTGCTTTGCG gBlock GTGGCAGAAAGCCCCTTGCATCCTGAGCTCCTTGGAGTAGGGAC CGCATATCGCCTGTGTGAGCCAGATCGCTCCGCAGCCGCTGACT TGTCCCCGTCTCCGGGAGGGCATTTAAATTTCGGCTCACCGCATT TCTGACAGCCGGAGACGGACACTCGGCGTCCCGCCCGCCTGTC CCCGCGGCGATTCCAACACCATGGTGAGCAAGGGCGA
GAPDH F CTGTTGCTGTAGCCAAATTCGT GAPDH R ACCCACTCCTCCACCTTTGAC RLUC F TCGGTTGGCAGAAGCTATGA RLUC R CCGATAAATAACGCGCCCAA FLUC F ACAAAGGCTATCAGGTGGCT FLUC R CGTGCTCCAAAACAACAACG qcircGFP F GCAGTGCTTCAGCCGCTAC qcircGFP R GTGTCGCCCTCGAACTTCAC
131 REFERENCES
1. Adeli K. 2011. Translational control mechanisms in metabolic regulation: critical role of RNA binding proteins, microRNAs, and cytoplasmic RNA granules. American Journal of Physiology - Endocrinology and Metabolism 301:E1051- E1064.
2. Pyronnet S, Sonenberg N. 2001. Cell-cycle-dependent translational control. Current Opinion in Genetics & Development 11:13-18.
3. Jan E, Mohr I, Walsh D. 2016. A Cap-to-Tail Guide to mRNA Translation Strategies in Virus-Infected Cells. Annual Review of Virology 3:283-307.
4. Hinnebusch AG. 2014. The scanning mechanism of eukaryotic translation initiation. Annual Review of Biochemistry 83:779-812.
5. Sonenberg N, Hinnebusch AG. 2009. Regulation of translation initiation in eukaryotes: mechanisms and biological targets. Cell 136:731-745.
6. Wek RC, Jiang H-Y, Anthony TG. 2006. Coping with stress: eIF2 kinases and translational control. Biochemical Society Transactions 34:7-11.
7. Grifo JA, Tahara SM, Morgan MA, Shatkin AJ, Merrick WC. 1983. New initiation factor activity required for globin mRNA translation. Journal of Biological Chemistry 258:5804-5810.
8. Tahara SM, Morgan MA, Shatkin AJ. 1981. Two forms of purified m7G-cap binding protein with different effects on capped mRNA translation in extracts of uninfected and poliovirus-infected HeLa cells. The Journal of Biological Chemistry 256:7691-7694.
9. Sonenberg N, Shatkin AJ. 1978. Nonspecific effect of m7GMP on protein-RNA interactions. The Journal of Biological Chemistry 253:6630-6632.
10. Shin B-S, Kim J-R, Walker SE, Dong J, Lorsch JR, Dever TE. 2011. Initiation Factor eIF2γ Promotes eIF2–GTP–Met-tRNAiMet Ternary Complex Binding to the 40S Ribosome. Nature structural & molecular biology 18:1227-1234.
132 11. Méthot N, Song MS, Sonenberg N. 1996. A region rich in aspartic acid, arginine, tyrosine, and glycine (DRYG) mediates eukaryotic initiation factor 4B (eIF4B) self-association and interaction with eIF3. Molecular and Cellular Biology 16:5328-5334.
12. Asano K, Clayton J, Shalev A, Hinnebusch AG. 2000. A multifactor complex of eukaryotic initiation factors, eIF1, eIF2, eIF3, eIF5, and initiator tRNAMet is an important translation initiation intermediate in vivo. Genes & Development 14:2534-2546.
13. Valásek L, Nielsen KH, Zhang F, Fekete CA, Hinnebusch AG. 2004. Interactions of eukaryotic translation initiation factor 3 (eIF3) subunit NIP1/c with eIF1 and eIF5 promote preinitiation complex assembly and regulate start codon selection. Molecular and Cellular Biology 24:9437-9455.
14. Hellen CUT, Sarnow P. 2001. Internal ribosome entry sites in eukaryotic mRNA molecules. Genes & Development 15:1593-1612.
15. Le SY, Maizel JV. 1997. A common RNA structural motif involved in the internal initiation of translation of cellular mRNAs. Nucleic Acids Research 25:362-369.
16. Yang Q, Sarnow P. 1997. Location of the internal ribosome entry site in the 5' non-coding region of the immunoglobulin heavy-chain binding protein (BiP) mRNA: evidence for specific RNA-protein interactions. Nucleic Acids Research 25:2800-2807.
17. Stoneley M, Paulin FE, Le Quesne JP, Chappell SA, Willis AE. 1998. C-Myc 5' untranslated region contains an internal ribosome entry segment. Oncogene 16:423-428.
18. Johannes G, Carter MS, Eisen MB, Brown PO, Sarnow P. 1999. Identification of eukaryotic mRNAs that are translated at reduced cap binding complex eIF4F concentrations using a cDNA microarray. Proceedings of the National Academy of Sciences 96:13118-13123.
19. Johannes G, Sarnow P. 1998. Cap-independent polysomal association of natural mRNAs encoding c-myc, BiP, and eIF4G conferred by internal ribosome entry sites. RNA (New York, NY) 4:1500-1513.
133 20. Pelletier J, Sonenberg N. 1988. Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature 334:320- 325.
21. Nomoto A, Kitamura N, Golini F, Wimmer E. 1977. The 5'-terminal structures of poliovirion RNA and poliovirus mRNA differ only in the genome-linked protein VPg. Proceedings of the National Academy of Sciences of the United States of America 74:5345-5349.
22. Pelletier J, Flynn ME, Kaplan G, Racaniello V, Sonenberg N. 1988. Mutational analysis of upstream AUG codons of poliovirus RNA. Journal of Virology 62:4486-4492.
23. Hambidge SJ, Sarnow P. 1992. Translational enhancement of the poliovirus 5' noncoding region mediated by virus-encoded polypeptide 2A. Proceedings of the National Academy of Sciences of the United States of America 89:10272-10276.
24. Borman AM, Bailly JL, Girard M, Kean KM. 1995. Picornavirus internal ribosome entry segments: comparison of translation efficiency and the requirements for optimal internal initiation of translation in vitro. Nucleic Acids Research 23:3656-3663.
25. Meyer K, Petersen A, Niepmann M, Beck E. 1995. Interaction of eukaryotic initiation factor eIF-4B with a picornavirus internal translation initiation site. Journal of Virology 69:2819-2824.
26. Pestova TV, Shatsky IN, Hellen CU. 1996. Functional dissection of eukaryotic initiation factor 4F: the 4A subunit and the central domain of the 4G subunit are sufficient to mediate internal entry of 43S preinitiation complexes. Molecular and Cellular Biology 16:6870-6878.
27. Lomakin IB, Hellen CU, Pestova TV. 2000. Physical association of eukaryotic initiation factor 4G (eIF4G) with eIF4A strongly enhances binding of eIF4G to the internal ribosomal entry site of encephalomyocarditis virus and is required for internal initiation of translation. Molecular and Cellular Biology 20:6019-6029.
28. Anderson EC, Hunt SL, Jackson RJ. 2007. Internal initiation of translation from the human rhinovirus-2 internal ribosome entry site requires the binding of Unr to two distinct sites on the 5' untranslated region. The Journal of General Virology 88:3043-3052.
134 29. Fletcher SP, Jackson RJ. 2002. Pestivirus Internal Ribosome Entry Site (IRES) Structure and Function: Elements in the 5′ Untranslated Region Important for IRES Function. Journal of Virology 76:5024-5033.
30. Kaminski A, Howell MT, Jackson RJ. 1990. Initiation of encephalomyocarditis virus RNA translation: the authentic initiation site is not selected by a scanning mechanism. The EMBO journal 9:3753-3759.
31. Lukavsky PJ. 2009. Structure and function of HCV IRES domains. Virus Research 139:166-171.
32. Mocarski Jr. ES. 2007. Betaherpes viral genes and their functions. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
33. Stinski MF, Meier JL. 2007. Immediate–early viral gene regulation and function. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
34. Tanaka S, Furukawa T, Plotkin SA. 1975. Human cytomegalovirus stimulates host cell RNA synthesis. Journal of Virology 15:297-304.
35. Stinski MF. 1977. Synthesis of proteins and glycoproteins in cells infected with human cytomegalovirus. Journal of Virology 23:751-767.
36. Walsh D, Perez C, Notary J, Mohr I. 2005. Regulation of the translation initiation factor eIF4F by multiple mechanisms in human cytomegalovirus- infected cells. Journal of Virology 79:8057-8064.
37. Perez C, McKinney C, Chulunbaatar U, Mohr I. 2011. Translational control of the abundance of cytoplasmic poly(A) binding protein in human cytomegalovirus- infected cells. Journal of Virology 85:156-164.
38. Lenarcic EM, Ziehr B, De Leon G, Mitchell D, Moorman NJ. 2014. Differential role for host translation factors in host and viral protein synthesis during human cytomegalovirus infection. Journal of Virology 88:1473-1483.
135 39. Grainger L, Cicchini L, Rak M, Petrucelli A, Fitzgerald KD, Semler BL, Goodrum F. 2010. Stress-inducible alternative translation initiation of human cytomegalovirus latency protein pUL138. Journal of Virology 84:9472-9486.
40. Weingarten-Gabbay S, Elias-Kirma S, Nir R, Gritsenko AA, Stern-Ginossar N, Yakhini Z, Weinberger A, Segal E. 2016. Systematic discovery of cap- independent translation sequences in human and viral genomes. Science 351:aad4939.
41. Ziehr B, Lenarcic E, Vincent HA, Cecil C, Garcia B, Shenk T, Moorman NJ. 2015. Human cytomegalovirus TRS1 protein associates with the 7- methylguanosine mRNA cap and facilitates translation. Proteomics 15:1983- 1994.
42. Braggin JE, Child SJ, Geballe AP. 2016. Essential role of protein kinase R antagonism by TRS1 in human cytomegalovirus replication. Virology 489:75-85.
43. Bierle CJ, Semmens KM, Geballe AP. 2013. Double-stranded RNA binding by the human cytomegalovirus PKR antagonist TRS1. Virology 442:28-37.
44. Hakki M, Geballe AP. 2005. Double-stranded RNA binding by human cytomegalovirus pTRS1. Journal of Virology 79:7311-7318.
45. Ziehr B, Vincent HA, Moorman NJ. 2016. Human Cytomegalovirus pTRS1 and pIRS1 Antagonize Protein Kinase R To Facilitate Virus Replication. Journal of Virology 90:3839-3848.
46. Vincent HA, Ziehr B, Moorman NJ. 2017. Mechanism of Protein Kinase R Inhibition by Human Cytomegalovirus pTRS1. Journal of Virology 91.
47. Wang Y, Wang Z. 2015. Efficient backsplicing produces translatable circular mRNAs. RNA (New York, NY) 21:172-179.
48. Romanowski MJ, Shenk T. 1997. Characterization of the human cytomegalovirus irs1 and trs1 genes: a second immediate-early transcription unit within irs1 whose product antagonizes transcriptional activation. Journal of Virology 71:1485-1496.
136 49. Balvay L, Rifo RS, Ricci EP, Decimo D, Ohlmann T. 2009. Structural and functional diversity of viral IRESes. Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1789:542-557.
50. Bieleski L, Talbot SJ. 2001. Kaposi's sarcoma-associated herpesvirus vCyclin open reading frame contains an internal ribosome entry site. Journal of Virology 75:1864-1869.
51. Borchardt EK, Meganck RM, Vincent HA, Ball CB, Ramos SBV, Moorman NJ, Marzluff WF, Asokan A. 2017. Inducing circular RNA formation using the CRISPR endoribonuclease Csy4. RNA (New York, NY) 23:619-627.
52. Buchkovich NJ, Yu Y, Pierciey FJ, Alwine JC. 2010. Human cytomegalovirus induces the endoplasmic reticulum chaperone BiP through increased transcription and activation of translation by using the BiP internal ribosome entry site. Journal of Virology 84:11479-11486.
53. Buchkovich NJ, Maguire TG, Alwine JC. 2010. Role of the endoplasmic reticulum chaperone BiP, SUN domain proteins, and dynein in altering nuclear morphology during human cytomegalovirus infection. Journal of Virology 84:7005-7017.
54. Buchkovich NJ, Maguire TG, Paton AW, Paton JC, Alwine JC. 2009. The endoplasmic reticulum chaperone BiP/GRP78 is important in the structure and function of the human cytomegalovirus assembly compartment. Journal of Virology 83:11421-11428.
55. Buchkovich NJ, Maguire TG, Yu Y, Paton AW, Paton JC, Alwine JC. 2008. Human cytomegalovirus specifically controls the levels of the endoplasmic reticulum chaperone BiP/GRP78, which is required for virion assembly. Journal of Virology 82:31-39.
56. Hagemeier C, Walker SM, Sissons PJ, Sinclair JH. 1992. The 72K IE1 and 80K IE2 proteins of human cytomegalovirus independently trans-activate the c- fos, c-myc and hsp70 promoters via basal promoter elements. The Journal of General Virology 73 ( Pt 9):2385-2393.
57. Monick MM, Geist LJ, Stinski MF, Hunninghake GW. 1992. The immediate early genes of human cytomegalovirus upregulate expression of the cellular genes myc and fos. American Journal of Respiratory Cell and Molecular Biology 7:251-256.
137 58. Lenarcic EM, Ziehr BJ, Moorman NJ. 2015. An unbiased proteomics approach to identify human cytomegalovirus RNA-associated proteins. Virology 481:13-23.
59. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition. Cell 122:901-913.
60. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. 2001. Tight Binding of the Phosphorylated α Subunit of Initiation Factor 2 (eIF2α) to the Regulatory Subunits of Guanine Nucleotide Exchange Factor eIF2B Is Required for Inhibition of Translation Initiation. Molecular and Cellular Biology 21:5018- 5030.
61. Chaumorcel M, Lussignol M, Mouna L, Cavignac Y, Fahie K, Cotte-Laffitte J, Geballe A, Brune W, Beau I, Codogno P, Esclatine A. 2012. The Human Cytomegalovirus Protein TRS1 Inhibits Autophagy via Its Interaction with Beclin 1. Journal of Virology 86:2571-2584.
62. Child SJ, Hakki M, De Niro KL, Geballe AP. 2004. Evasion of cellular antiviral responses by human cytomegalovirus TRS1 and IRS1. Journal of Virology 78:197-205.
63. Nogalski MT, Collins-McMillen D, Yurochko AD. 2014. Overview of Human Cytomegalovirus Pathogenesis, p 15-28, Human Cytomegaloviruses. Humana Press, Totowa, NJ.
64. Sinzger C, Digel M, Jahn G. 2008. Cytomegalovirus Cell Tropism, p 63-83, Human Cytomegalovirus. Springer, Berlin, Heidelberg.
138 CHAPTER 4: HCMV PTRS1 INTERACTS WITH ACTIVE PP1 CATALYTIC SUBUNITS
OVERVIEW
Human cytomegalovirus (HCMV) is a ubiquitous pathogen that infects a majority of the world’s population. Lytic HCMV replication in immunocompromised individuals or neonates can lead to disease in multiple organ systems, morbidity, and even and mortality. The HCMV protein TRS1 (pTRS1) and its functional homologue (pIRS1) are necessary for lytic replication. Both proteins antagonize the antiviral protein kinase R
(PKR) to inhibit the host antiviral respsone and ensure efficient HCMV replication. pTRS1 has additional PKR-independent functions, including the ability to stimulate cap- dependent and cap-independent translation. However the mechanism behind these functions remains unknown. To determine how pTRS1 promotes translation in a PKR- independent manner we identified ribosome-associated cellular proteins that interact with pTRS1. We found that pTRS1 interacts with protein phosphatase 1 (PP1) catalytic subunits both in vitro and during HCMV infection. pTRS1-specific immune complexes were significantly associated with phosphatase activity, but pTRS1 expression had no effect on PP1 catalytic activity. Rather, pTRS1 expression changed the complement of proteins that interact with the PP1 alpha catalytic subunit. These data suggest that rather than affecting PP1 catalytic activity, pTRS1 changes the substrate specificity of the PP1 holoenzyme. Thus the interaction of pTRS1 with PP1 could affect the regulation of multiple signaling pathways during HCMV infection.
139 INTRODUCTION
Human cytomegalovirus (HCMV) is a significant human pathogen. Reactivation of a latent infection or primary infection in immunocompromised individuals can lead to significant morbidity and mortality (1). In addition, primary HCMV infection during pregnancy is a leading cause of congenital birth defects (2). HCMV infection is characterized by a prolonged replication cycle and substantial manipulation of cellular signaling processes (1). HCMV regulates cellular metabolism, cell cycle progression, and protein synthesis pathways to ensure that viral proteins are expressed and progeny virions are produced (3-6). Viral infection is an inherently stressful event, thus HCMV must also blunt cellular stress responses to ensure efficient replication (7).
HCMV expresses several viral proteins that regulate cellular stress response pathways, including the viral proteins pTRS1 and pIRS1, which inhibit activation of the antiviral kinase PKR (8-11). PKR is activated upon binding to viral double-stranded RNA
(dsRNA) (12, 13). Once activated, PKR phosphorylates the alpha subunit of the eIF2 translation initiation factor (eIF2α) (13). Phosphorylation of eIF2α prevents the formation of ternary complexes, which are essential for translation initiation (14, 15). Therefore, activation of PKR in response to viral infection leads to an overall inhibition of translation, which prevents the synthesis of viral proteins. pTRS1 blocks PKR activation by binding PKR and inhibiting PKR kinase activity (8). As a result, loss of pTRS1 expression, and expression of its functional homologue pIRS1, leads to robust PKR activation and eIF2α phosphorylation early in infection (9). The resulting inhibition of protein synthesis severely attenuates viral protein expression, and therefore virus
140 replication (8, 10). Thus pTRS1 and pIRS1 block the activation of the PKR-eIF2α stress response pathway to ensure the synthesis of viral proteins during infection.
Viruses use multiple strategies to prevent eIF2α phosphorylation during infection
(16). The vaccinia virus proteins E3L and K3L both antagonize PKR. E3L binds and sequesters viral dsRNA, preventing detection by PKR, while K3L acts as a PKR pseudosubstrate and prevents its interaction with eIF2α (17-19). Herpes simplex virus-1
(HSV-1) also encodes two viral proteins that block the PKR-eIF2α signaling pathway.
The HSV-1 US11 protein binds both dsRNA and PKR, and prevents PKR activation (20-
22). Further, the HSV-1 protein ICP34.5 recruits the cellular phosphatase PP1 to de- phosphorylate eIF2α (16, 23). A similar function is used by the papilloma virus (HPV) protein E6 and the African Swine Fever Virus protein DP71L, which both recruit PP1 to dephosphorylate eIF2α (24, 25). While targeting PP1 to phosphorylated eIF2α is a common viral strategy to maintain protein synthesis, the role of PP1 in HCMV infection has not been studied in detail. Both PP1 levels and overall levels of phosphatase activity increase during HCMV infection. Also, phosphorylation of eIF2α is regulated by phosphatase activity during infection, and treatment with phosphatase inhibitors at the start of infection inhibits overall levels of protein synthesis. These data suggest that PP1 may play an important role in HCMV replication (26).
Although PKR antagonism is a primary pTRS1 function, pTRS1 has multiple functions independent of its ability to inhibit PKR activation. pTRS1 increases overall levels of protein synthesis and stimulates the translation of reporter mRNAs in cells lacking PKR expression (chapter 3)(27). In addition, pTRS1 limits eIF2α phosphorylation in response to activation of another eIF2α kinase, HRI, in PKR KO cells
141 (8). These data suggest that pTRS1 regulates multiple stress response pathways to prevent eIF2α phosphorylation. However, the mechanism behind these additional pTRS1 functions remains unknown.
Here we show that pTRS1 interacts with multiple PP1 catalytic subunits and changes the complement of proteins that interact with the PP1 alpha catalytic subunit. A screen of pTRS1-interacting proteins revealed that pTRS1 interacts with all three PP1 catalytic subunits. We confirmed the interaction of pTRS1 with the PP1Cα, PP1Cβ, and
PP1Cγ catalytic subunits in vitro, and showed that pTRS1 interacts with the PP1Cβ catalytic subunit during HCMV infection. Phosphatase activity is associated with pTRS1- specific immune complexes, however pTRS1 expression does not affect PP1Cα catalytic activity. Rather we found that pTRS1 changes the complement of proteins that interact with PP1Cα, suggesting that pTRS1 may change PP1 substrate specificity.
Together, these data suggest that the interaction of pTRS1 with PP1 may regulate multiple signaling pathways during HCMV replication.
MATERIALS AND METHODS
Cell culture, viral infections, and drug treatments. HeLa cells, HEK293T cells, and MRC-5 primary human fibroblasts were maintained in Dulbecco’s Modified Eagle’s
Medium (DMEM, Sigma) supplemented with 1% penicillin/streptomycin (Sigma) and
10% fetal bovine serum (FBS, Gibco). HeLa cells in which PKR expression was ablated using CRISPR/Cas9 (9) were maintained in DMEM containing 1% penicillin/ streptomycin, 10% FBS, and 1 µg/ml puromycin.
142 MRC-5 cells were infected with HCMV AD169inGFP at a multiplicity of infection
(MOI) of 3 for all experiments. This virus contains a GFP open reading frame (ORF) under the control of an SV40 promoter that is expressed throughout infection (9).
Where indicated, cells were treated with 0.5 mM arsenite (Sigma) for 1 hour to induce eIF2α phosphorylation. For BioID experiments, cells were treated with 50 µM biotin (Sigma) diluted in serum-free DMEM for 18 hours before harvest.
Generation of recombinant plasmids. The full-length TRS1 plasmid was a kind gift from Adam Geballe (28). For the construction of pTRS1 BioID: the BirA open reading frame (ORF) was amplified using TRS1 BirA F and BirA R (Table 1) from the pcDNA3.1 mycBioID plasmid (Addgene #35700). The TRS1 ORF was amplified using
TRS1 F and TRS1 BirA R (Table 1). TRS1 in pcDNA3.1 was digested with BamHI and
XhoI to remove the TRS1 ORF and linearize the backbone plasmid. The BirA ORF and
TRS1 ORF were then cloned into linear pcDNA3.1 using a Gibson cloning strategy
(NEB). For the construction of TRS1 ΔRVTF: The TRS1 pcDNA3.1 plasmid was amplified using RVTF F and RVTF R (Table 1). The PCR product was ligated using
Gibson (NEB). These mutations lead to a 780RVTF783 to 780AAAA783 mutation in TRS1.
Plasmids containing the PPP1CA, PPP1CB, and PPP1CC ORFs were obtained from the ORFeome collection at the Tissue Culture Facility at University of North
Carolina at Chapel Hill. For the construction of carboxyl-terminal myc-tagged PP1 catalytic subunits: The PPP1CA ORF was amplified using PPP1CA mycF and PPP1CA mycR. The PPP1CB ORF was amplified using PPP1CB mycF and PPP1CB mycR. The
PPP1CC ORF was amplified using PPP1CC mycF and PPP1CC mycR. All primers are listed in Table 1. pcDNA3.1 was linearized following digestion with BamHI and XhoI and
143 the PPP1 catalytic subunits were cloned into the plasmid backbone using Gibson cloning (NEB). For the PPP1CA F293A mutant: PPP1CA in pcDNA3.1 was amplified using the primers PP1 F293A F and PP1 F293A R (Table 1). The PCR product was then cirularized using Gibson (NEB). For the amino-terminal BirA tag on PPP1CA: The
BirA ORF was amplified using BirA pcD F and BirA R2 and the PPP1CA ORF was amplified using PPP1CA BioIDF and PPP1CA BioIDR (Table 1). pcDNA3.1 was linearized following digestion with BamHI and XhoI and the BirA and PPP1CA ORFs were cloned into the plasmid backbone using Gibson cloning (NEB). The sequence of all clones was confirmed using Sanger sequencing.
PKR activation. HEK293T cells were transfected with either TRS1 BioID or a
GFP control for forty-eight hours. Cells were then lysed in 120 µl of PKR lysis buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% NP-40, 10% glycerol, and 1 mM EDTA with protease and phosphatase inhibitors (Roche)). Cell debris was removed following a 10 min, 10,000 x g spin at 4°C. 1 mg/ml poly(I:C) (LMW, Invivogen) was added to 100 µl of lysate and incubated at 30°C for 30 min to induce PKR activation. Levels of PKR activation, measured by phosphorylation at PKR Thr446, were then determined by
Western blot analysis (see below).
Metabolic labeling. HeLa cells were transfected with either TRS1 BioID or an empty vector control for 24 hours. Cells were then starved in methionine/cysteine-free media (Sigma) for 15 min and subsequently treated with 125 µCi of 35S-labeled methionine and cysteine (PerkinElmer EasyTag Express Labeling Mix) for 30 min. Cells were then washed twice in ice-cold 1X PBS and scraped. Cell pellets were lysed in
RIPA buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, 0.25% deoxycholate,
144 1 mM EDTA, containing protease and phosphatase inhibitors (Roche)). Protein was then precipitated by adding trichloroacetic acid (TCA) to a final concentration of 20%.
Precipitated proteins were transferred to a glass microfiber filter, washed twice with 20%
TCA, washed once with 100% ethanol and then allowed to air dry. Filters were transferred to vials containing scintillation fluid and levels of radioactivity were determined using a scintillation counter. Levels of radioactivity were normalized to the levels of protein within the lysate as determined by Bradford assay (Amresco).
Ribosome Isolation. Cells were transfected with TRS1, BioID, or TRS1 BioID for 24 hours and then treated with 50 µM biotin for 18 hours. Cells were then treated with 0.1 mg/ml cyclohexamide (CHX) for 10 min. Cells were washed twice in ice-cold
PBS containing 0.1 mg/ml CHX, scraped, and pelleted at 4°C. Cell pellets were lysed in polysome lysis buffer (20 mM Tris-HCl (pH 7.4), 140 mM KCl, 5 mM MgCl2, 0.1%
TritonX-100, 10 mM DTT, 0.1 mg/ml CHX), allowed to swell on ice for 10 min, and then passed through a 27-gauge needle five times. Lysates were spun for 5 min at 3,000 x g at 4°C to pellet nuclei. Supernatant was then moved to a new tube and spun for 10 min at 10,000 x g at 4°C to remove mitochondria. 0.5 ml of supernatant was then layered onto a 1 ml sucrose cushion (1M sucrose, 0.5M KCl, 5 mM MgCl2, 50 mM Tris-HCl (pH
7.4)) and spun for 2h at 245,000 x g. Supernatant was decanted and ribosomal pellets washed once with ice-cold water before being stored at -80°C.
Immunoprecipitations. For the TRS1 BioID streptavidin affinity chromatography, ribosome pellets from cells transfected with TRS1, BioID, or TRS1
BioID were lysed in RIPA buffer supplemented with 2M Urea (50 mM Tris-HCl (pH 7.4),
150 mM NaCl, 1% NP-40, 0.25% deoxycholate, 1 mM EDTA, 2M Urea, containing
145 protease and phosphatase inhibitors (Roche)) for 10 min on ice. Lysates were sonicated
(3 pulses, 30% output) and cell debris removed following a 10 min spin at 10,000 x g at
4°C. 20 µl of Dynabeads MyOne Streptavidin T1 beads (Thermo Scientific) were washed three times and re-suspended in 100 µl of RIPA + Urea buffer. Re-suspended beads were added to cleared lysates and nutated at 4°C for 3 hours. Beads were then washed three times with RIPA + Urea buffer and washed an additional three times in 50 mM ammonium bicarbonate (ABC). Beads were re-suspended in 100 µl ABC and submitted to the mass spectrometry facility for analysis.
To confirm the pTRS1 interactions during infection, MRC-5 fibroblasts were infected with HCMV (MOI of 3) for 48 or 72 hours. Cells were washed once with 1X
PBS, pelleted, and lysed in RIPA light buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl,
1% NP-40, 1 mM EDTA, containing protease and phosphatase inhibitors (Roche)) for
10 min on ice. Cell debris was removed following a 10,000 x g spin at 4°C. 20 µl of
Protein A/G bead slurry (Santa Cruz Biotechnology) were then washed three times and re-suspended in 100 µl of RIPA light buffer. Re-suspended beads were added to cleared lysates and nutated at 4°C for 30 min in a pre-clear step. Beads were then spun out and the pTRS1 antibody (29) or mouse igG control (Bethyl) or anti-IE1 antibody control (30) were added to the lysates and nutated at 4°C for 1 hour. 100 µl of re- suspended protein A/G beads were then added to each lysate and nutated for an additional 1.5 hours at 4°C. Beads were then washed three times with RIPA light buffer and re-suspended in 1X sample buffer before Western blot analysis (see below).
For the PP1 catalytic subunit immunoprecipitations, both transfected HeLa cells and infected MRC-5 fibroblasts were washed once with 1X PBS, pelleted, and lysed in
146 Strang buffer (31) (50 mM Hepes (pH 7.4), 150 mM NaCl, 1% TritonX-100, 10% glycerol, 2 mM EDTA, 0.1 mM DTT, containing protease and phosphatase inhibitors
(Roche)) for 10 min on ice. Cell debris was removed following a 10,000 x g spin at 4°C.
20 µl of Protein A/G bead slurry (Santa Cruz Biotechnology) were then washed three times and re-suspended in 100 µl of Strang buffer. Re-suspended beads were added to cleared lysates and nutated at 4°C for 30 min in a pre-clear step. Beads were then spun out and the appropriate antibody or mouse IgG control (Bethyl) was added to the lysates and nutated at 4°C for 1 hour. 100 µl of re-suspended protein A/G beads were then added to each lysate and nutated for an additional 1.5 hours at 4°C. Beads were then washed three times with Strang buffer and re-suspended in 1X sample buffer before Western blot analysis (see below).
Western blot analysis. Unless otherwise stated, cell pellets were lysed in RIPA buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, 0.25% deoxycholate, 1 mM
EDTA, containing protease and phosphatase inhibitors (Roche)) for 10 min on ice. Cell debris was then removed by a 10,000 x g spin at 4°C, and the lysate protein concentration was determined by Bradford assay. Lysates were brought to a final concentration of 1X sample buffer (0.1 M Tris-HCl (pH 6.8), 6% glycerol, 2% SDS, 0.1 mM DTT, 0.002% bromophenol blue), boiled for 10 min and spun at 10,000 x g for 2 min before loading onto SDS-PAGE gels. Equal amounts of protein were then resolved on 10% SDS-PAGE gels and transferred to 0.45 µM nitrocellulose membrane (GE healthcare). Membranes were blocked in 5% non-fat milk in TBS-T (20 mM Tris-HCl (pH
7.6), 150 mM NaCl, 0.1% Tween 20) for one hour at room temperature. All mouse primary antibodies were diluted in 1% bovine serum albumin (BSA, VWR) in TBS-T.
147 Membranes were incubated in mouse primary antibody for either one hour at room temperature or overnight at 4°C. Rabbit antibodies were diluted in 5% BSA in TBS-T, with the exception of anti-His which was diluted in 1% BSA in TBS-T. Membranes were incubated in rabbit primary antibody overnight at 4°C. Both anti-mouse and anti-rabbit secondary antibodies (KPL) were diluted 1:10,000 in 1% BSA in TBS-T. Membranes were incubated in secondary antibody for one hour at room temperature before being developed with WesternBlot ECL and peroxide (Advansta). Membranes probed with streptavidin-HRP (Pierce, diluted 1:500 in 1% BSA in TBS-T) were incubated with antibody for one hour at room temperature before being developed. Membranes probed for eIF2α phosphorylation were blocked overnight with 5% BSA in TBS-T followed by an overnight incubation in primary antibody. The following primary antibodies were used:
Tubulin (1:20,000; Thermo Scientific), TRS1 (1:100; (29)), Total PKR (1:1,000; Santa
Cruz Biotechnologies), Phospho-PKR (Thr446) (1:1,000; Abcam), Streptavidin-HRP
(1:500; Pierce), PAN3 (1:1,000; Bethyl), PABP (1:1,000; CST), CPEB4 (1:1,000;
Genetex), myc (1:1,000; CST), PP1Cα (1:1,000; CST), PP1Cβ (1:1,000; CST), PP1Cγ
(1:1,000; CST), Total eIF2α (1:1,000; CST), Phospho-eIF2α (1:1,000; CST).
Phosphatase assay. TRS1 and PP1 immune complexes were isolated as above and associated phosphatase activity was determined using the Millipore Ser/Thr
Phosphatase Assay Kit 1 (#17-127). Briefly, beads were washed once with Ser/Thr buffer following three washes with Strang buffer. Beads were then re-suspended in a final volume of 80 µl of Ser/Thr buffer and 750 µM phosphopeptide. The phosphatase assay was ran for 10 min at 30°C with shaking, whereupon free phosphate levels were determined using Malachite green according to manufacturer instructions. For the rate
148 of phosphatase activity experiments: PP1 immune complexes were isolated as above and washed once with Ser/Thr buffer. Phosphatase assays were ran with phosphopeptide concentrations of 750 µM, 500 µM, 350 µM, 175 µM, or 100 µM. All experiments were run in triplicate and significance determined using a Student’s t-test
(p<0.05).
RESULTS
pTRS1 BioID to identify cellular pTRS1 binding partners. pTRS1 has multiple
PKR-independent functions, including promotion of both cap-dependent and cap- independent translation and inhibition of eIF2α phosphorylation following HRI activation
(chapter 3)(8, 27). However, the mechanism behind these functions remains unclear. To elucidate the mechanism of PKR-independent pTRS1 functions, we used the BioID approach to identify cellular proteins that interact with pTRS1 (32). We fused a promiscuous biotin ligase (BioID) to the carboxyl-terminus of pTRS1 (pTRS1 BioID).
When expressed in the presence of excess biotin, proteins that interact with pTRS1
BioID, even transiently, will be covalently biotinylated. pTRS1-interacting proteins can then be isolated using streptavidin affinity chromatography, and identified by mass spectrometry (Fig 1A). We confirmed that the pTRS1 BioID fusion protein maintained wild-type function by assessing the ability of pTRS1 BioID to antagonize PKR and stimulate translation. We found that pTRS1 BioID prevented PKR autophosphorylaton in response to poly(I:C) treatment (Fig 1B), and increased overall levels of protein synthesis by 50% compared to cells expressing GFP, similar to untagged pTRS1 (Fig
149 1C)(27). Thus pTRS1 BioID retains the known functions of pTRS1, suggesting the fusion protein functions similarly to wild type pTRS1.
To visualize proteins that interact with pTRS1 we transfected cells with TRS1,
BioID, or TRS1 BioID and treated them with exogenous biotin. Biotinylated proteins in whole cell lysates were analyzed by Western blot (Fig 1D). No biotinylated proteins were observed in cells transfected with untagged pTRS1, as expected. Many proteins were labeled in cells transfected with BioID. However, cell expressing pTRS1 BioID showed a different banding pattern compared to the BioID control. pTRS1 binds the mRNA cap, associates with 40S ribosomal subunits, and enhances translation in PKR deficient cells (27). These data highlight a role for pTRS1 in translation regulation. Thus, to identify pTRS1-associated proteins likely involved in translation regulation, we identified ribosome-associated proteins that interact with pTRS1. We pelleted ribosomes from cells transfected with untagged TRS1, free BioID, or TRS1 BioID, isolated biotinylated proteins from ribosome pellets using streptavidin affinity chromatography, and examined the recovered proteins by Western blot (Fig 1E). No biotinylated proteins were observed in control cells expressing untagged pTRS1, and a unique banding pattern of biotinylated proteins was found in cells expressing pTRS1
BioID compared to cells expressing free BioID. These data suggest that pTRS1 interacts with a specific subset of ribosome-associated proteins.
pTRS1 interacts with proteins that regulate RNA metabolism and post- transcriptional control of gene expression We next identified ribosome-associated, pTRS1 interactors. Ribosome-associated, biotinylated proteins from cells transfected with untagged TRS1, free BioID, or TRS1 BioID were identified by mass spectrometry.
150 The identified proteins were considered to specifically interact with pTRS1 if they were only observed in the pTRS1 BioID sample. We consistently identified 114 cellular proteins that specifically interact with pTRS1 (Fig 2A). As expected, PKR was amongst the top hits in our screen, validating that the BioID approach identifies known pTRS1 binding partners. Gene ontogeny analysis (GO analysis) revealed the molecular functions most associated with pTRS1 interacting partners were double-stranded and single-stranded RNA binding, 3’ untranslated region (UTR) binding, and 3’-5’ exoribonuclease activity (Fig 2B). GO analysis of biological processes showed that proteins that interact with pTRS1 are significantly enriched for nuclear mRNA surveillance, RNA helicase activity, and stress granule assembly (Fig 2C).
We validated the interaction of pTRS1 with several proteins identified in our mass spectrometry analysis: PABP, PAN3 and CPEB4. Poly(A) binding protein (PABP) binds the poly(A) tail of mRNAs and regulates translation initiation and mRNA stability (33).
PAN3 is a poly(A)-specific ribonuclease involved in mRNA decay (34), while the cytoplasmic polyadenylation element binding protein 4 (CPEB4) regulates poly(A) tail length and stability of target mRNAs (35). The interaction of pTRS1 with PABP and
PAN3 was confirmed in HCMV infected cells, demonstrating the relevance of these results to pTRS1 function during infection (Fig 3A,B). We also confirmed the interaction of CPEB4 and pTRS1 in transfected cells (Fig 3C). These data show that BioID is a valid approach to identify novel interactions between pTRS1 and cellular proteins.
pTRS1 interacts with PP1 catalytic subunits. Our mass spectrometry analysis showed that pTRS1 specifically interacts with the alpha, beta and gamma catalytic subunits of protein phosphatase 1 (PP1Cα, β, γ, respectively). We confirmed the
151 interaction of pTRS1 with each PP1 catalytic subunit in transfected cells (Fig 4A-C) and also found that pTRS1 interacts with PP1Cβ during HCMV infection (Fig 4D). PP1 is a serine/threonine phosphatase that regulates a majority of cellular signaling pathways.
PP1 catalytic subunits have no substrate specificity themselves, but are recruited to specific substrates through interactions with PP1 regulatory proteins. The majority of
PP1 regulatory proteins contain an ‘RVxF’ motif that mediates their binding to PP1 (36,
37). pTRS1 contains an ‘RVTF’ motif in its carboxyl-terminus. To determine if this motif mediated pTRS1 binding to PP1 catalytic subunits, we mutated TRS1 780RVTF783 to
780AAAA783 (TRS1 ΔRVTF) and determined if pTRS1 still interacted with PP1Cα. We found that the pTRS1 RVTF motif was dispensable for this interaction. These data suggest that another pTRS1 motif mediates the interaction of pTRS1 with PP1.
A single amino acid (Phe293) in PP1 catalytic subunits mediates the binding of
PP1 to regulatory subunits containing an RVxF motif (36). As most PP1 regulatory subunits bind Phe293, we tested the effect of mutating Phe293 to an alanine residue
(PP1Cα F293A) on the ability of pTRS1 to interact with PP1Cα (Fig 4F). The PP1Cα
F293A mutation completely ablated pTRS1 binding, suggesting that pTRS1 either binds
PP1 through an indirect interaction, or a non-RVxF pTRS1 motif interacts with the PP1
F293 residue.
pTRS1 expression does not affect PP1Cα phosphatase activity. We next determined if pTRS1 interacts with active PP1 catalytic subunits by measuring phosphatase activity in pTRS1 immune complexes from transfected cells (Fig 5A).
There was a significant increase in phosphatase activity associated with pTRS1-specific immune complexes compared to the antibody control, suggesting that active
152 phosphatases are associated with pTRS1. To determine if pTRS1 changes the activity of the PP1Cα catalytic subunit, we measured the rate of PP1Cα phosphatase activity in cells expressing pTRS1 or a control protein (GFP) (Fig 5B). There was no significant difference in the rate of PP1Cα phosphatase activity in cells expressing pTRS1 compared to control cells. We conclude that pTRS1 associates with active PP1 catalytic subunits, and that pTRS1 does not affect PP1 phosphatase activity.
pTRS1 changes PP1Cα binding partners. The majority of PP1 regulatory binding proteins do not affect PP1 catalytic activity, but rather change the substrate specificity of the bound catalytic subunit (36). Since pTRS1 does not change PP1 catalytic activity, we hypothesized that pTRS1 may change PP1 substrate specificity by altering the complement of regulatory proteins bound to PP1 catalytic subunits. To test this hypothesis we examined the effect of pTRS1 expression on PP1 catalytic subunit binding partners in transfected cells. We fused BioID to the amino-terminus of PP1Cα to create PP1Cα BioID. Importantly, PP1Cα BioID maintained wild-type levels of phosphatase activity (Fig 6A). We then co-transfected WT or PKR KO cells with
PPP1CA BioID and GFP or TRS1, treated cells with exogenous biotin, and isolated biotinylated proteins. When PP1Cα-interacting proteins were analyzed by Western blot we found that there was a difference in the complement of proteins in the presence of pTRS1 (Fig 6B), which was independent of PKR expression. These data suggest that pTRS1 changes either the components of the PP1Cα holoenzyme, or alters PP1Cα substrate specificity.
153 DISCUSSION
In this study we identified ribosome-associated cellular proteins that interact with pTRS1, with the goal of understanding how pTRS1 stimulates translation in a PKR- independent manner. pTRS1 specifically interacted with multiple ribosome-associated cellular proteins, which were significantly enriched in functions associated with translation initiation, 3’UTR binding, and RNA metabolism (Fig 2A,2B). We confirmed multiple novel pTRS1 interactions in transfected and HCMV-infected cells, including interactions with all three PP1 catalytic subunits (Fig 3,4). Phosphatase activity was found in pTRS1-specific immune complexes (Fig 5A), demonstrating that pTRS1 associates with active PP1. However pTRS1 did not affect the rate of PP1 catalytic activity (Fig 5B). Rather, pTRS1 appears to change the complement of proteins that interact with PP1 catalytic subunits (Fig 6B), potentially altering the complement of PP1 regulatory subunits or changing the substrate specificity of PP1 catalytic subunits.
We found that ribosome-associated proteins that interact with pTRS1 shared similar functions (Fig 2A). GO analysis of molecular functions revealed that pTRS1- interactors were significantly enriched for both double-stranded and single-stranded
RNA binding, and RNA helicase activity (Fig 2B). pTRS1 stimulates mRNA translation, independent of its ability to inhibit PKR activation (27). We found pTRS1 interacts with multiple RNA helicases, including DDX3X, DDX6, and DHX29 (Fig 2A). Each of these helicases associates with translation initiation complexes, and regulates the translation of a subset of cellular mRNAs (38). In addition, we confirmed that pTRS1 interacts with the poly(A) binding protein (PABP). These data, together with the fact that pTRS1 binds the m7G cap, dsRNA, and associates with 40S ribosomal subunits (27, 38-40), suggest
154 that pTRS1 may stimulate translation by recruiting translation initiation factors and ribosomal subunits to mRNAs. While additional studies are needed to test this hypothesis, our data have identified novel pTRS1-interactors that are likely important for the ability of pTRS1 to stimulate translation, independent of its ability to inhibit PKR.
GO analysis also revealed potential new pTRS1 functions. In addition to the functions described above, pTRS1-interacting proteins were also significantly enriched for 3’-5’ exoribonuclease activity (Fig 3C). pTRS1 specifically interacted with multiple
RNA exosome components (Fig 2A). In addition, we confirmed that pTRS1 associates with PAN3 in HCMV-infected cells (Fig 3B). PAN3 is part of the PAN2/3 exoribonuclease complex, which mediates deadenylation of mRNAs targeted for degradation. pTRS1 also interacted with several 3’UTR binding proteins that regulate mRNA stability and translation efficiency, such as STAU1, FUBP3, and SYNCRIP (34,
35). Thus regulation of mRNA stability may be an additional mechanism used by pTRS1 to facilitate translation. Promotion of mRNA stability, or inhibition of mRNA degradation, could enhance the translation of pTRS1 target mRNAs.
We showed that pTRS1 interacts with all three PP1 catalytic subunits, and that pTRS1 is associated with active phosphatase(s) (Fig 4,5A). More than 80% of proteins that bind PP1 do so through an ‘RVxF’ consensus motif (36). pTRS1 contains an ‘RVTF’ motif in its carboxyl-terminus, however this motif was dispensable for pTRS1 association with PP1Cα (Fig 4E). In contrast, mutating the PP1 residue necessary for binding of RVxF motif-containing PP1 regulatory subunits led to a loss of pTRS1 association with PP1Cα (Fig 4F). These data suggest that either another motif within pTRS1 interacts with the PP1 RVxF binding pocket, or the interaction between pTRS1
155 and PP1 is not direct. As no other motif within pTRS1 matches even the least stringent
RVxF motif ‘rules’, we hypothesize that pTRS1 associates with PP1Cα through an indirect interaction, possibly through an interaction with a PP1 regulatory subunit. This is consistent with our result showing that disrupting the residue in the PP1 catalytic subunit recognized by most PP1 regulatory subunits ablated the interaction of pTRS1 with PP1Cα. We also show that pTRS1 expression changes the complement of proteins that interact with PP1Cα (Fig 6B), suggesting that pTRS1 interacts with PP1 regulatory subunits to either change PP1 holoenzyme structure and/or PP1 substrate specificity.
PP1 regulates dozens of cellular processes by dephosphorylating its target substrates (41). PP1 levels and overall levels of phosphatase activity increase during
HCMV infection, and phosphatase activity is required for HCMV protein synthesis. Hakki et al. also showed that phosphatase activity is required to limit eIF2α phosphorylation during infection, but is dispensable for preventing PKR activation (26). This is consistent with our data showing that pTRS1 binds PKR and prevents its autophosphorylation and activation (8). If PKR is never phosphorylated, then phosphatase activity is irrelevant for its control. However these data raise the possibility that pTRS1 may also regulate the
PKR-eIF2α pathway by recruiting PP1 to reverse eIF2α phosphorylation. We previously found that pTRS1 limits eIF2α phosphorylation in response to activation of an additional eIF2α kinase, HRI, in PKR deficient cells (8). Here we show that pTRS1 interacts with active PP1 catalytic subunits and may change PP1 substrate specificity (Fig 5A,6B).
Perhaps pTRS1 recruits PP1 to eIF2α that has been phosphorylated by other eIF2α kinases (16). If so, this would suggest a more general role for pTRS1 in mitigating virus- induced stress by limiting eIF2α phosphorylation through an interaction with PP1.
156 It is also possible that the interaction of pTRS1 with PP1 catalytic subunits impacts additional signaling pathways during HCMV infection. We recently found that
HCMV changes the activity of a large number of cellular kinases (4). Many kinases are regulated by phosphorylation, thus pTRS1 could recruit PP1 to phosphorylated kinases to alter their activity. Or pTRS1 could recruit PP1 to dephosphorylate kinase substrates and inhibit signaling. Future studies to determine how the association of pTRS1 with
PP1 globally impacts kinase activity and substrate phosphorylation are in order to elucidate the full effect of this interaction on cellular signaling.
157 FIGURES
Figure 4.1. pTRS1 interacts with specific ribosome-associated proteins. (A)
Schematic of BioID approach to identifying pTRS1-interacting proteins. Cellular proteins that interact with pTRS1 BioID, in the presence of excess biotin, will be biotinylated. (B)
Lysates of cells transfected with either GFP or TRS1 BioID were spiked with poly(I:C) to activate PKR. Levels of PKR activation, measured by phosphorylation at PKR Thr446, were determined by Western blot analysis. (C) Overall levels of protein synthesis of cells transfected with an empty vector control or pTRS1 were determined through metabolic labeling of nascent peptides. (D) Cells were transfected with TRS1, BioID, or
TRS1 BioID and treated with excess biotin. Biotinylated proteins from whole cell lysates were isolated by streptavidin affinity chromatography and examined by Western blot analysis. (E) As in (D), but ribosomes were extracted from lysates of cells transfected
158 with TRS1, BioID, or TRS1 BioID before ribosome-associated, biotinylated proteins were isolated.
Figure 4.2. Ribosome-associated proteins that specifically interact with pTRS1.
(A) STRING analysis of 114 cellular proteins that were identified as hits by mass
159 spectrometry data only in cells transfected with TRS1 BioID. Gene ontology (GO) analysis of the molecular function (B) and biological processes (C) associated with ribosome-associated proteins identified as pTRS1-interactors.
Figure 4.3. Validation of pTRS1-interacting proteins. Fibroblasts were infected with
HCMV at an MOI of 3 for 72 hours. The presence of PABP (A) and PAN3 (B) in pTRS1- specific immune complexes was determined by Western blot analysis. (C) HeLa cells were co-transfected with CPEB4 and TRS1 and the presence of pTRS1 in CPEB4- specific immune complexes was determined by Western blot analysis.
160
Figure 4.4. pTRS1 interacts with all three PP1 catalytic subunits. HeLa cells were co-transfected with TRS1 and PPP1CA (A), PPP1CB (B), or PPP1CC (C). The presence of pTRS1 in PP1-specific immune complexes was determined by Western blot. (D) Fibroblasts were infected with HCMV at an MOI of 3 for 48 hours. The presence of PP1Cβ in pTRS1-specific immune complexes was then determined by
Western blot. (E) HeLa cells were co-transfected with PPP1CA and a pTRS1 mutant lacking its RVTF motif (ΔRVTF). The presence of WT pTRS1 or ΔRVTF in PP1Cα- specific immune complexes was then determined by Western blot. (F) HeLa cells were co-transfected with TRS1 and either WT PPP1CA or PPP1CA F293A and the presence of pTRS1 in PP1-specific immune complexes was determined by Western blot.
161
Figure 4.5. pTRS1 does not affect PP1Cα phosphatase activity. (A) HeLa cells were transfected with pTRS1 and a phosphatase reaction was ran on pTRS1-specific or control immune complexes. Free phosphate was detected using a malachite green assay. (B) HeLa cells were co-transfected with PPP1CA and either GFP or TRS1.
Phosphatase assays were run on PP1Cα-specific immune complexes with increasing concentrations of phosphopeptide. Free phosphate concentrations were determined using a malachite green assay.
162
Figure 4.6. pTRS1 changes the complement of proteins that interact with PP1Cα.
(A) HeLa cells were transfected with PPP1CA or PPP1CA BioID. PP1 was immunoprecipitated and the level of phosphatase activity associated with PP1-specific immune complexes was determined. (B) WT or PKR KO HeLa cells were co-transfected with PPP1CA BioID and GFP or TRS1. Cells were treated with biotin, lysed and biotinylated proteins isolated by streptavidin affinity chromatography. Biotinylated proteins were identified by Western blot analysis.
163 TABLES
Table 4.1. Primer Sequences
Primer Name Sequence (5'à3') CACTACCATTACAATGCTCAAGAACAAAAACTCATCTCAG TRS1BirA F AAGAGG CGCGGGCCCTCTAGACTCGATCACTTCTCTGCGCTTCTC BirA R AGGGAG TAAGCTTGGTACCGAGCTCGATGGCCCAGCGCAACGGCA TRS1 F TG CTTCTGAGATGAGTTTTTGTTCTTGAGCATTGTAATGGTA TRS1BirA R GTG GACATGCGAGCAGCTGCAGCTTCTAATGTCTAATGTAGC RVTF F CACACACTACC CATTAGAAGCTGCAGCTGCTCGCATGTCGCGGCACAATC RVTF R TGGCAGC TAAGCTTGGTACCGAGCTCGGCCGCCACCATGTCCGACA PPP1CA mycF GCGAGAAGCTC CGCGGGCCCTCTAGACTCGACTACAGATCCTCTTCTGAG PPP1CA mycR ATGAGTTTTTGTTCTTTCTTGGCTTTGGCGGAATTGC TAAGCTTGGTACCGAGCTCGGCCGCCACCATGGCGGAC PPP1CB mycF GGGGAGCTGAACG CGCGGGCCCTCTAGACTCGATCACAGATCCTCTTCTGAG PPP1CB mycR ATGAGTTTTTGTTCCCTTTTCTTCGGCGGATTAGC TAAGCTTGGTACCGAGCTCGGCCGCCACCATGGCGGATT PPP1CB mycF TAGATAAACTC CGCGGGCCCTCTAGACTCGACTACAGATCCTCTTCTGAG PPP1CB mycR ATGAGTTTTTGTTCTTTCTTTGCTTGCTTTGTGATC CATGTGCTCTGCTCAGATCCTCAAGCCCGCCGACAAGAA PP1 F293A F C GAGGATCTGAGCAGAGCACATGAGGGTCTCGTCCACACT PP1 F293A R C TAAGCTTGGTACCGAGCTCGGCCGCCACCATGGAACAAA BiRA pcD F AACTCATCTCAG BirA R2 CTCGAGCTTCTCTGCGCTTCTC PP1CA BioID GAGAAGCGCAGAGAAGCTCGAGTCCGACAGCGAGAAGC F TCAACC PP1CA BioID CGCGGGCCCTCTAGACTCGACTATTTCTTGGCTTTGGCG R GAATTGC
164 REFERENCES
1. Landolfo S, Gariglio M, Gribaudo G, Lembo D. 2003. The human cytomegalovirus. Pharmacology & Therapeutics 98:269-297.
2. Pereira L, Maidji E, Fisher SJ, McDonagh S, Tabata T. 2007. HCMV persistence in the population: potential transplacental transmission. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge.
3. McArdle J, Moorman NJ, Munger J. 2012. HCMV targets the metabolic stress response through activation of AMPK whose activity is important for viral replication. PLoS pathogens 8:e1002502.
4. Arend KC, Lenarcic EM, Vincent HA, Rashid N, Lazear E, McDonald IM, Gilbert TSK, East MP, Herring LE, Johnson GL, Graves LM, Moorman NJ. 2017. Kinome Profiling Identifies Druggable Targets for Novel Human Cytomegalovirus (HCMV) Antivirals. Molecular & cellular proteomics: MCP 16:S263-S276.
5. Spector DH. 2015. Human cytomegalovirus riding the cell cycle. Medical Microbiology and Immunology 204:409-419.
6. Vincent HA, Ziehr B, Moorman NJ. 2016. Human Cytomegalovirus Strategies to Maintain and Promote mRNA Translation. Viruses 8.
7. Alwine JC. 2008. Modulation of host cell stress responses by human cytomegalovirus. Current Topics in Microbiology and Immunology 325:263-279.
8. Vincent HA, Ziehr B, Moorman NJ. 2017. Mechanism of Protein Kinase R Inhibition by Human Cytomegalovirus pTRS1. Journal of Virology 91.
9. Ziehr B, Vincent HA, Moorman NJ. 2016. Human Cytomegalovirus pTRS1 and pIRS1 Antagonize Protein Kinase R To Facilitate Virus Replication. Journal of Virology 90:3839-3848.
165 10. Braggin JE, Child SJ, Geballe AP. 2016. Essential role of protein kinase R antagonism by TRS1 in human cytomegalovirus replication. Virology 489:75-85.
11. Hakki M, Marshall EE, De Niro KL, Geballe AP. 2006. Binding and nuclear relocalization of protein kinase R by human cytomegalovirus TRS1. Journal of Virology 80:11817-11826.
12. Lemaire PA, Anderson E, Lary J, Cole JL. 2008. Mechanism of PKR Activation by dsRNA. Journal of molecular biology 381:351-360.
13. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition. Cell 122:901-913.
14. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. 2001. Tight Binding of the Phosphorylated Subunit of Initiation Factor 2 (eIF2 ) to the Regulatory Subunits of Guanine Nucleotide Exchange Factor eIF2B Is Required for Inhibition of Translation Initiation. Molecular and Cellular Biology 21:5018-5030.
15. Singh CR, Udagawa T, Lee B, Wassink S, He H, Yamamoto Y, Anderson JT, Pavitt GD, Asano K. 2007. Change in Nutritional Status Modulates the Abundance of Critical Pre-initiation Intermediate Complexes During Translation Initiation in Vivo. Journal of Molecular Biology 370:315-330.
16. Rojas M, Vasconcelos G, Dever TE. 2015. An eIF2 -binding motif in protein phosphatase 1 subunit GADD34 and its viral orthologs is required to promote dephosphorylation of eIF2 . Proceedings of the National Academy of Sciences of the United States of America 112:E3466-E3475.
17. Shors T, Kibler KV, Perkins KB, Seidler-Wulff R, Banaszak MP, Jacobs BL. 1997. Complementation of Vaccinia Virus Deleted of the E3L Gene by Mutants of E3L. Virology 239:269-276.
18. Chang H-W, Jacobs BL. 1993. Identification of a Conserved Motif That Is Necessary for Binding of the Vaccinia Virus E3L Gene Products to Double- Stranded RNA. Virology 194:537-547.
166 19. Seo EJ, Liu F, Kawagishi-Kobayashi M, Ung TL, Cao C, Dar AC, Sicheri F, Dever TE. 2008. Protein kinase PKR mutants resistant to the poxvirus pseudosubstrate K3L protein. Proceedings of the National Academy of Sciences of the United States of America 105:16894-16899.
20. Cassady KA, Gross M, Roizman B. 1998. The Herpes Simplex Virus US11 Protein Effectively Compensates for the 134.5 Gene if Present before Activation of Protein Kinase R by Precluding Its Phosphorylation and That of the Subunit of Eukaryotic Translation Initiation Factor 2. Journal of Virology 72:8620-8626.
21. Peters GA, Khoo D, Mohr I, Sen GC. 2002. Inhibition of PACT-mediated activation of PKR by the herpes simplex virus type 1 Us11 protein. Journal of Virology 76:11054-11064.
22. Cassady KA, Gross M. 2002. The Herpes Simplex Virus Type 1 US11 Protein Interacts with Protein Kinase R in Infected Cells and Requires a 30-Amino-Acid Sequence Adjacent to a Kinase Substrate Domain. Journal of Virology 76:2029- 2035.
23. Li Y, Zhang C, Chen X, Yu J, Wang Y, Yang Y, Du M, Jin H, Ma Y, He B, Cao Y. 2011. ICP34.5 protein of herpes simplex virus facilitates the initiation of protein translation by bridging eukaryotic initiation factor 2alpha (eIF2alpha) and protein phosphatase 1. The Journal of Biological Chemistry 286:24785-24792.
24. Kazemi S, Papadopoulou S, Li S, Su Q, Wang S, Yoshimura A, Matlashewski G, Dever TE, Koromilas AE. 2004. Control of alpha subunit of eukaryotic translation initiation factor 2 (eIF2 alpha) phosphorylation by the human papillomavirus type 18 E6 oncoprotein: implications for eIF2 alpha- dependent gene expression and cell death. Molecular and Cellular Biology 24:3415-3429.
25. Zhang F, Moon A, Childs K, Goodbourn S, Dixon LK. 2010. The African Swine Fever Virus DP71L Protein Recruits the Protein Phosphatase 1 Catalytic Subunit To Dephosphorylate eIF2 and Inhibits CHOP Induction but Is Dispensable for These Activities during Virus Infection. Journal of Virology 84:10681-10689.
26. Hakki M, Geballe AP. 2008. Cellular serine/threonine phosphatase activity during human cytomegalovirus infection. Virology 380:255-263.
167 27. Ziehr B, Lenarcic E, Vincent HA, Cecil C, Garcia B, Shenk T, Moorman NJ. 2015. Human cytomegalovirus TRS1 protein associates with the 7- methylguanosine mRNA cap and facilitates translation. Proteomics 15:1983- 1994.
28. Hakki M, Geballe AP. 2005. Double-stranded RNA binding by human cytomegalovirus pTRS1. Journal of Virology 79:7311-7318.
29. Romanowski MJ, Shenk T. 1997. Characterization of the human cytomegalovirus irs1 and trs1 genes: a second immediate-early transcription unit within irs1 whose product antagonizes transcriptional activation. Journal of Virology 71:1485-1496.
30. Zhu H, Shen Y, Shenk T. 1995. Human cytomegalovirus IE1 and IE2 proteins block apoptosis. Journal of Virology 69:7960-7970.
31. Strang BL, Geballe AP, Coen DM. 2010. Association of human cytomegalovirus proteins IRS1 and TRS1 with the viral DNA polymerase accessory subunit UL44. The Journal of General Virology 91:2167-2175.
32. Roux KJ, Kim DI, Raida M, Burke B. 2012. A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. The Journal of Cell Biology 196:801-810.
33. Brook M, McCracken L, Reddington JP, Lu Z-L, Morrice NA, Gray NK. 2012. The multifunctional poly(A)-binding protein (PABP) 1 is subject to extensive dynamic post-translational modification, which molecular modelling suggests plays an important role in co-ordinating its activities. The Biochemical Journal 441:803-812.
34. Norbury CJ. 2013. Cytoplasmic RNA: a case of the tail wagging the dog. Nature Reviews Molecular Cell Biology 14:643-653.
35. Richter JD. 2007. CPEB: a life in translation. Trends in Biochemical Sciences 32:279-285.
36. Peti W, Nairn AC, Page R. 2013. Structural basis for protein phosphatase 1 regulation and specificity. The FEBS journal 280:596-611.
168 37. Roy J, Cyert MS. 2009. Cracking the Phosphatase Code: Docking Interactions Determine Substrate Specificity. Sci Signal 2:re9-re9.
38. Hinnebusch AG, Ivanov IP, Sonenberg N. 2016. Translational control by 5 - untranslated regions of eukaryotic mRNAs. Science 352:1413-1416.
39. Lenarcic EM, Ziehr BJ, Moorman NJ. 2015. An unbiased proteomics approach to identify human cytomegalovirus RNA-associated proteins. Virology 481:13-23.
40. Sonenberg N, Hinnebusch AG. 2009. Regulation of translation initiation in eukaryotes: mechanisms and biological targets. Cell 136:731-745.
41. Shi Y. 2009. Serine/threonine phosphatases: mechanism through structure. Cell 139:468-484.
169 CHAPTER 5: DISCUSSION
In this chapter I summarize the work presented in my thesis, and discuss how my findings have advanced our knowledge of the role of pTRS1 in mRNA translation regulation. In addition I discuss several outstanding questions raised by my work, and the future studies needed to answer those questions, to further our understanding of translation regulation in the control of HCMV protein synthesis.
Summary of my findings
The HCMV protein TRS1 (pTRS1) is packaged in the tegument of the HCMV virion (1). pTRS1 is first expressed at immediate early times during infection, and pTRS1 abundance increases as infection progresses (2). Thus pTRS1 is present throughout the entire HCMV replication cycle. Both pTRS1 and its functional homologue pIRS1 are necessary for virus replication in primary human fibroblasts, as a virus lacking both pTRS1 and pIRS1 (ΔTRS1/ΔIRS1) does not replicate. Infection with
ΔTRS1/ΔIRS1 results in robust PKR activation and eIF2α phosphorylation, leading to a significant inhibition of overall levels of protein synthesis, and decreased expression of viral proteins (3, 4). Our group and others have shown that pTRS1 binding to PKR is necessary for PKR antagonism during infection (4, 5), however the molecular mechanism behind PKR inhibition was unknown. In chapter 2 I showed that the ability of pTRS1 to bind PKR is necessary for inhibition of PKR activation in vitro (Figure 2.2)(6). I found that pTRS1 does not sequester RNA ligands from PKR or prevent PKR
170 dimerization (Figure 2.3, 2.4). Rather, pTRS1 interacts with conserved residues in the
PKR kinase domain that are necessary for PKR to bind its substrate eIF2α (Figure 2.5).
I also showed that pTRS1 binding to PKR was necessary for pTRS1 to inhibit PKR kinase activity (Figure 2.6). Lastly, I found that pTRS1 limits eIF2α phosphorylation in response to activation of an additional eIF2α kinase, in cells lacking PKR (Figure 2.7), suggesting that pTRS1 may antagonize multiple eIF2α kinases during infection.
pTRS1 also promotes cap-dependent translation, independent of its ability to inhibit PKR. pTRS1 binds the m7G mRNA cap, stimulates the translation of reporter mRNAs containing viral or cellular 5’UTRs, and increases overall levels of protein synthesis (7). In chapter 3 I found that pTRS1 also stimulates cap-independent translation. pTRS1 enhanced the activity of multiple IRESs in a bicistronic reporter assay (Figure 3.1, 3.2), however, bicistronic reporter assays are prone to false positives. I therefore developed a new assay for IRES activity that circumvents several of the caveats of bicistronic reporters. Namely, this assay tests the ability of a purported
IRES to drive the translation of a circular mRNA lacking both an m7G mRNA cap and a poly(A) tail. Using this assay, I found that pTRS1 enhances the translation of circular
RNA (circRNA) reporters containing either viral or cellular IRESs (Figure 3.3, 3.4).
Importantly, the increase in translation was not solely due to PKR antagonism, as pTRS1 enhanced circular RNA translation in cells lacking PKR (Figure 3.5). Rather the ability of pTRS1 to enhance circRNA translation was dependent on the ability of pTRS1 to bind dsRNA (Figure 3.6). Together these data show that pTRS1 enhances both cap- dependent and cap-independent translation through a PKR-independent mechanism.
171 To begin to define the mechanism by which pTRS1 stimulates translation independent of PKR antagonism, in chapter 4 I identified pTRS1 cellular binding partners using the BioID approach (8). I fused a promiscuous biotin ligase domain (BirA) to the carboxyl-terminus of pTRS1 (pTRS1 BioID), and expressed pTRS1 BioID in the presence of exogenous biotin. Cellular proteins that interact with pTRS1 BioID, even transiently, are covalently biotinylated, and can be isolated with streptavidin affinity chromatography and identified by mass spectrometry (Figure 4.1). To identify pTRS1 interactors likely involved in translation, we isolated biotinylated proteins associated with ribosomes in cells expressing pTRS1 BioID. We identified 114 pTRS1-interacting partners, including the alpha, beta and gamma catalytic subunits of protein phosphatase
1 (PP1) (Figure 4.2). I confirmed that pTRS1 interacts each PP1 catalytic subunit in vitro, and interacts with the PP1Cβ subunit during HCMV infection (Figure 4.4). I found that phosphatase activity is associated with pTRS1-specific immune complexes, but that pTRS1 does not change the rate of PP1Cα catalytic activity (Figure 4.5). Rather, my data show that pTRS1 changes the complement of proteins that interact with
PP1Cα (Figure 4.6), suggesting that pTRS1 alters the structure of the PP1 holoenzyme, and potentially PP1 substrate specificity.
In summary, I have defined the molecular mechanism by which pTRS1 antagonizes PKR, identified an additional pTRS1 function (PKR-independent stimulation of cap-independent mRNA translation), and identified pTRS1-interacting proteins that are potentially important for HCMV replication. Below I discuss the outstanding questions that my data has raised, and how these questions might be addressed.
172 Outstanding questions related to pTRS1 and PKR
What are the dsRNA ligands that activate PKR during HCMV infection? In the absence of both pTRS1 and pIRS1, PKR is activated within 24 hours of infection (4).
However, the ligands responsible for PKR activation in HCMV-infected cells have not been identified. Binding to viral dsRNA activates PKR (9), and viral dsRNA accumulates during HCMV infection, presumably due to the presence of overlapping transcriptional units on opposite strands of the HCMV genome (3). These viral dsRNAs could act as
PKR ligands in the absence pTRS1/pIRS1. In addition, viral RNA packaged within the tegument of the virus could also activate PKR upon infection, although the complement of RNAs packaged within the virion is unknown (10). Thus, it is possible that RNAs packaged within the virion, newly transcribed viral RNA, or a combination of both, act as
PKR ligands in the absence of pTRS1 and pIRS1.
To determine which ligands activate PKR during HCMV infection we could examine PKR activation immediately after infection with a ΔTRS1/ΔIRS1 virus, before viral transcription begins (<3 hpi). If PKR were activated this early, it would suggest that viral RNAs packaged within the tegument are the PKR activating ligands. We could confirm this by examining PKR activation in the presence of the transcription inhibitor actinomycin D (ActD). If PKR is not active at very early times post infection with
ΔTRS1/ΔIRS1, or in the presence of ActD, it would suggest that newly synthesized viral
RNA activates PKR. RIP-seq or HITS-CLIP could be used to identify the RNAs bound by PKR in HCMV-infected cells. The RNAs that activate PKR have not been identified in any viral infection. Thus identifying HCMV RNAs bound by PKR would provide insight
173 into structural or sequence similarities that PKR uses to discriminate between host and viral RNAs.
Does pTRS1 prevent PKR binding to eIF2α? My data show that pTRS1 inhibits
PKR kinase activity to prevent PKR activation (Figure 2.6), and also suggest that pTRS1 blocks the interaction of PKR with its substrate eIF2α. Both pTRS1 and eIF2α bind the Thr487 residue in the PKR kinase domain. When Thr487 is mutated to an alanine residue, PKR undergoes activating autophosphorylation, but cannot bind or phosphorylate eIF2α (Figure 2.5)(6, 11). pTRS1 binds PKR Thr487, suggesting it prevents PKR binding to eIF2α. The vaccinia virus K3L protein uses a similar mechanism; K3L acts as a PKR pseudosubstrate and prevents the interaction of PKR with eIF2α (12). However, since pTRS1 inhibits PKR kinase activity, it seems unlikely that pTRS1 functions as an eIF2α pseudosubstrate. Rather I hypothesize that pTRS1 serves two roles, both blocking PKR autophosphorylation and acting as a physical barrier to PKR binding to eIF2α. To test this hypothesis we could conduct in vitro binding assays using purified proteins to determine if pTRS1 blocks an interaction between activated PKR and eIF2α. This would allow us to fully understand how pTRS1 prevents eIF2α phosphorylation by binding to PKR.
Does pTRS1 antagonize multiple eIF2α kinases during HCMV infection? The mechanism by which pTRS1 inhibits PKR suggests that pTRS1 may also inhibit additional eIF2α kinases. My data shows that pTRS1 binds HRI and limits eIF2α phosphorylation in response to an HRI agonist (Figure 2.7)(6). Thus in addition to inhibiting PKR, pTRS1 may also limit HRI activation. pTRS1 interacts with amino acids in the αG helix of the PKR kinase domain that are conserved in all four eIF2α kinases
174 (Figure 2.5). This suggests that pTRS1 can bind and inhibit multiple eIF2α kinases.
While I found that pTRS1 binds both PKR and HRI during HCMV infection (Figure 2.2,
2.7), it remains to be determined if pTRS1 also interacts with PERK and GCN2. We could test this by looking for the presence of PERK and GCN2 in pTRS1-specific immune complexes isolated from infected cells. If pTRS1 does interact with multiple eIF2α kinases we could then introduce the synonymous PKR Thr487 mutation into the other eIF2α kinases and see if this disrupts pTRS1 binding in vitro. If this mutation disrupts binding it would suggest that pTRS1 binds all eIF2α kinases through a conserved mechanism. Experiments measuring the activation of HRI, PERK, and GCN2 in the presence of pTRS1 would then be needed to determine if pTRS1 binding inhibits activation of the other eIF2α kinases, and potentially prevents their interaction with eIF2α. These experiments would further our understanding of how pTRS1 regulates stress response pathways that could inhibit translation during HCMV infection.
What drives eIF2α phosphorylation during the late stage of HCMV replication?
Whether or not pTRS1 inhibits multiple eIF2α kinases, eIF2α becomes phosphorylated during the late stage of HCMV replication (13-15). Thus an outstanding question is what causes eIF2α phosphorylation late in infection? PKR remains inactive throughout infection, showing that pTRS1/pIRS1 effectively antagonize PKR (14). In addition,
PERK depletion does not prevent the increase in eIF2α phosphorylation (13). Thus presumably either HRI or GCN2 are activated during infection and phosphorylate eIF2α.
However the activation state of HRI and GCN2 throughout infection has not been examined. To determine which kinase is responsible for eIF2α phosphorylation, we could measure the activation of HRI and GCN2 across a time course of infection. We
175 could then measure the ability of HCMV to induce eIF2α phosphorylation at late times after infection in cells where HRI or GCN2 expression has been ablated using
CRISPR/Cas9 mutagenesis. It is also possible that multiple eIF2α kinases contribute to eIF2α phosphorylation. In this case, the combined disruption/depletion of both HRI and
GCN2 would be needed to limit eIF2α phosphorylation. This will allow us to determine which eIF2α kinase phosphorylates eIF2α, and what kind of cellular stress leads to eIF2α phosphorylation during HCMV replication.
These data also suggest that HRI and/or GCN2 escape inhibition by pTRS1.
Because pTRS1 binds HRI and limits eIF2α phosphorylation in response to an HRI agonist (Figure 2.7), it is probable that pTRS1 binds HRI and inhibits its activation and/or binding to eIF2α during infection (6). Unlike the other three eIF2α kinases, GCN2 contains a pseudokinase domain in its amino terminus that binds the GCN2 protein kinase domain and prevents its activation. The GCN2 pseudokinase domain is homologous in structure to eIF2α protein kinase domains (16), suggesting it could be bound by pTRS1. Perhaps pTRS1 binds the pseudokinase domain rather than the active kinase domain. To test this hypothesis we could remove the GCN2 pseudokinase domain and see if the deletion disrupts pTRS1 binding to GCN2. If pTRS1 interacts with the GCN2 pseudokinase domain it could allow GCN2 to become activated in the presence of pTRS1 and phosphorylate eIF2α. Understanding how pTRS1 interacts with eIF2α kinases will allow us to determine the role of pTRS1 in regulating stress response pathways during HCMV replication.
What is the role of pIRS1 during infection? pTRS1 and pIRS1 are partially encoded by the repeat regions flanking the US region of the HCMV genome. As a result,
176 the amino terminal 550 amino acids of the two proteins are identical (17). While the carboxyl terminus of pTRS1 and pIRS1 differ, this region of both proteins binds PKR (2).
However the contribution of the pIRS1 PKR binding domain during infection has not been examined, nor has the mechanism by which pIRS1 prevents PKR activation.
Since the majority of pIRS1 and pTRS1 are homologous, and pIRS1 contains the residues necessary for PKR binding, it is probable that pIRS1 inhibits PKR through the same mechanism as pTRS1. Therefore, why does HCMV encode two proteins with seemingly identical functions? Expression of either pTRS1 or pIRS1 alone is sufficient to inhibit PKR activation (4). In addition, recombinant viruses lacking either TRS1 or
IRS1 replicate nearly as efficiently as wild type HCMV; however, the replication of
ΔTRS1 and ΔIRS1 viruses has only been characterized in unstressed cells (18). HCMV induces the expression of type I interferons (IFNs), which upregulate the expression of antiviral interferon stimulated genes (ISGs), including PKR, in neighboring uninfected cells (19-21). Perhaps the presence of both pIRS1 and pTRS1 is necessary when infecting a cell with a higher baseline level of PKR expression. To test this we could pre- treat cells with type I IFNs before infection with a ΔTRS1, ΔIRS1, or wild type virus. If the presence of both pIRS1 and pTRS1 is necessary to establish infection in cells previously primed to express ISGs, we would expect a defect in ΔTRS1 and ΔIRS1 replication compared to a wild type infection. We could also examine PKR activation during these infections. If PKR activation is observed after infection with ΔTRS1 or
ΔIRS1, it would suggest that the presence of both pTRS1 and pIRS1 is necessary to inhibit PKR activation in cells primed to express ISGs. Thus the redundant functions of
177 pTRS1 and pIRS1 may provide a fail safe for the virus when spreading to uninfected,
IFN-primed cells in vivo.
In addition to PKR, type I IFNs induce the expression of multiple innate immune sensors that recognize viral RNA and contribute to the antiviral response (21). Both pTRS1 and pIRS1 have non-canonical RNA binding domains in their amino terminus, and are packaged in the tegument of the virion (1, 22, 23). Thus pTRS1 and pIRS1 together could be necessary to sequester RNA ligands from innate immune sensors in cells previously exposed to type I IFNs. Such a scenario is probable in vivo, as HCMV- infected cells produce significant levels of type I IFNs (19). Comparing HCMV replication in cells pre-treated with type I IFNs and infected with wild type virus or a recombinant virus containing mutations in pTRS1/pIRS1 that disrupt dsRNA binding would shed light on the function of pTRS1/pIRS1 dsRNA binding during infection.
Sequestration of RNA ligands by pTRS1 and pIRS1 could be an additional HCMV strategy to block innate immune activation early in infection.
How does pTRS1 stimulate translation in a PKR-independent manner? pTRS1 promotes translation through both PKR-dependent and PKR-independent mechanisms
(Figure 3.5)(7). While the ability of pTRS1 to enhance translation is dampened in cells lacking PKR, pTRS1 increases overall levels of protein synthesis and the translation of reporter genes in PKR deficient cells. Thus inhibition of PKR alone cannot fully account for the ability of pTRS1 to enhance translation. The mechanism by which pTRS1 stimulates translation in a PKR-independent manner remains unknown. To begin to understand how pTRS1 promotes translation, in chapter 4 I identified ribosome- associated cellular proteins that interact with pTRS1. Based on these data I have two
178 hypotheses for how pTRS1 enhances translation: (1) pTRS1 antagonizes additional signaling pathways that inhibit translation or (2) pTRS1 facilitates the recruitment of ribosomes to specific mRNAs.
Hypothesis 1: pTRS1 antagonizes additional signaling pathways that inhibit translation. In support of my first hypothesis, I found that pTRS1 interacts with all three catalytic subunits of protein phosphatase 1 (PP1) (Figure 4.2). PP1 is a serine/threonine phosphatase that regulates many cellular signaling pathways, including the eIF2α pathway (24). The heterotrimeric PP1 holoenzyme consists of a catalytic subunit bound by two regulatory subunits that control the activity and/or specificity of the catalytic subunit (25, 26). I confirmed that pTRS1 interacts with all three PP1 catalytic subunits in transfected cells, and PP1Cβ during HCMV infection (Figure 4.4). pTRS1 did not change PP1Cα catalytic activity (Figure 4.5), but appeared to change the complement of proteins that interact with PP1Cα (Figure 4.6). PP1 is recruited to dephosphorylate eIF2α by GADD34 (24). Recruitment of GADD34 to eIF2α by viral proteins is a common strategy used by several viruses to dephosphorylate eIF2α during infection (27-29).
However pTRS1 does not seem to use this strategy, as we did not identify GADD34 as a pTRS1-associated protein in our proteomics screen. Rather, we hypothesize that the interaction of pTRS1 with PP1 changes the substrate specificity of the PP1 holoenzyme to enhance recruitment to eIF2α. Mass spectrometry analysis of proteins that interact with PP1Cα in the presence or absence of pTRS1 is currently ongoing to identify how pTRS1 may affect PP1 substrate specificity.
Another possibility is that pTRS1 nucleates the formation of a unique PP1 complex in infected cells. Perhaps pTRS1 recruits additional cellular or viral proteins to
179 the PP1 holoenzyme. To test this, we could use a tandem immunoaffinity approach to identify proteins associated with both PP1 catalytic subunits and pTRS1. We would construct a TRS1 BioID virus, and then infect cells with the pTRS1 BioID virus in the presence of exogenous biotin. Following immunoprecipitation of PP1 from HCMV TRS1
BioID-infected cells, we would then isolate biotinylated proteins from PP1 immune complexes. The recovered proteins would thus interact with both PP1 and pTRS1, and could potentially form a novel PP1 holoenzyme found only in infected cells. While PP1 dephosphorylates eIF2α, it also regulates many other cellular signaling pathways, which could be impacted by its association with pTRS1. Phosphoproteomics analyses of cells infected with HCMV ΔTRS1/ΔIRS1 could allow us to correlate changes in PP1 binding partners with changes to cellular signaling pathways observed during HCMV replication.
Hypothesis 2: pTRS1 facilitates the recruitment of ribosomes to specific mRNAs.
My analysis of pTRS1 interacting partners also supports my second hypothesis that pTRS1 facilitates the recruitment of ribosomes to specific mRNAs. Myself and others have shown that pTRS1 binds dsRNA, the m7G mRNA cap, multiple RNA helicases, translation initiation factors, and the poly(A) binding protein (PABP) (Figure 4.2, 4.3)(7,
22, 23). These interactions suggest that pTRS1 promotes mRNA translation initiation as a mechanism to stimulate translation. We could test this hypothesis by measuring the effect of pTRS1 on the formation of 48S translation initiation complexes. The 48S complex consists of multiple translation initiation factors, the 40S ribosomal subunit, and
Met the ternary complex, which consists of eIF2-GTP and tRNAi . The 48S scans an mRNA 5’UTR until recognizing the translation start codon, whereupon hydrolysis of the eIF2-bound GTP promotes 60S binding and the release of translation initiation factors
180 (30). Treating translation competent lysates with GMP-PNP prevents 60S binding and allows for the isolation of 48S complexes by sucrose density centrifugation. To determine if pTRS1 promotes translation initiation, we could see if translation competent cell lysates from cells expressing pTRS1 support increased levels of 48S formation on a reporter mRNA. If pTRS1 increases 48S formation, it would suggest that pTRS1 recruits translation initiation factors and ribosomal subunits to that mRNA to stimulate translation. This could be a mechanism used by pTRS1 to enhance the translation of target mRNAs during HCMV infection.
To determine if pTRS1 binds mRNAs and enhances their translation during infection we need to first identify which RNAs are bound by pTRS1 using a RIP-seq experiment. Once the pTRS1-bound mRNAs have been identified we could infect cells with a ΔTRS1 virus and see if we observe a change in the translation efficiency of pTRS1-bound mRNAs in the absence of pTRS1 expression using ribosome profiling. If there is a decrease in the translation efficiency of pTRS1-bound mRNAs it suggests that pTRS1 enhances the translation of target mRNAs during HCMV replication. If we observe no change in the translation efficiency of pTRS1-target mRNAs it would suggest that stimulation of translation is a redundant function between pTRS1 and pIRS1. If this were the case, we would repeat the experiment with a ΔTRS1/ΔIRS1 virus in PKR KO cells. We could then repeat the 48S accumulation experiment described above, but spike a labeled pTRS1-target mRNA into translation competent lysates from cells infected with wild type, ΔTRS1, or a ΔTRS1/ΔIRS1 virus. If we only see an accumulation of 48S complexes in lysates from cells infected with wild type virus it would suggest that pTRS1, and possibly pIRS1, promote translation initiation on specific
181 mRNAs. This mechanism could ensure that pTRS1-bound mRNAs are efficiently translated during HCMV infection when the availability of translation initiation factors and ribosomal subunits is limiting.
I also found that pTRS1 promotes the activity of both cellular and viral IRESs
(chapter 3). Importantly, pTRS1 enhances IRES activity in the absence of PKR expression (Figure 3.5), thus the enhancement of IRES activity is not solely due to PKR inhibition. HCMV enhances the expression of several cellular proteins required for efficient virus replication that are encoded on IRES-containing mRNAs (31-33). For example, the BiP mRNA contains an IRES whose activity increases during HCMV infection. BiP is necessary for efficient virus replication, and we found that pTRS1 enhances BiP IRES activity (Figure 3.4)(34-36). Therefore, pTRS1 expression may be necessary for the increased translation of cellular mRNAs containing IRESs during
HCMV infection. pTRS1 could also facilitate the translation of IRES-containing HCMV mRNAs. While only a single HCMV IRES has been identified to date, ~100 HCMV
5’UTRs were recently shown to possess IRES activity in an in vitro screen (37, 38). This suggests that the role of cap-independent translation in HCMV protein synthesis may be more extensive than currently appreciated, and that the ability of pTRS1 to stimulate cap-independent translation could affect the expression of multiple viral proteins.
The pTRS1 RIP-Seq data (described above) will determine if pTRS1 enhances the translation of mRNAs containing IRES elements during infection. However, it will not reveal the RNA sequences and/or structures needed for pTRS1 to bind IRES elements.
Because pTRS1 stimulates the activity of multiple IRESs that share little sequence homology (chapter 3), and the stimulation of IRES activity is dependent on pTRS1
182 dsRNA binding (Figure 3.6), I hypothesize that pTRS1 binds common structural elements within these IRESs. We could test this hypothesis using HITS-CLIP to determine where pTRS1 binds IRES-containing mRNAs. We could then determine if pTRS1-binding sites share similar RNA secondary structure using mFold or SHAPE-
MaP analysis. This experiment would also allow us to determine if pTRS1 binding to all mRNAs is mediated through an interaction with a specific RNA secondary structure or specific RNA sequences. Thus pTRS1 could promote the translation of target mRNAs during HCMV replication by binding a common RNA secondary structure and/or sequence and recruiting translation initiation factors and ribosomal subunits to facilitate translation initiation.
The ability of pTRS1 to stimulate translation on RNAs that lack a m7G mRNA cap suggests that pTRS1 could play a role in the eIF4F-independent translation of viral mRNAs (chapter 3). The eIF4F complex binds the m7G mRNA cap and recruits 43S translation initiation complexes to the 5’ end of the mRNA to begin 5’UTR scanning (30). eIF4F formation is heavily regulated, and the eIF4F complex is required for the translation of many cellular mRNAs (39-41). eIF4F component expression and the abundance of assembled eIF4F complexes increase during HCMV infection (42, 43). In fact, the eIF4F complex is necessary for virus replication early in infection (43, 44).
However, we previously showed that the translation of HCMV mRNAs becomes resistant to eIF4F complex disruption or depletion late in infection (44). Thus HCMV mRNAs can recruit ribosomes and translation initiation factors through an eIF4F- independent mechanism. Based on my data showing that pTRS1 stimulates translation independent of an m7G cap, I hypothesize that pTRS1 also stimulates translation
183 independently of eIF4F. We could test this by determining if translation of viral mRNAs remains resistant to eIF4F disruption following infection with a ΔTRS1/ΔIRS1 virus. We would perform this experiment in PKR KO cells to control for the effect of pTRS1 inhibition of PKR activation. The ability of pTRS1 to promote translation of HCMV mRNAs in the absence of the eIF4F complex could allow viral protein synthesis to escape regulation by the cell, and limit competition between host and viral mRNAs for translation initiation factors. Thus if pTRS1 facilitates the eIF4F-independent translation of HCMV mRNAs, this likely plays a significant role in HCMV spread and pathogenesis within a host.
Outstanding questions pertaining to translation regulation during HCMV infection
What is the complement of full-length transcripts produced during HCMV replication? We have shown that pTRS1 stimulates the translation of mRNAs containing both cellular and viral 5’UTRs, with a greater effect on mRNAs containing viral 5’UTRs
(chapter 3)(7). This data suggests that pTRS1 binds a specific RNA secondary structure and/or sequence within mRNA 5’UTRs that mediates the enhancement of their translation. Above I proposed to use HITS-CLIP to identify the structure and/or sequence that give pTRS1 specificity to target mRNAs. A caveat of this approach is that the full-length HCMV transcriptome has not been defined. While RNA-Seq has revealed that the majority of the HCMV genome on both strands is transcribed, the presence of overlapping and antisense transcription units complicates the identification of the ends of HCMV mRNAs (45). As a result we have limited knowledge of the 5’ and 3’UTRs of
HCMV mRNAs, which play important roles in regulating translation (46, 47). Therefore I
184 could define the viral sequences bound by pTRS1, but might not be able to identify the full complement of viral mRNAs that contain that sequence. Thus we need to define the full-length HCMV transcriptome to understand how pTRS1 regulates mRNA translation.
We also need define the full-length HCMV transcriptome in order to understand how HCMV mRNA translation is regulated during infection. Ribosome profiling data from Stern-Ginossaur et al. redefined the number of HCMV ORFs, increasing the number of peptide coding regions from a previous estimate of ~200 to more than 750
(48). Thus, the HCMV proteome was shown to be much more complex than previously thought. Translation is largely regulated at the 5’ and 3’UTR of an mRNA. 5’UTRs serve as binding sites for factors that regulate translation initiation, while sequences in the
3’UTR play important roles in mRNA stability and localization (46, 47). Thus, to fully understand how the translation of an HCMV transcript is controlled during replication, and the potential affect(s) of pTRS1 on that transcript, we must define the full sequence of that transcript, from m7G cap to poly(A) tail.
As mentioned above, standard RNA-Seq approaches are not sufficient to define the full-length HCMV transcriptome. Gatherer et al. used short (2X75 bp) sequencing reads to show that nearly the entirety of both stands of the HCMV genome is transcribed during infection (45). However the short sequencing reads in this dataset were not capable of detecting full-length transcripts. Current high throughput sequencing reads (2x300 bp) are also not long enough to capture the full length of all
HCMV transcripts, which in some cases exceed 5,000 bp in length (49). In addition, long-read sequencing platforms, such as PacBio, provide many fewer reads and have very high error rates compared to high throughput approaches. Thus neither approach
185 alone is sufficient to define the full-length HCMV transcriptome. To overcome these problems, we have used a hybrid-sequencing approach that combines these two technologies. This approach uses long-read sequencing data to define full-length transcripts. High throughput ‘short reads’, which have very low error rates, are then mapped to full-length transcripts to define the exact sequence of the transcript.
Together, long-read and short-read datasets allow us to define full-length transcripts with certainty.
Once full-length transcripts are identified, the published ribosome profiling dataset will allow us to define how the 5’ and 3’UTRs of specific viral mRNAs affect their translation efficiency during infection. The transcription start sites of many HCMV genes change over the course of infection. Using this approach we could determine if changes in 5’UTR length and sequence affect the translation of those mRNAs. Additionally, we could identify which mRNAs contain pTRS1-binding sites and determine how their translation efficiency is regulated throughout infection. We could also identify viral mRNAs whose translation is coordinately regulated. Identifying sequence and/or structural similarities in coordinately regulated HCMV 5’ and 3’UTRs could reveal new mechanisms controlling the regulation of viral protein synthesis.
What are the RNAs packaged within the HCMV virion, and are they functional?
Both cellular and viral mRNAs are packaged alongside pTRS1 and pIRS1 in the tegument of the HCMV virion (1, 10). This suggests that pTRS1/pIRS1 recruit mRNAs to the tegument. However the role of pTRS1 and pIRS1 in packaging RNAs in the virion, or their role in the regulation of tegument mRNAs after infection, is unknown.
Conflicting reports suggest that either specific mRNAs are incorporated in the tegument,
186 or that tegument RNAs are packaged according to their abundance within the infected cell (10, 50-52). To date a comprehensive analysis of HCMV virion RNAs has not been performed. This could easily be accomplished using high throughout sequencing of
RNA isolated from purified HCMV virions. Defining the complement of HCMV RNAs packaged within the tegument could identify RNA features that dictate their packaging.
Comparing these data to the complement of RNAs bound by pTRS1 (see above) could identify which RNAs are specifically incorporated into the virion by pTRS1/pIRS1.
If pTRS1 and pIRS1 bind mRNAs in the tegument, they could promote the translation of their associated virion mRNAs immediately after HCMV entry. While it is currently unknown if virion mRNAs are translated, it is tempting to speculate that HCMV incorporates mRNAs encoding viral antagonists of the innate antiviral response. These mRNAs could then be immediately translated into proteins that inactivate cellular defenses. Alternatively, virion mRNAs could encode for host or viral proteins that remodel cell signaling pathways to support virus replication even before the incoming virion has reached the nucleus. The hypothesis that virion mRNAs are translated could be tested by determining the fate of virion RNA in newly infected cells. Virion RNA could be labeled by treating infected cells with 4-thiouridine (4SU) late in infection when virions are being produced. If labeled virion mRNAs associate with polyribosomes early in infection, it would show that mRNAs packaged within the tegument are translated.
We could also determine which, if any, virion mRNAs are translated using RNA-Seq of
4SU-labeled mRNAs associated with polyribosome fractions in newly infected cells treated with ActD. We could also examine the role of pTRS1 and pIRS1 in virion mRNA translation by repeating the experiment above with wild type and ΔTRS1/ΔIRS1 virus
187 grown in PKR KO cells (ΔTRS1/ΔIRS1 must be grown in PKR KO cells to achieve viral replication and virion production). We would then determine if the translation efficiency of virion mRNAs changes in the absence of pTRS1/pIRS1. Characterizing the function of tegument mRNAs immediately upon infection will be crucial for understanding if these very early events are required to prepare the infected cell for efficient virus replication.
It would also be interesting to determine if the complement of RNAs packaged in the virion is consistent across cell types. Do virions produced from infected fibroblasts contain the same RNAs as virions produced from infected epithelial cells? If they differ, is there a change in the fitness of virions produced in one cell type or another? These differences could affect the efficiency of lytic replication in different tissues in vivo, and ultimately affect pathogenesis.
Does mRNA stability regulate mRNA translation during HCMV infection? Our pTRS1 BioID screen revealed that pTRS1-associated proteins are significantly enriched for functions involved in mRNA degradation. For example, both PAN3 and PABP were identified as pTRS1-interactors in our proteomics screen, and we confirmed both interactions during HCMV infection (Figure 4.3). The PAN2/3 exoribonuclease complex is recruited to mRNAs marked for degradation. PAN3 displaces PABP from the poly(A) tail to allow for progressive deadenylation by PAN2. Once the poly(A) tail is <15 nt in length, the CCR4/NOT complex completely removes the poly(A) tail and recruits the
RNA exosome to mediate 3’à5’ degradation of the mRNA (47). We also found that pTRS1 specifically interacted with multiple components of the RNA exosome machinery, suggesting that pTRS1 may affect mRNA degradation. One possibility is
188 that pTRS1 blocks mRNA degradation, which would stabilize mRNAs and thus enhance their translation.
mRNA degradation is partially governed by the length of the mRNA poly(A) tail.
Increased poly(A) length increases the amount of bound PABP, which promotes mRNA circularization and translation, and thus mRNA stability (53, 54). While mRNAs are polyadenylated in the nucleus, cytoplasmic polyadenylation element binding proteins
(CPEBs) bind mRNA 3’UTRs and mediate the recruitment of deadenylases or poly(A) polymerases in the cytoplasm. Four CPEBs (CPEB 1-4) are expressed in mammalian cells, and their activation state determines if they promote deadenylation or polyadenylation of their bound mRNAs (55). A recent study showed that CPEB1 expression is necessary for efficient polyadenylation of HCMV mRNAs, expression of viral proteins, and efficient HCMV replication (56). These data suggest that viral mRNAs with shorter poly(A) tails are translated less efficiently, or are more prone to mRNA degradation, leading to decreased levels of viral proteins. We identified CPEB4 as a pTRS1-interacting protein and confirmed this interaction in transfected cells. Thus I hypothesize that pTRS1 may also play a role in the regulation of viral mRNA poly(A) tail length. We could test this by determining the poly(A) tail length of viral mRNAs using
TAIL-Seq following infection with a wild-type virus or ΔTRS1 virus. If we see no difference in poly(A) tail length we would repeat the experiment using a ΔTRS1/ΔIRS1 virus in PKR KO cells to control for a possibly redundant function of pTRS1 and pIRS1.
We could then determine if there is a change in translation efficiency (using ribosome profiling experiments proposed above) or mRNA stability (see below) for the mRNAs that show a change in poly(A) tail length in the absence of pTRS1 and/or pIRS1
189 expression. Stabilization of poly(A) tail-length by pTRS1, possibly through an interactions with PABP or CPEB4, could promote mRNA stability and translation of pTRS1-target mRNAs.
These data highlight an aspect of post-transcriptional regulation that has not been characterized during HCMV infection. Does the stability of viral mRNAs change during infection? And if so, does that change regulate the translation efficiency of viral mRNAs? Regulation of mRNA stability and/or poly(A) tail length would add yet another level of control that could allow the virus to rapidly change the translation efficiency of an mRNA, and therefore regulate the levels of viral and/or cellular proteins. To identify changes in mRNA stability caused by infection, we could label newly transcribed RNAs with 4SU for a defined period of time during infection, and then add ActD to prevent further transcription. 4SU-labeled RNA could be captured at progressive times after
ActD addition, and the amount of each labeled RNA remaining determined by RNA-Seq.
Performing this analysis at different times after infection would allow us to determine if there are changes in the stability of specific host and viral mRNAs during different stages of infection. By combining this data with previously published ribosome profiling datasets, we could determine if changes in the abundance or translation efficiency of cellular and viral mRNAs during infection results from changes in mRNA stability. We could then repeat this experiment in cells infected with a ΔTRS1 (or ΔTRS1/ΔIRS1, in case of a redundant pIRS1 function) virus to see if pTRS1 and/or pIRS1 expression has an effect on mRNA stability during infection. If we find that pTRS1 affects mRNA translation efficiency by regulating mRNA stability, it would suggest that pTRS1-
190 dependent regulation of mRNA stability is another layer of regulation utilized by the virus to ensure the efficient translation of mRNAs necessary for infection.
What is the role of 3’UTR binding proteins in HCMV mRNA translation regulation? One of the striking, and surprising, results from our pTRS1 BioID screen was the significant association of pTRS1 with 3’UTR binding proteins. Interestingly, a majority of the 3’UTR binding proteins identified as pTRS1 interactors have roles in regulating target mRNA stability, and therefore potentially translation efficiency, as discussed above (47). We have found that pTRS1 lacking its previously identified dsRNA binding domain (pTRS1 ΔRBD) still associates with poly(A) mRNA. Thus pTRS1 may interact with poly(A) mRNA through an interaction with PABP, a 3’UTR binding protein, or through an unindentified additional RNA binding domain. These data suggest that pTRS1 could regulate the stability and translation efficiency of mRNAs through interactions with cellular 3’UTR binding proteins.
The role of 3’UTR binding proteins in HCMV replication has not been thoroughly examined. This is surprising as the number of transcription start sites far outnumbers the number of polyadenylation signals in the HCMV genome (manuscript in preparation), suggesting that multiple HCMV transcripts share a common 3’UTR or polyadenylation site. Perhaps the most extreme example is the HCMV UL74A locus, where 34 splice donor sites scattered across 30 kB of the HCMV genome splice into a single acceptor site within UL74A (unpublished data). Thus, there are at least 34 HCMV transcripts that share a common 3’UTR. A similar splicing pattern is observed for several other HCMV mRNAs, although not to the extent observed in the UL74A locus.
Shared 3’UTRs have the potential to commonly regulate the stability, localization, and
191 translation of many HCMV mRNAs. It is therefore likely that 3’UTR binding proteins play a major role in the regulation of HCMV mRNA stability and translation. pTRS1 binding to a specific 3’UTR binding protein could allow pTRS1 to regulate the translation or stability of groups of HCMV mRNAs through a single interaction. pTRS1 binding to
3’UTR binding proteins could allow pTRS1 to have a broader effect on translation than if its interaction with mRNA was solely mediated through its ability to bind dsRNA. Thus it is likely that the translation of many mRNAs, in addition to those directly bound by pTRS1, is regulated by pTRS1 expression during HCMV infection.
Do mRNA modifications affect translation regulation during HCMV infection? We also found that pTRS1 interacted with the cellular proteins YTHDC2, YTHDF2, and
YTHDF3, which bind RNA marked with an N6-methyladenosine (m6A) modification. m6A is the most abundant RNA modification, and was recently shown to play a critical role in stress-resistant translation (57, 58). As pTRS1 is involved in translation regulation following cellular stress and interacts with multiple m6A readers, I hypothesize that m6A modifications play a role in the ability of pTRS1 to stimulate translation. Determining the effect of depleting m6A readers or m6A methyltransferases on the ability of pTRS1 to enhance reporter mRNA translation would allow us to determine if pTRS1 specifically stimulates the translation of m6A modified mRNAs. While the exact function of YTHDC2 and YTHDF3 is unknown, YTDHF2 has been shown to promote mRNA degradation of its bound mRNAs (59). Perhaps pTRS1 binds YTHDF2 and prevents degradation of m6A modified mRNAs. This could be an additional means by which pTRS1 modulates mRNA stability and therefore mRNA translation.
192 On a larger scale, the role of mRNA modifications in translation regulation during
HCMV infection has not been examined. RNA methylation affects both mRNA decay and translation initiation (57), however it is not known if HCMV mRNAs contain RNA modifications. HCMV mRNAs are processed similarly to cellular mRNAs, which often contain multiple RNA modifications. Thus it seems likely that HCMV RNAs contain similar RNA modifications. Recently, cap-independent translation was shown to require m6A modification of 5’UTRs (60, 61). Perhaps m6A modification of HCMV mRNAs enhances their translation in conditions that limit eIF4F abundance. We could test this by depleting m6A methyltransferases prior to infection and determining if HCMV mRNAs maintain the ability to translate in the absence of a complete eIF4F complex. If m6A modifications were necessary for HCMV mRNAs to translate in an eIF4F-independent manner, it would be interesting to see if the interaction of pTRS1 with m6A readers would be necessary for this phenotype. It is possible that m6A modifications on viral mRNAs are bound by the same m6A readers that pTRS1 binds, and thus could target pTRS1 to viral mRNAs to promote their translation. In any case, the continued translation of HCMV mRNAs in the face of cellular stress is a hallmark of HCMV translation regulation. Inclusion of mRNA modifications that promote stress-resistant translation on viral mRNAs may be another strategy utilized by HCMV to ensure efficient synthesis of viral proteins.
How are mRNAs translated in the presence of eIF2α phosphorylation during
HCMV infection? Many viruses inhibit the translation of cellular mRNAs to free ribosomes to translate viral mRNAs (62, 63). However both host and viral mRNAs are translated throughout the HCMV replicative cycle, even though phosphorylated eIF2α
193 accumulates later in infection (48, 64). Phosphorylation of eIF2α increases its affinity for the guanine nucleotide exchange factor (GEF) eIF2B. Tight binding of phosphorylated eIF2α inhibits the ability of eIF2B to exchange GDP for GTP on eIF2, preventing the formation of new ternary complexes and leading to a global inhibition of protein synthesis (65, 66). Thus an outstanding question is how are host and viral mRNAs efficiently translated in the presence of phosphorylated eIF2α? Even a small amount of eIF2α phosphorylation significantly inhibits protein synthesis, as eIF2α is present at much greater levels than eIF2B (eIF2α is ~10X more abundant) (67). Therefore even low levels of phosphorylated eIF2α can sequester all available eIF2B and prevent ternary complex formation. A potential mechanism to circumvent eIF2α phosphorylation would be for HCMV to increase expression levels of eIF2B. However, this does not seem to be the case, as eIF2α and eIF2B levels increase proportionally during infection
(68). An alternative would be for HCMV to encode a viral GEF, however a GEF domain has not been identified in an HCMV protein.
Another possible mechanism to maintain protein synthesis despite eIF2α phosphorylation would be for HCMV mRNAs to use an alternative translation initiation
Met factor(s). The eIF2A and eIF2D initiation factors bind tRNAi , associate with 40S ribosomal subunits, and can substitute for eIF2 in ternary complex formation (69-72). eIF2A is important for viral protein synthesis during hepatitis C virus and alphavirus infection, which induce eIF2α phosphorylation (73, 74). The role of eIF2A and eIF2D during HCMV infection has not been determined. Cells lacking expression of eIF2A, eIF2D, or both, are viable (75). Thus overall levels of protein synthesis and HCMV
194 replication could be examined in cells lacking eIF2A and eIF2D expression to determine if these proteins contribute to mRNA translation at late times in HCMV replication.
Another possibility is that HCMV alters the composition of eIF2B to make it insensitive to eIF2α phosphorylation. eIF2B is a decameric complex consisting of two copies of a five protein subunit, eIF2Bα-ε. One subunit, the eIF2Bα protein, is not necessary for eIF2B GEF activity. In fact, loss of eIF2Bα renders eIF2B activity resistant to inactivation when eIF2α is phosphorylated (76). HCMV could alter eIF2Bα expression or binding to the eIF2B complex to allow for continued mRNA translation despite eIF2α phosphorylation. Whole cell proteomics analysis of HCMV infected cells found that eIF2Bα levels are relatively unchanged by infection (68), however it is not known if infection prevents inclusion of eIF2Bα into the eIF2B complex. If eIF2Bα was dissociated from the eIF2B complex, perhaps by an HCMV protein, this could facilitate
GTP exchange on eIF2 despite eIF2α phosphorylation, allowing for overall levels of protein synthesis to remain unaffected. Proteomics analysis to define the components of the eIF2B complex during infection could identify changes to the eIF2B holoenzyme complex that potentially allow HCMV mRNAs to translate in the face of eIF2α phosphorylation.
It is also possible that HCMV specifically induces eIF2α phosphorylation late in infection for the benefit of the virus. As discussed above, eIF2α phosphorylation inhibits the translation of most mRNAs. However eIF2α phosphorylation also promotes translation initiation on mRNAs that contain short upstream ORFs (uORFs) in their
5’UTR preceding the mRNA coding determinant sequence (cds). Under non-stressed conditions, translation initiation on 5’UTR uORFs inhibits the translation of the
195 downstream cds. When eIF2α is phosphorylated, 48S complexes are more likely to scan through uORFs to the translation start codon for the cds (39, 46). uORFs are common in cellular mRNAs encoding proteins involved in alleviating cellular stress (41).
Virus replication and assembly induces significant cellular stress (77), thus the expression of cellular proteins needed to mitigate stress would likely be beneficial to the virus at this point of infection. Further HCMV ribosome profiling data identified several hundred short (<30 aa) peptide coding regions, which were often situated just upstream of known viral genes (48). While the full-length transcriptome of HCMV is poorly defined, this data suggests that uORFs may be a common feature of HCMV mRNAs.
Ribosome profiling data also showed that HCMV mRNAs are translated more efficiently than host mRNAs late in infection, when eIF2α is phosphorylated (64). I hypothesize that the inclusion of uORFs on HCMV mRNAs allows them to be translated more efficiently than host mRNAs when eIF2α is phosphorylated. This would allow for ongoing viral protein synthesis despite the harsh intracellular environment found late in infection, and could help HCMV mRNAs compete for host ribosomes by decreasing the translation of cellular mRNAs. To test this hypothesis, we could measure HCMV mRNA translation and replication in cells where the phosphorylation site of eIF2α (Ser51) has been mutated using CRISPR/Cas9. We could then compare the translation efficiency of host and viral mRNAs in control and eIF2α phosphosite mutant cells at different times after infection, and measure the impact on HCMV replication. No matter the mechanism, further work is needed to understand how HCMV mRNAs avoid translation inhibition in the presence of eIF2α phosphorylation, as this mechanism is likely
196 necessary for the efficient translation of viral mRNAs late in infection, and therefore necessary for efficient HCMV replication.
Conclusion
In this thesis I defined the molecular mechanism of PKR antagonism by pTRS1, showed that pTRS1 stimulates cap-independent translation, and identified pTRS1- interacting proteins that may contribute to the ability of pTRS1 to stimulate translation in a PKR-independent manner. These data add to a growing list of pTRS1 functions and also implicate a role for pTRS1 in the regulation of mRNA stability and mRNA modifications as a means to control protein synthesis during HCMV infection. Post- transcriptional regulation of gene expression is a largely unexplored area of HCMV research. The role of mRNA modifications, mRNA stability, and cellular RNA binding proteins in translation regulation during HCMV, and most herpesvirus, infections has not been thoroughly characterized. As viruses are completely reliant on the host translation machinery for the synthesis of viral proteins, factors that are necessary for the synthesis of viral proteins are attractive antiviral drug targets. Thus studies identifying the cellular factors necessary for the translation of viral mRNAs are important for both our complete understanding of the HCMV replication cycle, and the development of novel antiviral therapeutics.
197 REFERENCES
1. Romanowski MJ, Garrido-Guerrero E, Shenk T. 1997. pIRS1 and pTRS1 are present in human cytomegalovirus virions. Journal of Virology 71:5703-5705.
2. Hakki M, Marshall EE, De Niro KL, Geballe AP. 2006. Binding and nuclear relocalization of protein kinase R by human cytomegalovirus TRS1. Journal of Virology 80:11817-11826.
3. Marshall EE, Bierle CJ, Brune W, Geballe AP. 2009. Essential role for either TRS1 or IRS1 in human cytomegalovirus replication. Journal of Virology 83:4112-4120.
4. Ziehr B, Vincent HA, Moorman NJ. 2016. Human Cytomegalovirus pTRS1 and pIRS1 Antagonize Protein Kinase R To Facilitate Virus Replication. Journal of Virology 90:3839-3848.
5. Braggin JE, Child SJ, Geballe AP. 2016. Essential role of protein kinase R antagonism by TRS1 in human cytomegalovirus replication. Virology 489:75-85.
6. Vincent HA, Ziehr B, Moorman NJ. 2017. Mechanism of Protein Kinase R Inhibition by Human Cytomegalovirus pTRS1. Journal of Virology 91.
7. Ziehr B, Lenarcic E, Vincent HA, Cecil C, Garcia B, Shenk T, Moorman NJ. 2015. Human cytomegalovirus TRS1 protein associates with the 7- methylguanosine mRNA cap and facilitates translation. Proteomics 15:1983- 1994.
8. Roux KJ, Kim DI, Raida M, Burke B. 2012. A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. The Journal of Cell Biology 196:801-810.
9. Lemaire PA, Anderson E, Lary J, Cole JL. 2008. Mechanism of PKR Activation by dsRNA. Journal of molecular biology 381:351-360.
10. Terhune SS, Schröer J, Shenk T. 2004. RNAs are packaged into human cytomegalovirus virions in proportion to their intracellular concentration. Journal of Virology 78:10390-10398.
198 11. Dey M, Cao C, Dar AC, Tamura T, Ozato K, Sicheri F, Dever TE. 2005. Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition. Cell 122:901-913.
12. Seo EJ, Liu F, Kawagishi-Kobayashi M, Ung TL, Cao C, Dar AC, Sicheri F, Dever TE. 2008. Protein kinase PKR mutants resistant to the poxvirus pseudosubstrate K3L protein. Proceedings of the National Academy of Sciences of the United States of America 105:16894-16899.
13. Yu Y, Pierciey FJ, Maguire TG, Alwine JC. 2013. PKR-like endoplasmic reticulum kinase is necessary for lipogenic activation during HCMV infection. PLoS pathogens 9:e1003266.
14. Hakki M, Geballe AP. 2008. Cellular serine/threonine phosphatase activity during human cytomegalovirus infection. Virology 380:255-263.
15. Isler JA, Skalet AH, Alwine JC. 2005. Human cytomegalovirus infection activates and regulates the unfolded protein response. Journal of Virology 79:6890-6899.
16. Lageix S, Rothenburg S, Dever TE, Hinnebusch AG. 2014. Enhanced interaction between pseudokinase and kinase domains in Gcn2 stimulates eIF2α phosphorylation in starved cells. PLoS genetics 10:e1004326.
17. Romanowski MJ, Shenk T. 1997. Characterization of the human cytomegalovirus irs1 and trs1 genes: a second immediate-early transcription unit within irs1 whose product antagonizes transcriptional activation. Journal of Virology 71:1485-1496.
18. Blankenship CA, Shenk T. 2002. Mutant human cytomegalovirus lacking the immediate-early TRS1 coding region exhibits a late defect. Journal of Virology 76:12290-12299.
19. DeFilippis VR, Alvarado D, Sali T, Rothenburg S, Früh K. 2010. Human Cytomegalovirus Induces the Interferon Response via the DNA Sensor ZBP1. Journal of Virology 84:585-598.
20. Zhu H, Cong JP, Shenk T. 1997. Use of differential display analysis to assess the effect of human cytomegalovirus infection on the accumulation of cellular
199 RNAs: induction of interferon-responsive RNAs. Proceedings of the National Academy of Sciences of the United States of America 94:13985-13990.
21. Amsler L, Verweij M, DeFilippis VR. 2013. The Tiers and Dimensions of Evasion of the Type I Interferon Response by Human Cytomegalovirus. Journal of molecular biology 425.
22. Bierle CJ, Semmens KM, Geballe AP. 2013. Double-stranded RNA binding by the human cytomegalovirus PKR antagonist TRS1. Virology 442:28-37.
23. Hakki M, Geballe AP. 2005. Double-stranded RNA binding by human cytomegalovirus pTRS1. Journal of Virology 79:7311-7318.
24. Choy MS, Yusoff P, Lee IC, Newton JC, Goh CW, Page R, Shenolikar S, Peti W. 2015. Structural and functional analysis of the GADD34:PP1 eIF2α phosphatase. Cell reports 11:1885-1891.
25. Peti W, Nairn AC, Page R. 2013. Structural basis for protein phosphatase 1 regulation and specificity. The FEBS journal 280:596-611.
26. Roy J, Cyert MS. 2009. Cracking the Phosphatase Code: Docking Interactions Determine Substrate Specificity. Sci Signal 2:re9-re9.
27. Kazemi S, Papadopoulou S, Li S, Su Q, Wang S, Yoshimura A, Matlashewski G, Dever TE, Koromilas AE. 2004. Control of alpha subunit of eukaryotic translation initiation factor 2 (eIF2 alpha) phosphorylation by the human papillomavirus type 18 E6 oncoprotein: implications for eIF2 alpha- dependent gene expression and cell death. Molecular and Cellular Biology 24:3415-3429.
28. Cassady KA, Gross M, Roizman B. 1998. The Herpes Simplex Virus US11 Protein Effectively Compensates for the γ134.5 Gene if Present before Activation of Protein Kinase R by Precluding Its Phosphorylation and That of the α Subunit of Eukaryotic Translation Initiation Factor 2. Journal of Virology 72:8620-8626.
29. Rojas M, Vasconcelos G, Dever TE. 2015. An eIF2α-binding motif in protein phosphatase 1 subunit GADD34 and its viral orthologs is required to promote dephosphorylation of eIF2α. Proceedings of the National Academy of Sciences of the United States of America 112:E3466-E3475.
200 30. Hinnebusch AG. 2014. The scanning mechanism of eukaryotic translation initiation. Annual Review of Biochemistry 83:779-812.
31. Lacerda R, Menezes J, Romão L. 2017. More than just scanning: the importance of cap-independent mRNA translation initiation for cellular stress response and cancer. Cellular and Molecular Life Sciences 74:1659-1680.
32. Collins-McMillen D, Kim JH, Nogalski MT, Stevenson EV, Chan GC, Caskey JR, Cieply SJ, Yurochko AD. 2015. Human Cytomegalovirus Promotes Survival of Infected Monocytes via a Distinct Temporal Regulation of Cellular Bcl-2 Family Proteins. Journal of Virology 90:2356-2371.
33. de Wit RH, Mujić-Delić A, van Senten JR, Fraile-Ramos A, Siderius M, Smit MJ. 2016. Human cytomegalovirus encoded chemokine receptor US28 activates the HIF-1α/PKM2 axis in glioblastoma cells. Oncotarget 7:67966-67985.
34. Buchkovich NJ, Maguire TG, Alwine JC. 2010. Role of the endoplasmic reticulum chaperone BiP, SUN domain proteins, and dynein in altering nuclear morphology during human cytomegalovirus infection. Journal of Virology 84:7005-7017.
35. Buchkovich NJ, Maguire TG, Paton AW, Paton JC, Alwine JC. 2009. The endoplasmic reticulum chaperone BiP/GRP78 is important in the structure and function of the human cytomegalovirus assembly compartment. Journal of Virology 83:11421-11428.
36. Buchkovich NJ, Maguire TG, Yu Y, Paton AW, Paton JC, Alwine JC. 2008. Human cytomegalovirus specifically controls the levels of the endoplasmic reticulum chaperone BiP/GRP78, which is required for virion assembly. Journal of Virology 82:31-39.
37. Grainger L, Cicchini L, Rak M, Petrucelli A, Fitzgerald KD, Semler BL, Goodrum F. 2010. Stress-inducible alternative translation initiation of human cytomegalovirus latency protein pUL138. Journal of Virology 84:9472-9486.
38. Weingarten-Gabbay S, Elias-Kirma S, Nir R, Gritsenko AA, Stern-Ginossar N, Yakhini Z, Weinberger A, Segal E. 2016. Systematic discovery of cap- independent translation sequences in human and viral genomes. Science 351:aad4939.
201 39. Sonenberg N, Hinnebusch AG. 2009. Regulation of translation initiation in eukaryotes: mechanisms and biological targets. Cell 136:731-745.
40. Hara K, Yonezawa K, Kozlowski MT, Sugimoto T, Andrabi K, Weng Q-P, Kasuga M, Nishimoto I, Avruch J. 1997. Regulation of eIF-4E BP1 Phosphorylation by mTOR. Journal of Biological Chemistry 272:26457-26463.
41. Liu B, Qian S-B. 2014. Translational reprogramming in stress response. Wiley interdisciplinary reviews RNA 5:301-305.
42. Walsh D, Perez C, Notary J, Mohr I. 2005. Regulation of the translation initiation factor eIF4F by multiple mechanisms in human cytomegalovirus- infected cells. Journal of Virology 79:8057-8064.
43. McKinney C, Perez C, Mohr I. 2012. Poly(A) binding protein abundance regulates eukaryotic translation initiation factor 4F assembly in human cytomegalovirus-infected cells. Proceedings of the National Academy of Sciences 109:5627-5632.
44. Lenarcic EM, Ziehr B, De Leon G, Mitchell D, Moorman NJ. 2014. Differential role for host translation factors in host and viral protein synthesis during human cytomegalovirus infection. Journal of Virology 88:1473-1483.
45. Gatherer D, Seirafian S, Cunningham C, Holton M, Dargan DJ, Baluchova K, Hector RD, Galbraith J, Herzyk P, Wilkinson GWG, Davison AJ. 2011. High- resolution human cytomegalovirus transcriptome. Proceedings of the National Academy of Sciences 108:19755-19760.
46. Hinnebusch AG, Ivanov IP, Sonenberg N. 2016. Translational control by 5 - untranslated regions of eukaryotic mRNAs. Science 352:1413-1416.
47. Norbury CJ. 2013. Cytoplasmic RNA: a case of the tail wagging the dog. Nature Reviews Molecular Cell Biology 14:643-653.
48. Stern-Ginossar N, Weisburd B, Michalski A, Le VTK, Hein MY, Huang S-X, Ma M, Shen B, Qian S-B, Hengel H, Mann M, Ingolia NT, Weissman JS. 2012. Decoding human cytomegalovirus. Science (New York, NY) 338:1088-1093.
202 49. Kulesza CA, Shenk T. 2004. Human cytomegalovirus 5-kilobase immediate- early RNA is a stable intron. Journal of Virology 78:13182-13189.
50. Greijer AE, Dekkers CAJ, Middeldorp JM. 2000. Human Cytomegalovirus Virions Differentially Incorporate Viral and Host Cell RNA during the Assembly Process. Journal of Virology 74:9078-9082.
51. Mohammad A-A, Costa H, Landázuri N, Lui W-O, Hultenby K, Rahbar A, Yaiw K-C, Söderberg-Nauclér C. 2017. Human cytomegalovirus microRNAs are carried by virions and dense bodies and are delivered to target cells. The Journal of General Virology 98:1058-1072.
52. Sarcinella E, Brown M, Tellier R, Petric M, Mazzulli T. 2004. Detection of RNA in purified cytomegalovirus virions. Virus Research 104:129-137.
53. Eckmann CR, Rammelt C, Wahle E. 2011. Control of poly(A) tail length. Wiley Interdisciplinary Reviews: RNA 2:348-361.
54. Brook M, McCracken L, Reddington JP, Lu Z-L, Morrice NA, Gray NK. 2012. The multifunctional poly(A)-binding protein (PABP) 1 is subject to extensive dynamic post-translational modification, which molecular modelling suggests plays an important role in co-ordinating its activities. The Biochemical Journal 441:803-812.
55. Richter JD. 2007. CPEB: a life in translation. Trends in Biochemical Sciences 32:279-285.
56. Batra R, Stark TJ, Clark E, Belzile J-P, Wheeler EC, Yee BA, Huang H, Gelboin-Burkhart C, Huelga SC, Aigner S, Roberts BT, Bos TJ, Sathe S, Donohue JP, Rigo F, Ares M, Spector DH, Yeo GW. 2016. RNA binding protein CPEB1 remodels host and viral RNA landscapes. Nature structural & molecular biology 23:1101-1110.
57. Zhao BS, Roundtree IA, He C. 2016. Post-transcriptional gene regulation by mRNA modifications. Nature Reviews Molecular Cell Biology 18:nrm.2016.2132.
58. Zhou J, Wan J, Gao X, Zhang X, Qian S-B. 2015. Dynamic m6A mRNA methylation directs translational control of heat shock response. Nature 526:591- 594.
203 59. Wang X, Lu Z, Gomez A, Hon GC, Yue Y, Han D, Fu Y, Parisien M, Dai Q, Jia G, Ren B, Pan T, He C. 2014. N6-methyladenosine-dependent regulation of messenger RNA stability. Nature 505:117-120.
60. Coots RA, Liu X-M, Mao Y, Dong L, Zhou J, Wan J, Zhang X, Qian S-B. 2017. m6A Facilitates eIF4F-Independent mRNA Translation. Molecular Cell 0.
61. Meyer KD, Patil DP, Zhou J, Zinoviev A, Skabkin MA, Elemento O, Pestova TV, Qian S-B, Jaffrey SR. 2015. 5 UTR m6A Promotes Cap-Independent Translation. Cell 163:999-1010.
62. Jan E, Mohr I, Walsh D. 2016. A Cap-to-Tail Guide to mRNA Translation Strategies in Virus-Infected Cells. Annual Review of Virology 3:283-307.
63. Walsh D, Mathews MB, Mohr I. 2013. Tinkering with Translation: Protein Synthesis in Virus-Infected Cells. Cold Spring Harbor Perspectives in Biology 5:a012351.
64. Tirosh O, Cohen Y, Shitrit A, Shani O, Le-Trilling VTK, Trilling M, Friedlander G, Tanenbaum M, Stern-Ginossar N. 2015. The Transcription and Translation Landscapes during Human Cytomegalovirus Infection Reveal Novel Host-Pathogen Interactions. PLOS Pathogens 11:e1005288.
65. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. 2001. Tight Binding of the Phosphorylated α Subunit of Initiation Factor 2 (eIF2α) to the Regulatory Subunits of Guanine Nucleotide Exchange Factor eIF2B Is Required for Inhibition of Translation Initiation. Molecular and Cellular Biology 21:5018- 5030.
66. Pavitt GD, Ramaiah KV, Kimball SR, Hinnebusch AG. 1998. eIF2 independently binds two distinct eIF2B subcomplexes that catalyze and regulate guanine-nucleotide exchange. Genes & Development 12:514-526.
67. Singh CR, Udagawa T, Lee B, Wassink S, He H, Yamamoto Y, Anderson JT, Pavitt GD, Asano K. 2007. Change in Nutritional Status Modulates the Abundance of Critical Pre-initiation Intermediate Complexes During Translation Initiation in Vivo. Journal of Molecular Biology 370:315-330.
68. Weekes MP, Tomasec P, Huttlin EL, Fielding CA, Nusinow D, Stanton RJ, Wang ECY, Aicheler R, Murrell I, Wilkinson GWG, Lehner PJ, Gygi SP. 2014.
204 Quantitative Temporal Viromics: An Approach to Investigate Host-Pathogen Interaction. Cell 157:1460-1472.
69. Starck SR, Tsai JC, Chen K, Shodiya M, Wang L, Yahiro K, Martins-Green M, Shastri N, Walter P. 2016. Translation from the 5' untranslated region shapes the integrated stress response. Science (New York, NY) 351:aad3867.
70. Skabkin MA, Skabkina OV, Dhote V, Komar AA, Hellen CUT, Pestova TV. 2010. Activities of Ligatin and MCT-1/DENR in eukaryotic translation initiation and ribosomal recycling. Genes & Development 24:1787-1801.
71. Komar AA, Gross SR, Barth-Baus D, Strachan R, Hensold JO, Goss Kinzy T, Merrick WC. 2005. Novel characteristics of the biological properties of the yeast Saccharomyces cerevisiae eukaryotic initiation factor 2A. The Journal of Biological Chemistry 280:15601-15611.
72. Merrick WC, Anderson WF. 1975. Purification and characterization of homogeneous protein synthesis initiation factor M1 from rabbit reticulocytes. The Journal of Biological Chemistry 250:1197-1206.
73. Kim JH, Park SM, Park JH, Keum SJ, Jang SK. 2011. eIF2A mediates translation of hepatitis C viral mRNA under stress conditions. The EMBO journal 30:2454-2464.
74. Ventoso I, Sanz MA, Molina S, Berlanga JJ, Carrasco L, Esteban M. 2006. Translational resistance of late alphavirus mRNA to eIF2α phosphorylation: a strategy to overcome the antiviral effect of protein kinase PKR. Genes & Development 20:87-100.
75. Sanz MA, González Almela E, Carrasco L. 2017. Translation of Sindbis Subgenomic mRNA is Independent of eIF2, eIF2A and eIF2D. Scientific Reports 7.
76. Gordiyenko Y, Schmidt C, Jennings MD, Matak-Vinkovic D, Pavitt GD, Robinson CV. 2014. eIF2B is a decameric guanine nucleotide exchange factor with a γ2ε2 tetrameric core. Nature Communications 5:ncomms4902.
77. Alwine JC. 2008. Modulation of host cell stress responses by human cytomegalovirus. Current Topics in Microbiology and Immunology 325:263-279.
205