Characterization of the Transcripts that Encode pUL138, a Latency Determinant, During Human Cytomegalovirus Infection

Item Type text; Electronic Dissertation

Authors Grainger, Lora Ann

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 24/09/2021 15:04:11

Link to Item http://hdl.handle.net/10150/195915 1

CHARACTERIZATION OF THE TRANSCRIPTS THAT ENCODE PUL138, A LATENCY DETERMINANT, DURING HUMAN CYTOMEGALOVIRUS INFECTION

by

Lora A. Grainger

Copyright © Lora A. Grainger 2010

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF IMMUNOBIOLOGY

In Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2010 2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Lora A. Grainger entitled: Characterization of the Transcripts that Encode pUL138, a Latency Determinant, During Human Cytomegalovirus Infection. We recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy.

______Date: 6/15/10 Dr. Nafees Ahmad

______Date: 6/15/10 Dr. Lonnie Lybarger

______Date: 6/15/10 Dr. Carol Dieckmann

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement

______Date: 6/15/10 Dissertation Director: Dr. Felicia Goodrum 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.

SIGNED: Lora A. Grainger 4

ACKNOWLEDGEMENTS

There are many people I would like to thank that have contributed to my completion of the Ph.D. but I owe Dr. Felicia Goodrum special thanks as she ‘rescued’ me from a string of bad luck that would have left me wavering if she had not invited me to join her laboratory. With that, I owe a tremendous amount of gratitude to her as well as Alex Petrucelli and Mike Rak who have been instrumental as team players in the development of the internationally recognized Goodrum laboratory. The four of us truly became a family while the lab was being established. I’d also like to thank the current and past members of the Goodrum laboratory. Thank you for all of your support!

Much of this work was done in collaboration with Louis Cicchini who has contributed greatly to the identification of the HCMV IRES. He has become an extraordinary scientist and I am eternally grateful for his insight, enthusiasm and organizational skills through the duration of our time together in the laboratory.

I would also like to thank the faculty and graduate students of the Department of Immunobiology for all of their support through this process, in particular, the late Dr. Deluca and Dr. Friedman who allowed me to work in their laboratories. Further, I am sincerely grateful for the friendship, support and knowledge extended by the Immunobiology graduate students during times of extreme happiness as well as when I needed them the most. Thank you!

Throughout my time as a graduate student I have enjoyed sharing my love of science with others. Therefore, I would like to extend thanks to Nadja Anderson and Stacey Forsyth who have been instrumental in allowing me to extend my research to others through various biotechnology outreach initiatives. Thanks!

Last but definitely not least, I would like to thank the members of my family and close friends for their support through the pursuit of my Ph.D. I’d especially like to thank my mom and dad who have provided unconditional love, support and independence that has made me who I am today! Thank you! 5

DEDICATION

To my loving parents, John and Marylin Grainger. 6

TABLE OF CONTENTS

LIST OF TABLES…………………………………………………………………...…10

LIST OF FIGURES…………………………………………………………………….11

ABSTRACT…………………………………………………………………………….14

CHAPTER 1 Literature Review………………………………………………………15

1.1 Human Cytomegalovirus ………………………………………………...15

1.2 HCMV Latency…………………………………………………………….21

1.3 Mechanisms of Translation Initiation……………………………………31

1.4 IRESs……………………………………………………………………….36

1.5 The Type I Interferon Response…………………………………………40

1.6 Type I IFN Response and HCMV Latency…………………………...... 44

CHAPTER 2 Materials and Methods………………………………………………...46

2.1 Cells………………………………………………………………………...46

2.2 Viruses……………………………………………………………………...47

2.3 Plasmids……………………………………………………………………48

2.4 Transient Transfection and Luciferase Reporter Assays………….….50

2.5 Immunoblotting………………………………………………………….…51

2.6 Northern Blotting…………………………………………………………..52

2.7 RACE……………………………………………………………………….53

2.8 In Vitro Transcription and Translation…………………………………..55

2.9 RNA Structure Prediction………………………………………………...55 7

TABLE OF CONTENTS-Continued

2.10 qRT-PCR Transcript Length Analysis…………………………………56

2.11 qRT-PCR IE, UL138 Expression and IFN Analysis………………….58

2.12 Interferon Manipulation………………………………………………….60

CHAPTER 3 UL138 transcript expression during productive and latent Human

cytomegalovirus infection……………………………………………..63

3.1 Introduction………………………………………………………………...63

3.2 UL138 is encoded on multiple transcripts in fibroblasts and

hematopoietic progenitor cells……………………………………….65

3.3 UL138 transcripts are specific and are present in

FIX∆UL138STOP infected fibroblasts…………………………………68

3.4 The UL138 transcripts in FIX∆UL138STOP infected

cells are stable…………………………………………………………70

3.5 Kinetics of UL138 expression in fibroblasts and CD34+

HPCs……………………………………………………………………71

3.6 Discussion and Future Directions……………………………………….74

CHAPTER 4 Protein coding capacity of UL138 transcripts……………………….76

4.1 Introduction………………………………………………………………...76

4.2 Putative ORFs upstream of UL138 encode proteins

during HCMV infection……………………………………………...78

4.3. pUL138 is expressed from polycistronic mRNAs……………...……...81

4.4 Model of the coding potential of the polycistronic UL138 transcripts..84 8

TABLE OF CONTENTS-Continued

4.5 Discussion and Future Directions……………………………………….85

CHAPTER 5 pUL138 is expressed by alternative mechanisms of

translation initiation………………………………………………….…88

5.1 Introduction…………………………………………………………….…..88

5.2 A sequence overlapping the 5′ end of the UL136 ORF

promotes translation of a downstream cistron………………….…...91

5.3 Transcripts originating from the dual luciferase vector are full

length and ICS-7 do not have inherent promoter activity………..…94

5.4 ICS-7 Mfold prediction suggests extensive RNA

secondary structure…………………………………………………….98

5.5 ICS-7 IRES activity is dependent on eIF4G…………………………..100

5.6 Discussion and Future Directions……………………………………...101

CHAPTER 6 pUL138 expression from the 3.6- and 2.7-kb

transcripts is stress inducible……………………………………….104

6.1 Introduction……………………………………………………………….104

6.2 ICS-7 is a stress-inducible IRES……………………………………….106

6.3 UL138 cistron expression from the 3.6- and 2.7-kb transcripts,

but not the 1.4-kb transcript, is stress-inducible…………………..107

6.4 IRES mediated UL138 cistron expression from

the 2.7-kb transcript is dependent on eIF4G………………….…..111

9

TABLE OF CONTENTS-Continued

6.5 UL138 protein expression is stress-inducible from the 3.6- and

2.7-kb UL138 transcripts, but not the 1.4-kb transcript…………..113

6.6 Discussion and Future Directions……………………………………...116

CHAPTER 7 The role of type I IFN in human cytomegalovirus latency………..119

7.1 Introduction……………………………………………………………….119

7.2 The ULb′ ORFs UL136-UL142 are associated with decreased

resistance to the type I IFN response in MRC5 fibroblasts……...121

7.3 The ULb′ ORFs UL136-UL142 are associated with

decreased resistance to IFNα challenge…………………………..124

7.4 The ULb′ ORFs UL136-UL142 are associated with increased

resistance to the type I IFN response in CD34+ HPCs………….126

7.5 ORFs UL136-UL142 are necessary for FIX to inhibit ISG

expression in CD34+ HPCs…………………………………………129

7.6 Discussion and Future Directions……………………………………...132

CHAPTER 8 Discussion……………………………………………………………..136

REFERENCES……………………………………………………………………….154 10

LIST OF TABLES

Table 1 Primers used for cloning expression vectors……………………………...60

Table 2 Primers used for recombinant virus generation…………………………..62

11

LIST OF FIGURES

Figure 1. Human cytomegalovirus virion organization…………………………….17

Figure 2. Genome organization of laboratory adapted and

clinical strains of HCMV……………………………………………………….20

Figure 3. Classification of human herpesviruses…………………………………..22

Figure 4. in vitro HCMV latency model system………………………..…………..24

Figure 5. pUL138 promotes a latent infection in CD34+

hematopoietic progenitor cells……………………………………………….26

Figure 6. Identification of UL138 transcripts………………………………………..30

Figure 7. Eukaryotic mechanisms of translation initiation…………………………34

Figure 8. The type I IFN response in HCMV infected cells……………………….43

Figure 9. UL138 is encoded on 3.6-, 2.7- and 1.4-kb transcripts

in MRC5 fibroblasts……………………………………………………………67

Figure 10. UL138 transcripts are specific and are present

in FIX∆UL138STOP infected MRC5 fibroblasts………………………………69

Figure 11. The UL138 transcripts in FIX∆UL138STOP infected

cells are stable…………………………………………………………………70

Figure 12. Kinetics of UL138 expression……………………………………………73

Figure 13. UL133, UL135 and UL136 encode proteins……………………………80

Figure 14. UL138 transcripts are polycistronic……………………………………..83

Figure 15. Model of the coding potential of the polycistronic UL138 transcripts..85 12

LIST OF FIGURES-Continued

Figure 16. A 663-nt sequence overlapping UL136 supports

expression of a downstream cistron…………………………………………93

Figure 17. Downstream cistron expression is not due to the

presence of monocistronic transcripts specific to ICS-7…………………..97

Figure 18. Predicted RNA secondary structure of HCMV the IRES……………..99

Figure 19. ICS-7 IRES activity is dependent on eIF4G…………………………..101

Figure 20. ICS-7 is stress inducible………………………………………………..106

Figure 21. IRES mediated expression from the UL138 cistron

is induced during stress…………………………………………………….110

Figure 22. IRES-mediated expression of the UL138 cistron

is dependent on eIF4G……………………………………………………..113

Figure 23. IRES mediated expression of pUL138 from the 3.6-

and 2.7-kb transcripts is induced during stress………………………….115

Figure 24. ULb′ ORFs UL136-UL142 are associated with decreased

resistance to the type I IFN response in MRC5 fibroblasts…………….123

Figure 25. ULb′ ORFs UL136-UL142 are associated with decreased

resistance to IFN challenge………………………………………………..125

Figure 26. ULb′ ORFs UL136-UL142 are associated with increased

resistance to the type I IFN response in CD34+ HPCs…………………128

13

LIST OF FIGURES-Continued

Figure 27. ORFs UL136-UL142 are necessary for FIX to inhibit

ISG expression in CD34+ HPCs……………………………………………131 14

ABSTRACT

Mechanisms involved in the establishment of HCMV latency are poorly understood, however, work in our laboratory has demonstrated the ULb′ encoded protein, pUL138, as the first viral determinant to function in the establishment of

HCMV latency in CD34+ hematopoietic progenitor cells (HPCs). This work characterizes the transcripts that encode pUL138, identifies three novel ULb′ proteins (pUL133, pUL135, and pUL136) and represents the first demonstration of an internal ribosome entry site (IRES) mediated expression of pUL138. pUL138 is encoded on three polycistronic transcripts of 3.6-, 2.7- and 1.4-kb in length. pUL133, pUL135 and truncated pUL136, are expressed on the 3.6-, 2.7- and 1.4-kb transcripts, respectively, in addition to pUL138. We demonstrate that pUL138 expression is inducible from the IRES on the 3.6- and 2.7-kb transcripts under conditions of cellular stress, whereas pUL138 expression from the 1.4-kb transcript is inhibited under these same conditions. Differential utilization of the

UL138 transcripts and their respective encoded proteins may regulate the outcome of viral infection in a cell type or cell context dependent manner. The interaction of these proteins during HCMV latency is the focus of ongoing research. In addition, this work represents preliminary data regarding the type I interferon (IFN) response during HCMV during productive infection in MRC5 fibroblasts and during the establishment of HCMV latency in CD34+ HPCs. 15

CHAPTER 1

Literature Review

1.1 Human Cytomegalovirus

Human cytomegalovirus (HCMV) is a ubiquitous virus that persists through a latent infection in 60-90% of the population worldwide. The highest proportions of infected individuals are found in developing countries and areas dominated by lower socioeconomic groups. This number reflects people who, after asymptomatic primary infection, went on to harbor a latent infection (76). HCMV can be found in, and is transmitted through, bodily fluids from infected individuals. In infected children, viral shedding can occur for years before latency is established; consequently, children in group care centers make up an important infectious source of the virus. HCMV is not typically harmful to the host unless it reactivates, which happens when the host becomes immunocompromised, which is common with patients with advanced HIV disease or solid organ and bone marrow transplant recipients (116, 190). However, the long term effects of HCMV latency are unknown and current research implicates an association between atherosclerosis and immune senescence during latent

HCMV infection, in otherwise healthy individuals (116, 158, 168).

HCMV is the leading cause of infectious disease related birth defects and expectant mothers who do not have immunity to HCMV have the greatest risk for transmitting the virus to the fetus (30). Congenital HCMV infection affects 1 in 16

100 live births in the US with 10% of those infections progressing to permanent disabilities such as hearing loss, cognitive defects, seizures or even death (19,

53, 116). Although the rates of virus transmission do not vary according to the stage of pregnancy, birth defect severity increased in fetuses infected during the first half of the pregnancy (163). These effects have prompted HCMV to be designated as a high priority for vaccine development by both the National

Vaccine Advisory Board and also the Institute of Medicine of the National

Academy of Sciences (6, 19).

HCMV is a member of the beta herpes family of herpesviruses and is characterized by slow replication in cell culture, a restricted, species-specific host range and a latent cellular reservoir consisting of hematopoietic progenitor cells

(HPCs). HCMV is an enveloped virus and viral DNA (~230-kb) is packaged within an icosahedral (T=16) nucleocapsid and the space between the nucleocapsid and the envelope is filled with a mixture of viral proteins and mRNAs termed the tegument (Fig. 1) (110). Interestingly, betaherpesvirus genomes constitute the greatest amount of genetic diversity among all the herpesviruses. For example, the human and rodent CMV genomes are very diverse in that only ~70 genes are conserved between the viruses out of an estimated coding potential of ~160 gene products (38, 137). The organization of the HCMV genome is divided into the unique short (US) and (UL) regions that are flanked by terminal repeats that are involved in cleavage and packaging of viral genomes during replication. HCMV 17

terminal repeats also allow for genome isomer formation amongst the US and UL regions.

Figure 1. Human cytomegalovirus virion organization. HCMV has a dsDNA genome that is protected by an icosahedral nucleocapsid surrounded by tegument proteins. The virion’s envelope is derived from host cell membrane and expresses viral glycoproteins. Each virion component is listed and diagramed in an electron micrograph (left) of a purified HCMV virion and a schematic (right). The virion is approximately 200 nm in diameter.

The mechanism of virus attachment and entry into the host cells is unclear, as many as 5 envelope glycoproteins likely mediate this process in a wide variety of cell types (110). Direct fusion of the viral envelope to the plasma membrane as well as receptor-mediated endocytosis, are thought to be responsible for virus attachment (31). This interaction may be mediated by viral envelope glycoproteins gB and gM that tether the virus to heparan sulfate proteoglycans

(HSPGs) on the cell surface (32). Once membrane fusion has occurred the 18

tegument proteins and the viral nucleocapsid are released into the cytoplasm.

Upon release, many of the tegument proteins elicit an innate immune response in the host cell, yet many tegument proteins also aid in viral replication once the

HCMV genome is delivered to the nucleus via a microtubule network and nuclear pore docking (110).

Once the viral genome enters the nucleus, viral gene expression follows a highly regulated cascade during a productive (lytic) infection (90). The immediate early (IE) genes are the first to be expressed following infection and the resulting protein products act by regulating host cell functions while promoting viral gene expression. The IE1 and IE2 proteins are transcriptional regulators that are needed for viral gene expression during a productive infection (43, 109).

Complete lytic viral replication involves the early (E) genes followed by the late

(L) genes that function to replicate viral DNA and assemble the virion, respectively. Once the replicated DNA has been packaged into a nucleocapsid the tegument proteins along with other viral nonstructural proteins facilitate a two- stage envelopment and egress process that eventually leads to the release by exocytosis at the plasma membrane (110).

Interestingly, a degree of cellular tropism exists among strains of HCMV that is associated with infection with particular strains of HCMV. The two most predominant strains are clinical and laboratory-adapted strains of HCMV that are distinguished by the presence of the ULb′ region (Fig. 2A) (23, 38, 112). The

ULb′ region consists of ~20 putative ORFs that was lost after extensive serial 19

passage of HCMV clinical isolates in permissive fibroblast cell lines with the goal of generating an attenuated virus to use as a vaccine candidate (Fig. 2B). Strains that have lost the ULb′ region are termed laboratory adapted strains (ie AD169 and Towne) and while they have a growth advantage in cultured MRC5 fibroblasts, they have lost the ability to replicate well in primary macrophages and dendritic cells, whereas clinical strains of HCMV are unaffected (110).

Consequently, work has been done to maintain the genome of clinical isolates of

HCMV (ie FIX, Toledo and TB40E). This was accomplished through limited passage in MRC5 fibroblasts as well as the generation of the viral genome as a bacterial artificial chromosome (BAC) that allowed for genetic manipulation in bacteria and maintenance of the ULb′ region the genome of HCMV isolates (57).

Interestingly, recent work has implicated the ULb′ region, and specifically pUL138 encoded by the UL138 ORF within the in the ULb′ region, with the ability of clinical strains of HCMV to establish a latent infection in its host (51, 124). HCMV latency is the focus of work in our laboratory.

20

Figure 2. Genome organization of laboratory adapted and clinical strains of HCMV. (A) Schematic of the HCMV genome [adapted from (113)]. The genome is approximately 230 kb and is organized into a unique long region (UL) and a unique short (US) region that is flanked by inverted repeat sequences (R) that associate to either the long (L), short (S), or internal (I) or terminal (T) regions of the genome. The arrows portray the relative orientation of each segment and predicted ORFs are numbered. Laboratory adapted strains (top) lack ORFs UL133-UL152 and instead have a duplication of the RL 1-14 region. Clinical strains of HCMV (bottom) have retained the ULb′ region. (B) A detailed representation of the ~15 kb, ULb′ region that is necessary for the establishment of HCMV latency (51). Predicted ORFs are shown as arrows. Little is know about the protein coding capacity of these ORFs, however, UL138 (highlighted) was found to encode a protein needed for HCMV latency (124).

21

1.2 HCMV Latency

Viral latency is a hallmark of the herpesvirus family. During latency, viral genomes are able to persist within the host without viral replication (116), however the virus retains the ability to reactivate and resume replication after a reactivation stimulus. Host range, growth kinetics and also the cell type specificity of the latent viral reservoir are a variety of characteristics that are used to further define the members of the herpesvirus family (Fig. 3). The alpha herpesviruses, characterized by herpes simplex virus (HSV-1 & HSV-2) and varicella zoster virus (VSV), establish latency in neuronal cells; have a relatively short reproductive cycle and a variable host range. The beta herpesviruses, of which

HCMV is the prototypical member, establish latency in hematopoietic progenitor cells (HPCs) of the myeloid lineage, have a very long replicative cycle and a restricted host range, whereas the gamma herpesviruses such as Kaposi’s sarcoma-associated herpesvirus (KSHV) and Epstein-Barr virus (EBV) establish latency in B lymphocytes and have a restricted host range. Members of each herpesvirus subfamily use unique viral factors and mechanisms in the establishment of latency.

22

Figure 3. Classification of human herpesviruses. The herpesviruses are divided into alpha, beta, and gamma subfamilies based on characteristics listed in the table. Infection characteristics related to latency and disease are listed for representative examples from each subfamily.

The mechanisms of HCMV latency are poorly understood especially in comparison with the other herpesviruses. The strict host range of HCMV, predominant use of laboratory-adapted HCMV strains, that cannot establish a latent infection, and lack of a latency model system has hindered research in this area. Despite these limitations a substantial amount of work has been done to determine the reservoir of latent HCMV. Early work identified HCMV DNA in the 23

leukocyte fraction of peripheral blood, specifically in CD14+ monocytes, in seropositive individuals (14, 173). Further, HCMV genomes were also found in leukocyte precursors such as CD34+ hematopoietic progenitor cells (HPCs)

(105, 157). It was also found that monocytes harboring viral genomes in the absence of the major transactivator immediate early gene expression could be reactivated to produce lytic gene expression and the release of infectious virus upon terminal differentiation of the cells into macrophages or dendritic cells (140,

159, 160, 174). This finding provoked increased interest in the development of latency model systems that utilizes CD34+ cells from cord blood, fetal liver and bone marrow sources.

Recent work has described the development of an in vitro model system that could maintain the differentiation state of naive infected CD34+ cells over time while retaining the ability to experimentally reactivate the virus. This approach allowed for quantitative analysis of the ability of HCMV virus strains to establish a latent infection in vitro (Fig. 4) (50). Early experiments comparing clinical to laboratory-adapted strains of HCMV showed that a 15-kb region of the

HCMV genome termed the ULb′ (UL133 through UL152) was necessary for the establishment of HCMV latency (51). Further analysis focused on testing individual ORFs within the ULb′ region for a contribution to HCMV latency. This work revealed UL138 as a predominant viral factor necessary for the establishment of HCMV latency (Fig. 5A). Specifically, after infection with recombinant HCMV viruses containing either deletion of the UL138 coding region 24

or insertion of a premature stop mutation resulted in viruses that had a partial loss of latency phenotype (Fig. 5B) (51, 124). These results suggest that pUL138 is necessary for the latent infection, although pUL138 is not sufficient for latency and that other viral determinants likely contribute to the latency phenotype.

Interestingly, pUL138 is a 21 kDa type-1 transmembrane protein that localizes to the Golgi apparatus. Taken together these results suggest that pUL138 supports

HCMV latency through a novel mechanism (124).

25

Figure 4. in vitro HCMV latency model system. CD34+ hematopoietic progenitor cells are isolated and immediately infected with viruses that have been engineered to express the green fluorescent protein (GFP) which is used as a marker for infection. At 20 hpi the GFP+ cells are purified by FACS and immediately put into long-term bone marrow culture that consists culturing cells over a stromal cell layer that can support long-term maintenance and self- renewal of hematopoietic stem cells. After 14 days in long-term culture the hematopoietic progenitor cells are collected and are split equally into to groups. The cells from the first group (reactivation) are co-cultured with MRC5 fibroblasts in a limiting dilution format. The cells from the second group (pre-formed) are lysed and then put into MRC5 fibroblast co-culture in limiting dilution format. Both cell groups are then cultured in the presence of pro-inflammatory cytokines to stimulate reactivation from latently infected cells and after 10-14 days the number of GFP positive, fibroblast infectious centers are counted. Comparison between the reactivation and pre-formed groups is used to determine the extent of HCMV latency. For example, if the amount of virus from the reactivation and the pre- formed groups is equal then the virus had no potential for reactivation and therefore did not establish a latent infection (lytic phenotype). Alternatively, if the amount of virus in the reactivation group is greater than that of the pre-formed group then the reactivation stimulus was able to reactivate the virus from a latent state, thus a latent infection was established (latent phenotype).

26

27

Figure 5. pUL138 promotes a latent infection in CD34+ hematopoietic progenitor cells. [adapted from (51) and (124)] (A) The latency phenotype of virus infection of CD34+ HPCs (MOI = 2) was analyzed using the in vitro latency model system described in Fig. 4. The bars represent the fraction of GFP+ wells in the reactivation (green bars) and pre-formed (blue bars) cell groups after 10-15 days co-culture with MRC5 fibroblasts. The HCMV clinical strain FIX is used as a positive control for the latency phenotype. Alternatively the laboratory adapted strains AD169 and Towne were used as controls for the lytic phenotype. FIX∆2 is a recombinant FIX virus that lacks ORFs UL136-UL142. The latency phenotype of each of recombinant viruses lacking each of these ORFs is also shown (FIX∆ORF). (B) Similar latency phenotype analysis of FIX, FIX∆2, FIX∆UL138 and FIXUL138STOP (a recombinant virus with a premature stop codon inserted into the UL138 ORF to inhibit UL138 protein production). Comparison of the fraction of GFP+ wells between the reactivation and pre-formed cell groups resulted in P values of 0.004, 0.21, 0.03 and 0.09 for FIXwt, FIX∆2, FIX∆UL138 ad FIXUL138STOP, respectively.

The establishment of HCMV latency, while poorly understood, likely involves coordinated interactions between multiple viral and cellular determinants. While pUL138 is currently the only viral determinant found to directly function in HCMV latency, other viral factors have been identified in

HPCs from seropositive donors. For example, latency associated immediate early transcripts and proteins were found following infection of granulocyte- macrophage progenitor cells (GM-Ps) (86, 89). In addition, a viral homologue to interleukin-10 (vIL-10) has been identified in infected GM-Ps, however the exact role of the vIL-10 has yet to be determined as vIL-10 transcripts were also detected after infection with HCMV laboratory strains and expression of vIL-10 did not affect antigen presentation in neighboring cells (77, 121). Finally, an antisense transcript to UL81-82, termed LUNA, was identified in monocytes from 28

healthy seropositive donors however an exact mechanism for LUNA in latency has yet to be determined (12, 141). It will be interesting to elucidate the mechanisms by which these virally encoded latency factors promote HCMV latency.

Cellular factors have been found to be indirectly associated with the establishment of HCMV latency and also with reactivation from latency. Early work in CD34+ HPCs showed that progenitor cells with CD34+/CD38- phenotype were able to support a latent HCMV infection compared to CD34+/CD38-/Lin-

/Thy+ cells in which HCMV replicated productively, and yet other CD34+ lineages resulted in abortive infections (49). These results suggest that cell type specific factors, or lineage commitment profiles may contribute to the ability of HCMV to establish a latent infection.

Additional research suggests that viral transcription from the major immediate early promoter (MIEP) is inhibited upon the establishment of HCMV latency and MIEP transcriptional repression is relieved upon HCMV reactivation

(142). Cellular factors involved in this process include inhibition of the MIEP by promyelocytic leukemia nuclear body (PML-NB) proteins; death associated protein-6 (Daxx), ets-2 repressor factor (ERF), and ying yang-1 (YY1). The sum of the effects of these cellular factors on the establishment of HCMV latency is repression of the MIEP through chromatin modifications that render the MIEP inaccessible for transcriptional activation (96, 132, 148, 171, 182, 183). It should be noted, however, that a correlation has not been found between the viral 29

determinant of HCMV latency, pUL138, and cellular repression of the MIEP (124) suggesting that these processes may be independent of one another.

Interestingly, cellular differentiation of latently infected cells is associated viral reactivation and consequent reactivation measured by expression of IE1 and IE2 proteins. For example, it is thought that cellular differentiation triggers chromatin acetylation surrounding the MIEP that facilitates an open formation allowing for an increased IE gene expression upon reactivation (114, 139, 140).

Interestingly, signaling pathways involved in cellular differentiation, allogenic stimulation and/or reaction to pro-inflammatory cytokines promote reactivation of

HCMV from latency, especially during stress or immunocompromized states

(134, 159). These results suggest that chromatin remodeling plays an important role in HCMV latency and reactivation, however the interplay between latency associated cellular factors and viral determinants that mediate chromatin regulation during HCMV latency and reactivation have yet to be determined.

Viral gene expression in latently infected cells has not been fully characterized, however, viral gene array data demonstrates an overall decrease in viral gene expression as viral latency is established (27, 50). Microarray data shows that IE1 and IE2 genes are not expressed, or expressed transiently at low levels, during HCMV latency (50). Preliminary studies show that UL138 is encoded on two transcripts during HCMV infection that share a 3′ end, as there are two polyadenylation signals downstream of UL138b (Fig. 6). These transcripts were transcribed in the classic early phase of infection based on their 30

lack of expression with cyclohexamide treatment and presence during PAA treatment (124). It is not clear whether the UL138 transcripts play a role in HCMV latency.

Figure 6. Identification of UL138 transcripts. [(124)] Total RNA was isolated from FIX infected (MOI = 2) MRC5 fibroblasts over a time course of infection and analyzed using Northern blotting with an antisense UL138-specific probe. Cells were treated with either cycloheximide (CHX) or phosphonoacetic acid (PAA) to inhibit translation or viral DNA replication respectively and to determine the kinetics of UL138 expression.

Understanding of the interplay of viral and cellular factors in the establishment of latency will undoubtedly aid in HCMV vaccine development. Current vaccine strategies have focused attention on eliciting strong neutralizing immune responses with attenuated virus strains (Towne) alone, or in combination with, non-attenuated virus (Toledo/Towne), or subunit vaccines using viral proteins glycoprotein B (gB) and pp65 with little success. These strategies, while able to 31

promote an immune response, have failed to protect against subsequent viral challenge (2, 62, 117). Unique vaccination strategies that incorporate an understanding of HCMV latency may provide a basis for the next generation of clinical trials. For instance, development of a vaccine that could inhibit HCMV reactivation from latency would be beneficial for decreasing transplant rejection in organ recipients, however, other facets of HCMV infection (eg. congenital infection) would require a more conventional vaccine to prevent infection. The long-term goal of this research is to determine the viral and cellular factors that contribute to HCMV latency in a physiologically relevant setting. This may lead to new treatment options or vaccine strategies against HCMV infection.

1.3 Mechanisms of Translation Initiation

Canonical translation initiation, also termed the scanning model of translation initiation, starts with the formation of the 48S complex at the 5′-end of the mRNA. This process is mediated by various eukaryotic initiation factors

(eIFs). Briefly, attachment of the 43S preinitiation complex that consists of an

Met initiator tRNA (Met-tRNA i)/eIF2.GTP, eIF3 and eIF4A is attached to the mRNA

5′-end cap structure (m7GTP) by eIF4F to form the 48S preinitiation complex. eIF4F is a heterotrimer made up of three functionally distinct subunits including eIF4E that binds the cap, eIF4A functions as a RNA helicase and eIF4G that acts as a scaffolding protein to bridge eIF4a, eIF4E and poly(A)binding protein

(PABP) to the ribosome through eIF3. This complex then scans the mRNA in a 5′ 32

to 3′ direction until it recognizes an initiation codon AUG in a favorable sequence context (88) (ACCAUGG). Then early initiation factors are released and the 60S ribosome subunit is recruited and binds the scanning complex to form the 80S initiation complex and elongation of the polypeptide chain begins (67, 74, 188)

(Fig. 7A).

Translational regulation of gene expression often involves use of alternative mechanisms of translational initiation that allow for rapid responses to changes in the environment. Examples include translation mediated by (i) a short upstream

ORF (uORF), (ii) ribosome shunting and (iii) expression through internal ribosomal entry site (IRES) (Fig. 7B-E).

The expression of an upstream uORF can positively or negatively regulate the translation of a downstream ORF alone or in combination with other alternative mechanisms of translation initiation (83, 84, 177, 187). Leaky scanning and/or translation reinitiation events account for the downstream cistron expression (87), although in some instances the peptide derived from the uORF can inhibit initiation of a downstream cistron (104) or can enhance translation through eliciting a conformational change that allows for IRES mediated translation (187).

Ribosomal shunting is a cap-dependent mechanism in which the 40S ribosomal subunit binds the 5′ cap, scans the mRNA and then translocates from an upstream shunt donor site past large spans of mRNA, often containing regions of complex secondary structure and uORFs, to a downstream shunt 33

acceptor site where translation ensues (63, 147).

Finally, IRES-mediated translation is considered cap-independent, however, some classes of IRESs use cap-binding translation initiation factors to stabilize

RNA secondary structure and recruit ribosomes (44, 75, 120, 152, 153). IRES elements have complex secondary structures that serve to directly recruit ribosomes for translation initiation, however, conserved secondary structures and/or sequence among IRESs is still elusive (67). Further, IRES mediated translation allows for downstream cistron expression independent of the position of the IRES in the mRNA (41). For example, IRES elements of HCV and poliovirus are found in leader sequences and 5′ UTRs whereas other IRES elements have been located in between two cistrons such as with the dicistroviridae IGR IRES, and the KSHV IRES employs a unique position contained within the upstream cistron. Overall, these mechanisms of translation initiation provide a means for translation of a selective set of mRNAs under certain cellular contexts.

34

35

Figure 7. Eukaryotic mechanisms of translation initiation. [adapted from (44, 81, 82)] (A) A simplified schematic of canonical cap-dependent translation initiation. Briefly, sequential attachment of eIF4E followed by attachment of the Met 40S ribosome along with initiator tRNA (Met-tRNA i)/eIF2.GTP, eIF3 and eIF4A bind the 5′-end cap structure (m7GTP). eIF4F then attaches to the cap to form the translation initiation complex. eIF4F is a heterotrimer made up of eIF4E, eIF4A and eIF4G. This complex then scans the mRNA in a 5′ to 3′ direction until it recognizes an initiation codon (AUG) in a favorable sequence context. Once an initiation codon is found, early initiation factors are released from the complex and the 60S ribosome subunit is recruited and binds the scanning complex to form the 80S initiation complex and elongation of the polypeptide chain begins. Alternative mechanisms of translation initiation are depicted in (B)-(E). (B) Regulation of translation initiation by a small upstream ORF (uORF). Canonical translation of the uORF upstream of a second cistron can either positively or negatively regulate the translation initiation of the downstream cistron. Briefly, uORF translation can stall the ribosome altering the mRNA structure that then allows for internal ribosomal entry at a downstream site through an internal ribosome entry site (IRES). RNA secondary structure is often found between the uORF and the downstream cistron and may facilitate translation initiation. Translation of the downstream cistron may also occur through reinitiation. Interestingly the protein product from the uORF may directly inhibit or activate translation of the downstream cistron. The exact translation initiation factors that are responsible for translation of the downstream cistron are not completely understood. (C) Ribosomal shunting is a cap-dependent mechanism that involves binding of eIFs and the 40S ribosomal subunit to the 5′ cap. The preinitiation complex scans the mRNA and then translocates from an upstream shunt donor site past large spans of mRNA often containing regions of complex secondary structure to a downstream shunt acceptor site where translation ensues. Two examples of IRES mediated translation initiation that highlight IRESs that do not rely on eIFs/ITAFs compared to IRESs that do require these factors are shown in (D) and (E) respectively. (D) Translation initiation of type I IRESs includes internal ribosome entry through direct binding of the 40S ribosome through highly compact structured RNA followed by 60S ribosomal recruitment and subsequent translation elongation. It should be noted that IRESs in this category are on viral genomic RNA that is not capped and the GCU start codon is utilized instead of AUG. (E) Translation initiation of the type 4 IRESs is independent of the 5′ cap structure and involves recruitment of eIFs, ITAFs and the 40S ribosomal subunit by RNA secondary structure that can be in an extended form. Scanning of the initiation complex ensues until a start codon is recognized and translation elongation begins. The involvement of specific eIFs and ITAFs varies among IRESs in this class.

36

1.4 IRESs

IRES mediated translation was first described for poliovirus (PV) and encephalomyocarditis virus (EMCV) RNAs (75, 120). This work elucidated the viral mechanisms involved in translation of the PV encoded polyprotein despite the lack of a 5′ cap structure and the potent inhibition of host translation. Briefly, an IRES structure at the 5′ end of PV RNAs allowed for polyprotein translation through the direct recruitment of cellular translation machinery to the site of translation initiation. Further, PV encoded 2A protease promoted viral replication and expression through cleavage of the eukaryotic initiation factor 4G (eIF4G), which is necessary for classical cap-dependent translation initiation, thus inhibiting host translation (13). This example highlights an IRES function that is important to the viral life cycle of the PV RNA viruses that rely on IRES mediated expression for both viral gene expression and replication.

Interestingly, the first cellular IRES was identified in poliovirus infected cells that had continued expression of immunoglobulin heavy chain binding protein,

BiP, despite inhibition of host cell translation (99, 151). Since then, IRES mediated translation been described in a number of DNA viruses and cellular transcripts (42, 63, 102, 161). The biological importance of internal initiation is highlighted through the many complex regulatory mechanisms that allow for

IRES mediated expression under varied cellular contexts.

IRES mediated translation initiation from cellular IRESs are most active 37

during mitosis, cell stress, or viral infection when canonical cap-dependent translation mechanisms are inhibited (9, 162, 188). For example, IRES mediated expression of the X-chromosome-linked inhibitor of apoptosis (XIAP) is induced under cellular stresses, such as serum starvation, γ-irradiation and pro- inflammatory cytokine (IL-6) treatment (66, 69, 186). Increased XIAP protein levels are crucial for cell survival through the inhibition of apoptosis induced caspase activation (95). Collectively, these results implicate IRES mediated translation during cellular contexts in which canonical translation mechanisms are inhibited.

Cellular context specificity is often achieved through the IRES trans- activating factors (ITAFs) that are involved in IRES mediated expression through their role as non-canonical translation initiation factors (65, 92, 100, 162). ITAFs provide an additional level of regulation for IRES mediated expression. For example, ITAF relocalization and cell type specificity have been found to regulate

IRES activity (36, 91, 106, 126, 162). Further, attenuated poliovirus strains, used in vaccination, have substitutions in the IRES region that render the virus inactive in translation initiation in neuronal cells (165, 167). It is thought that this effect is due to the inability of ITAFs to bind the mutated IRES (63). It is tempting to speculate about the importance of the loss of IRES function in the decreased neurovirulence of attenuated PV strains.

ITAFs can bind IRES elements in both the cytoplasm and the nucleus, however, some mRNAs may require a ‘nuclear experience’ to acquire 38

association with a particular nuclear ITAF that is required for IRES activity (156).

The requirement of the nuclear experience may explain why some cytoplasmic

RNA virus IRESs are exceptionally active compared to cellular IRESs or DNA virus IRESs following RNA transfection into cells (42, 166). Common regulatory

ITAFs include polypyrimidine tract binding protein (PTB), heterogeneous nuclear ribonuclearproteins (hnRNPs), La autoantigen (La) and poly(rC)-binding protein 2

(PCBP2). Each ITAF contains multiple RNA-binding motifs and likely acts as

RNA chaperones to direct or stabilize IRES formation to facilitate internal translation initiation.

ITAF usage is a defining characteristic from which to categorize viral IRESs.

Generally, there are 4 distinct groups of viral IRESs based on use of: translation initiation factors and/or ITAFs, proposed secondary structure, the distance between the start codon and the IRES, and IRES activity in rabbit reticulocyte extracts (RRL) (81). The first group, exemplified by the Cricket paralysis virus

(CrPV), does not use protein factors to bind the ribosome. Instead this group relies on tightly-folded, globular RNA tertiary structure that includes three essential pseudoknots and two conserved stem loops for translation initiation

(Fig. 7D). IRESs from the second group also directly bind the 40S ribosomal subunit, however the RNA structure of this group is in an extended form and a subset of the canonical translation initiation factors including eIF3 and eIF2 must be recruited before translation can occur. Hepatitis C virus (HCV) is an example of this group. EMCV is a member of the third group of IRESs that use canonical 39

translation initiation factors and ITAFs to initiate translation at the 3′ end of the

IRES. These IRESs, as well as the previously described IRES groups are active in RRL. In contrast, IRESs in the fourth group, which includes poliovirus, are not active in RRL and require cell type specific ITAFs for full activity. In addition, these IRESs also use some translation initiation factors and initiate translation at an AUG downstream of the IRES (Fig. 7E). Interestingly, there is an inverse relationship between the ability for RNA secondary structure formation alone and the number of factors that are required for IRES functionality (41). Classification of cellular IRESs has been difficult, however recent work suggests that strong cellular IRESs have weak secondary structure, which may suggest that more translation initiation factors and/or ITAFs may aid in IRES function in the cellular context (184).

Interestingly, IRES elements have been found in many herpesviruses but not in HCMV. The work comprised in this doctoral thesis describes the first IRES characterized in HCMV. Further, IRESs found in other herpesvirus are associated with polycistronic transcripts generated from latency-associated regions of herpesvirus genomes and may represent a conserved strategy for the regulation of protein expression during latency (16, 29, 54, 72, 97, 169, 170). For example, Kaposi’s sarcoma-associated herpesvirus (KSHV) encodes a cluster of genes, LANA (ORF73), v-cyclin (ORF72) and v-flip (ORF71) that are important for latency (4, 119). Protein expression of v-flip is mediated by an IRES element that overlaps the upstream v-cyclin ORF (16, 54, 97). Similarly, the Epstein-Barr 40

virus (EBV) nuclear antigen-1 (EBNA1), the only nuclear protein expressed during all types of EBV latency, is also expressed via an IRES which is present in the U leader exon inherent to all EBNA1 mRNA despite which promoter is utilized

(72). Further, IRES-mediated translation of immediate-early proteins RLORF9

(170) and pp14 (169) in Marek’s disease virus and the expression of the immune evasion ORF, MK3 in murine gammaherpesvirus 68 are more examples of IRES- mediated protein expression related to herpesvirus latency (29). Collectively, these findings demonstrate an attractive mechanism by which herpesviruses may tightly coordinate the expression of multiple proteins required for complex host- virus interactions that impact the outcome of infection.

1.5 The Type I IFN Response

HCMV infection requires successful circumvention of both the innate and adaptive immune responses. The innate immune response is the first line of defense against viral infection. Briefly, an antiviral state within an infected cell is established through the production of the type I interferons (IFNα and IFNβ) that elicit a signal transduction cascade that results in the expression of cellular factors that protect the cell, and neighboring cells, against viral infection (Fig. 8)

(150, 164). For example, during HCMV infection, toll like receptor 2 (TLR2) stimulation subsequently activates NFκB, interferon response factor 3 (IRF-3), and AP1 (ATF-2/c-Jun) to promote the production of the type I interferons (IFNs) 41

(138). Next, secreted type I IFN binds to the IFN α/β receptor starting a signaling cascade through the JAK/STAT pathway that ultimately leads to increased expression of a family of genes termed the IFN stimulated genes (ISGs).

The proteins expressed from ISGs actively work to inhibit virus replication

(1). The main mechanisms by which an antiviral state is achieved include expression of 2’,5’-oligoadenylate synthetase (OAS) which inhibits viral gene expression by activation of RNase L (80, 103). Upregulation of MHC class I molecules is also prominent in IFN stimulated cells (149). Furthermore, other anti-viral factors induced in the innate immune response include RNA-activated protein kinase R (PKR) and myxovirus resistance A (MxA) protein which inhibit translation initiation or target viral RNA synthesis respectively (3, 150). In addition, the type I IFN response also recruits other cellular members of the innate and adaptive immune system to the site of infection and increases the cytolytic activity of NK cells (17, 175).

Throughout evolutionary time, viruses have developed various strategies to successfully evade the innate immune response. Many of these target the type I

IFN response signaling pathway. For example, the major immediate early gene products during HCMV infection, IE1 and IE2, have been implicated in inhibiting the innate anti-viral response in fibroblasts (Fig. 8) (118, 172). Current data show that IE2 effectively blocks induction of IFNβ by inhibiting NFκB and IRF-3 activation after infection with laboratory-adapted strains of HCMV in fibroblasts

(172). Additionally, IE1 binds to signal transducer and activator of transcription 42

(STAT) proteins STAT1 and STAT2 to inhibit association with interferon response factor 9 (IRF-9) (118). Both of these mechanisms promote viral replication by inhibiting the downstream production of antiviral host factors.

Further, microarray experiments found that the number of IFN-inducible genes, including IFNβ, was significantly decreased in HCMV infected fibroblasts in comparison to cells infected with a UV inactivated virus (18). This finding demonstrates that HCMV infection induces a robust innate IFN response and that viral gene expression is required to counteract that response. HCMV is also able to inhibit 2’5’-oligoadenylate synthetase, and subsequent MHC class I upregulation by JAK1 and IRF9 expression in IFNα treated cells (22).

Collectively, these data suggest that HCMV is able to evade the innate immune response. It is unknown whether the ability to evade the innate immune response contributes the establishment of HCMV latency.

43

Figure 8. The type I IFN response in HCMV infected cells. There are three distinct phases of the type I IFN response in HCMV infected cells. First, induction of the type I IFN response includes HCMV entry into the cell and subsequent stimulation of toll-like receptors (TLR), which activates NF-ΚB, IRF3 and AP1. These transcription factors translocate into the nucleus where they activate the expression of IFNα and IFNβ. The second phase includes IFNα/β receptor (IFNα/β R) signal transduction in infected and neighboring cells. Briefly, binding of IFNα/β to the receptor allows for association of Tyk2 and Jak1 to the receptor. These proteins then phosphorylate STAT1 and STAT2 allowing them to dimerize and associate with p48. This complex translocates into the nucleus where IRF9 associates to allow for the third phase of the IFN response that includes IFN induced gene transcription for effector protein expression. These proteins include PKR, OAs and MxA that act to inhibit viral translation, inhibit viral replication and viral RNA synthesis respectively. HCMV IE2 and IE1 proteins are also shown and can inhibit the type I IFN response at the first and third stages of the IFN response respectively. IE2 inhibits NF-ΚB and IRF3 activation to inhibit IFNα/β production. Alternatively, IE1 binds to phosphorylated STAT1 and STAT2 to inhibit association with IRF9 to inhibit the transcription of the IFN responsive effector proteins. 44

1.6 Type I IFN Response and HCMV Latency

A connection between the type I IFN response and the establishment of latency is found in other models of herpesvirus infection. For example, herpes simplex virus (HSV) encodes for a protein, ICP0, which antagonizes STAT1, thus effectively promoting viral replication. However, in the absence of ICP0, the type I

IFN response is able to suppress HSV replication into a latent state (58).

Similarly, type I IFN has been shown to inhibit reactivation of murine gammaherpesvirus from latency (11). Transcriptional regulation of Epstein-barr virus (EBV) latency associated proteins was found to be regulated by interferon response factor 7 (IRF-7). Further, during EBV latency, IRF-7 binds to the interferon-stimulated response elements (ISRE) found in the promoter region upstream of the Epstein-barr nuclear antigen 1 (EBNA1) which is associated with

EBV type III latency (191). In addition, viral homologues of IRF-1 through IRF-4 have been found in KSHV infection and have been implicated in regulation of the type I IFN response by altering histone acetyltransferase activities of proteins associated with the cellular IRFs to alter transcription of the type I IFNs (45).

Some of these proteins also inhibit PKR and alter the activation state of NFκB to downregulate type I IFN induction. Of particular interest is vIRF3, also termed latency associated nuclear antigen 2 (LANA2), which is constitutively expressed during latency and has been found to promote IFNα production in certain cell types (98, 143). Collectively, these data suggest that the herpesvirus family of 45

viruses have evolved mechanisms to subvert the antiviral state of the type I interferon system to promote latency.

It is unclear if HCMV uses similar mechanisms during the establishment of latency. However, it is reasonable to postulate that the latent viral state has evolved largely due to the pressure exerted by innate immunity and the type I

IFN response. In support of this hypothesis, key innate immune response evasion factors are within the ULb′ region that is essential for HCMV to establish a latent infection (51, 180). These include a viral MHC homologue and an inhibitor of NK activating ligand CD155 which suggests that the interplay of immune evasion and CMV latency may be more closely related than previously thought (90, 175, 180). In addition, other immune evasion gene products, including a TNF receptor homologue and UL142 which confers increased resistance to NK lysis also map to the ULb’ region (127, 180). Collectively, it is tempting to speculate a contribution of the innate immune response to HCMV latency.

46

CHAPTER 2

Materials and Methods

2.1 Cells

Human embryonic lung fibroblasts (MRC5) and human embryonic kidney

293 (HEK-293) (ATCC, Manassas, VA) cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum

(FBS), 10 mM HEPES, 1 mM sodium pyruvate, 2 mM L-glutamine, 0.1 mM nonessential amino acids 100 U/ml penicillin, and 100 µg/ml streptomycin. For experiments requiring serum starvation, MRC5 cells were cultured in normal growth media without serum for 24 hr prior to nucleofection and through the duration of the experiment. Mononuclear cells were isolated from cord blood using Ficoll/Paque Plus (Amersham Pharmacia Biosciences) density gradient centrifugation from donors at University Medical Center at the University of

Arizona with a protocol approved by the Institutional Review Board. Magnetic cell separation (Miltenyi Biotec) was used to enrich for CD34+ hematopoietic progenitor cells. CD34+ cells were maintained in IMDM supplemented with 10%

BIT serum substitute (StemCell Technologies), 2 mM l-glutamine, 20 ng/ml human low-density lipoprotein (Sigma), and 50 µM 2-mercaptoethanol.

47

2.2 Viruses

FIX virus strain bacterial artificial chromosome (BAC) that expresses the green fluorescent protein was used as the parental strain (FIXwt) for all recombinations described. Viruses were propagated by electroporating infectious

BAC DNA into MRC5 cells and purified by density gradient centrifugation through a 20% D-sorbitol cushion at 20,000 rpm in an SW28 rotor (Bechkman Coulter,

Fullerton, CA) for 80 min at 22°C. Virions were resuspended in Iscoe’s modified

DMEM containing 2% bovine serum albumin (BSA) and stored at -80°C.

Infectious virus yields were determined by tissue culture infective dose-50

(TCID50) on MRC5 cells.

Recombinant viruses (FIX-UL133myc, FIX-UL135myc, FIX-UL136myc, FIX-

LUC-ΔUL138) were constructed in E. coli by linear recombination in a two-step positive/negative selection method that leaves no trace of the engineering process (124, 178, 189). Briefly, in the first step, PCR products containing galK flanked by homologous viral sequences target the galK insertion into the viral

BAC by homologous recombination. In the second step, the galK insertion was replaced with PCR products containing the myc epitope tag flanked by viral sequences homologous to the desired insertion site. Similarly, the FIX-LUC-Δ138 recombinant virus was generated by replacing galK with PCR product encoding firefly luciferase amplified from the pGL3-Basic plasmid (Promega).

Recombinants were selected on M63 minimal plates containing 0.2% 2- deoxygalactose and 15 µg/ml chloramphenicol and further screened by BAC 48

digestion, PCR, and sequencing. The generation of FIX∆2, FIX-UL138myc and

FIX∆UL138STOP recombinant viruses was described previously (51, 124). All primers used for virus recombination are listed in Table 1.

2.3 Plasmids

All primers used for plasmid cloning are listed in Table 2. Individual HCMV

ORFs were cloned into the pGEM-T Easy vector (Promega) to generate antisense riboprobes recognizing ULb′ coding sequences for mapping of the

UL138 transcripts. The bicistronic, dual luciferase reporter construct (pCDNA3- rLuc-polIRES-fLuc) containing the poliovirus IRES, was kindly provided by N.

Sonenberg, (McGill University, Canada) (131). The poliovirus IRES was excised using KpnI and BamHI restriction sites and portions of UL138 cDNA upstream of

UL138 or the KSHV ORF72 IRES (nt pos. 123206 to 122973) (54) were cloned into the plasmid using primers flanked with KpnI and BamHI restriction sites and the FIXwt BAC DNA or KSHV genomic DNA as the template. The KSHV genome

DNA was derived from the KSHV-positive primary effusion lymphoma (PEL) cell line BCBL-1, a kind gift from P. Schaffer (University of Arizona). Sequences 5′ of

UL138 or the KSHV-IRES were cloned into the promoterless pGL3-Basic plasmid

(Promega) using primers with KpnI and HindIII restriction sites. Generation of pGL3-MIEP1400 was described previously (124). cDNAs (pFL) encoding each of the UL138 transcripts where FL had been substituted for UL138 were cloned into pEF1/myc-His-B (Invitrogen) using XbaI and PmeI restriction sites and FIX-LUC- 49

∆UL138 as the template. The PV 2A protease was cloned into pEF1/myc-His-B

(Invitrogen) using BamHI and EcoRI restriction sites using the complete PV genome expression construct kindly provided by B. Semler (UC Irvine) (59). This construct was engineered with in-frame start and stop codons using primers.

The pCIG lentivirus expression vector (73) obtained from L. Lybarger

(University of Arizona) was further modified to (i) provide a more versatile multiple cloning site-IRES-GFP cassette resulting in pCIG2-IRES-EGFP and (ii) to remove all post-transcriptional regulatory elements including the IRES, GFP, and WPRE resulting in pCIG2-IGW. Briefly, pCIG2-IRES-EGFP was generated with an expanded multiple cloning site (MCS) by annealing and elongating two primers with partial homology (bold) with EcoRI and BamHI restriction enzyme sites (underlined) at each end [FWD: (5′-GGGG GAATTC GG TGA TCA GGG

TAT ACG GCT CGA GGG GAT ATC GGG TTT AAA CGG GGC GC-3′) REV:

(5′-GGGG GGATCC CCT TAA TTA ACC TTC GAA CCG CGA TCG CCC GGC

GCG CCC CGT TTA AAC C-3′). Next, the ECMV IRES was PCR amplified from pIRESpuro (Clonetech) using primers containing BamHI and BglII enzyme sites

(underlined) [FWD: (5′-GGGG GGATCC GCC CCT CTC CCT CCC CCC CCC

C-3′) REV: (5′-GGGG AGATCT GGT TGT GGC AAG CTT ATC ATC G-3′).

EGFP was amplified from pCMS-EGFP (Clonetech) using primers containing

BglII and SalI enzyme sites (underlined) while maintaining the contextual Kozak sequence (bold) immediately preceding the ATG [FWD: (5′-GGGG AGATCT

CGC CAC CAT GGT GAG CAA GGG CGA GGA GC-3′) REV: (5′-GGGG 50

GTCGAC CTG CCC CAG CTG GTT CTT TCC GCC-3′). The MCS, ECMV-IRES and EGFP PCR amplicons were digested, ligated and inserted into the EcoRI and SalI sites of pCIG lacking the parental IRES and GFP to form pCIG2-IRES-

EGFP. The modified MCS contains the following enzyme sites in 5′ to 3′ order:

XbaI, NheI, EcoRI, BclI, BstZ17I, XhoI, EcoRV, PmeI, AscI, AsiSI, BstBI, PacI, and BamHI. The pCIG2-IRES-EGFP vector was used to clone individual myc- tagged ORFs amplified from pEF1/myc-His-B expression plasmids containing each myc-tagged ORF using the EcoRV restriction site. For vectors containing

UL138 cDNAs, the pCIG2-IRES-EGFP expression vector was further modified to pCIG-IGW. Briefly, the IRES, EGFP and the WPRE post-transcriptional regulatory element were excised using PacI and KpnI followed by re-ligation.

UL138 cDNAs were cloned into pCIG-IGW using XbaI and PmeI, or EcoRV and

AsiSI restriction sites with FIXwt BAC as the template.

2.4 Transient transfection and luciferase reporter assays

For transient expression in HEK-293 cells molar equivalents (~1-7E10 copies) of the indicated constructs were transfected into confluent HEK-293 cells using Lipofectamine 2000 (Invitrogen) and Opti-MEM I (Invitrogen) according to the manufacturer’s recommendations. At 4 hr post transfection, the medium was replaced with normal growth medium and protein expression was assayed at 48 h post-transfection. Luciferase activity was measured using either the Dual-

Luciferase Reporter Assay System (Promega) or the Luciferase Assay System 51

(Promega) as described previously (124). Nucleofection of sub-confluent MRC5 cells was performed with molar equivalents (~3E11 copies) of constructs using the Basic Nucleofector Kit for Primary Mammalian Fibroblasts (Lonza) program

U-023 according to the manufacturer’s recommendations. Protein expression was analyzed at 6 hr post nucleofection. For RNA experiments, molar equivalents (~3E11 copies) of in vitro transcribed RNA was transfected into HEK-

293 cells using Lipofectamine 2000 (Invitrogen) and Opti-MEM I (Invitrogen) and

FL activity was assayed 6 hr post transfection.

2.5 Immunoblotting

Protein lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to a 0.45 µm polyvinylidene diflouride

(Immobilon-FL, Millipore) membrane. Membranes were blocked in Tris-buffered saline (TBS) (25mM Tris [pH 8.0], 137mM NaCl, 3 mM KCL, 1.5 mM MgCl2, pH

8.0) plus 5% nonfat dry milk and 2.5 mg/ml BSA overnight at 4ºC. Blots were incubated with primary antibody in TBS supplemented with 2.5 µg/ml BSA and

0.05% Tween 20 (TBS-BT) for 1 h at room temperature. Primary antibody conditions are as follows: rabbit anti-myc (71D10-Cell Signaling @ 1:800), mouse anti-myc (Clone 9E10-Millipore @ 1:5,000), mouse anti-α-tubulin (DM1A-Sigma

@ 1:12,000), and mouse anti-HCMV IE1 and IE2 (Generous gift from T. Shenk, clone 3H4 @1:100). Custom rabbit polyclonal, primary antibodies

(OpenBiosystems) were generated for pUL133 (Peptide: 52

EPKPKKGRAKDKPKGKLK), pUL135 (Peptide: ELAPPPRWSDIEELLEK) and pUL136 (Peptide: AVQATVQVSGPRENAVSPAT). Each antibody was affinity purified using the immunizing peptide and used at the following concentrations: anti-UL133 (2 µg/mL), anti-UL135 (2 µg/mL), and anti-UL136 (5 µg/mL). Blots were washed three times in TBS-BT for 15 min each and then incubated with either goat anti-mouse IgG (H+L) DyLight 680-conjugated or goat anti-rabbit IgG

(H+L) DyLight 800-conjugated secondary antibodies (Pierce) and visualized using the Odyssey infrared imaging system (Li-Cor).

2.6 Northern Blotting

Total RNA was isolated from FIX infected cells at an MOI of 2 or from HEK-

293 cells after transfection with bicistronic luciferase constructs, using TRIzol LS reagent (Invitrogen) according to the manufacturer’s instructions. Where indicated, polyadenylated RNA was further purified using the NucleoTrap mRNA mini kit (Machery-Nagel). To analyze transcript stability, 4 µg/ml actinomycin D was added to infected cells at 16 hpi and RNA was isolated over a time course.

RNA was glyoxylated (8), subjected to 1% agarose gel electrophoresis, and transferred to a Nytran nylon membrane using a UV Stratalinker (Stratagene).

Antisense riboprobes recognizing ULb′ coding sequences were generated from

PCR products cloned into the pGEM-T Easy vector (Promega, or for FL analysis, the pEF1/myc-His-B (Invitrogen) plasmid containing the reverse complement of firefly luciferase was linearized with PmeI and used to generate a [α-32P]CTP- 53

radiolabeled antisense riboprobe using T7 in the SP6/T7 Riboprobe Combination

System (Promega) according to the manufacturer’s instructions. Probes were hybridized to blots in 4X SSC (0.6 M NaCl, 0.06 M sodium citrate [pH 7.0], 5X

Denhardt’s solution 90.01% [wt/vol] Ficoll 400, 0.01% [wt/vol] polyvinylpyrrolidone, 0.01% [wt/vol] BSA, 1% sodium dodecyl sulfate, 50% formamide and 10 µg/ml denatured salmon sperm DNA at 60ºC overnight.

Membranes were washed (8), and hybridization was detected by exposure to autoradiographic film or by phosphorilluminescence.

2.7 RACE

RNA expressed during HCMV (FIXwt) infection was analyzed by RNA-

Ligase Mediated Rapid Amplification of cDNA Ends (RLM-RACE), using the

FirstChoice RLM-RACE kit (Ambion). Briefly, primary human CD34+ cells or

MRC5 fibroblasts were infected at an MOI of 5. At 18 hpi, total RNA was isolated,

DNAse-treated using the RNA-II kit (Macherey-Nagel), processed as directed by the FirstChoice kit recommendations, and reverse transcribed with 90U of

MMLV-RT (Ambion) and 20U of Superscript III (Invitrogen) and 10% DMSO. The reverse transcription proceeded for 45 minutes at 45°C, then 20 min at 55°C.

RNA template was degraded by the addition of 10U RNAse H for 30 minutes at

37°C. As a negative control, RNA was processed as above, however the tobacco acid pyrophosphatase enyzme (TAP-null) was omitted during the RNA preparation and ligation. The cDNA generated from capped mRNA species was 54

amplified using the 5' FirstChoice outer primer and the following gene-specific primers: UL138 REV - 205nt (5′-TAGCTGCACTGGGAAGACACTT-3′), UL135

REV - 43nt (5′-ACCGCCTTTTCCAAGAGTT-3′), and UL133 REV - 18nt (5′-

ATGCTGCTGCTGCTGTTGTTGTA-3′); using Phusion Polymerase (New

England Biolabs), 500 nM primers, 200 nM DNTPs, 3% DMSO, and 1x Phusion

GC buffer. Reactions were denatured at 98°C for 1 minute, then 35 cycles of

98°C for 12 seconds, 65°C for 30 seconds, 72°C for 40 seconds, followed by a final extension time of 5 minutes at 72°C. The primary PCR amplicon was then subjected to 35 cycles of nested PCR, using the 5' FirstChoice inner primer and the following primers: UL138 REV - 411nt / XbaI (5′-

GTATCTAGAGTCGTGCCAATGGTAAGCTAGA-3′), UL135 REV - 762nt / XbaI

(5′-GTATCTAGAGCAGGCACCACGCTTATCAGAA-3′), and UL133 REV - 187nt

/ XbaI (5′-GTATCTAGACACGAAGGTTTGTTCTTCAGTTTA-3′), using the same thermocycling parameters that were used during the primary PCR. Resulting products were digested with BamHI-HF and XbaI (New England Biolabs), gel purified, cloned and sequenced. For the 3' RACE, total RNA was reverse transcribed as described above, primed by a Poly-A adapter primer supplied with the FirstChoice RLM-RACE kit (Ambion). cDNA was then amplified with the 3'

Outer Primer and a primer for UL138 FWD (5′-

ATGGACGATCTGCCGCTGAACGT-3′); using Phusion Polymerase with GC buffer, 500 nM primers, 200 nM DNTPs, and 3% DMSO. Cycling parameters 55

were: 98°C for 30 seconds, then 35 cycles of 98°C for 10 seconds, 65°C for 30 seconds, and 72°C for 30 seconds, followed by a final extension at 72°C for 5 minutes. Primary PCR amplicon was then subjected to 35 cycles of nested PCR, using the 3' FirstChoice Inner primer and UL138 FWD + 20nt (5′-

TCTAGCTTACCATTGGCACGAC-3′), using the same cycling parameters. The resulting product was digested with BamHI-HF (New England Biolabs), gel purified, cloned, and sequenced.

2.8 In-vitro Transcription and Translation

pEF1/myc-His-B expression plasmids containing FIX-LUC-Δ138 cDNAs were linearized using PmeI, phenol-chloroform extracted, and transcribed using the mMessage mMachine T7 (Ambion) kit and subsequently poly-A tailed using the Poly-A Tailing (Ambion) kit, according to the manufacturer’s instructions.

Retic Lysate IVT (Ambion) was used for translation of in vitro transcribed RNA using a 1:1 mix of the -met and -leu 20X translation mix (80mM potassium acetate concentration) according to the manufacturer’s instructions.

2.9 RNA Structure Prediction

RNA secondary structure prediction was generated using the RNA Mfold server version 3.2 (192). (http://mfold.bioinfo.rpi.edu/cgi-bin/rna-form1.cgi).

Default parameters were used except for the following: Structure draw Mode: untangle with loop fix, Exterior Loop Type: flat, Structure Annotation: p-num. Free 56

energy of stable stem loops was calculated by submitting the sequence for each stem loop individually to the Mfold server.

2.10 qRT-PCR Transcript Length Analysis

RNA was isolated and DNase treated using the NucleoSpin RNA II kit

(Machery-Nagel) according to manufacturer’s instructions. cDNA was generated from 1 µg input RNA or in vitro transcribed and poly A tailed RNA using an anchored-oligo[dT]18 primer and the Transcriptor First Strand cDNA Synthesis Kit

(Roche) according to the manufacturer’s instructions. Quantitive RT-PCR was performed with the LightCycler 480 Probes Master (Roche) according to the manufacture’s instructions along with Universal Probe Library (Roche) probes and primers specific for UL133 (FWD 5′-TTTCTGGACCTGGCATCG-3′, REV 5′-

TACGCGCAGAGGAAAATCAT-3′, Probe #114 5′-CTGTGGTC-3′), UL135 (FWD

5′-GCGGTGTACGTCGCTCTAC-3′, REV 5′-GGAAACTCGGGTTTATCTATCG-

3′, Probe #55 5′-GGCAGACCATCCTCTCCTATG-3′) UL136 (FWD 5′-

GCGGCTGTCATTATCCTGAG-3′, REV 5′-TAGTACATGGCCCCGTTCA-3′,

Probe #152 5′-TCGCCGTC-3′) UL138 (FWD 5′-GGTTCATCGTCTTCGTCGTC-

3′, REV 5′-CACGGGTTTCAACAGATCG-3′, Probe #126 5′-TCTCTGGT-3′) and firefly luciferase (FWD 5′-TGAGTACTTCGAAATGTCCGTTC-3′, REV 5′-

GTATTCAGCCCATATCGTTTCAT-3′ Probe #29 5′-GAAGACGG-3′) and the prevalidated UPL Human Beta Actin Gene Assay (Roche). Linear fragments for the generation of UL133, UL135, UL136, UL138 and firefly luciferase standard 57

curves were synthesized using the following primer sets using the FIX BAC or pGL3-Basic plasmid (Promega) as a template: UL133 (FWD 5′-

ATGGGTTGCGACGTGCACG-3′, REV 5′-TTACGTTCCGGTCTGATGCTGC-3′),

UL135 (FWD 5′-ATGTCCGTACACCGGCC-3′, REV 5′-

TCAGGTCATCTGCATTGACTCG-3′), UL136 (FWD 5′-

AGAGAGAAAGAGAGTATGTCAGTC-3′, REV 5′-

TTACGTAGCGGGAGATACGGC-3′) UL138 (FWD 5′-

ATGGACGATCGTCCGCTGAA-3′, REV 5′-TCACGTGTATTCTTGATGATAA-3′) firefly luciferase (FWD 5′-ATGGAAGACGCCAAAAACATAAAGAAAG-3′, REV 5′-

TTGGCCTTTATGAGGATCTCTCTGATTT-3′). qRT-PCR reaction conditions were as follows: 95°C for 10 min; 60 cycles of 95°C for 10 sec and 60°C for 30 sec. Reactions were run on a LightCycler 480 (Roche) and data analysis was performed using the LightCycler 480 Software (Version 1.5) using second derivative max method.

Fold change of 5′ and 3′ probe amplification for each transcript was used to quantify transcript ends compared to cDNA ends derived from in vitro transcribed

RNA (iv-RNA) using the following equation (125):

where Eff is efficiency, Ref is reference sample (iv-RNA), Exp is the experimental sample, and Ct is the crossing point. The efficiency of each probe set was determined by standard curve amplification: UL133 (1.646), UL135 (1.979), 58

UL136 (1.921), UL138 (1.953) and FL (1.897).. In our analysis we considered a

2-fold change to be within the confidence interval for full-length transcripts.

2.11 qRT-PCR IE, UL138 Expression and IFN Analysis

RNA was isolated and DNase treated using the NucleoSpin RNA II kit

(Machery-Nagel) according to manufacturer’s instructions. cDNA was generated from either 2µg RNA from MRC5 fibroblasts or total sample RNA from CD34+

HPCs using oligo(dT) primers and Superscript II reverse transcriptase

(Invitrogen) according to the manufacturer’s instructions. qRT-PCR was performed using D-Lux primers (Invitrogen) designed specifically for the following targets: UL122 (5′-cgggACAGGAAGACATCAAGCCcG-3′ and 5′-

TTGTTGCGGTACTGGATGGTAAA-3′), UL138 (5′- cgggCGTCGATCTGTTGAAACCcG-3′ and 5′-CATGGCTACGGTGGTGAACTG-

3′), Viperin (5′-cggcGCTGTGTGGAAACAGAAGCcG-3′ and 5′-

CCCAACCAGCGTCAACTATCAC-3′), IRF7 (5′- cgaagTACCATCTACCTGGGCTTcG-3′ and 5′-CACCAGGACCAGGCTCTTCTC-

3′), IFNβ (5′-caacaatCTGGCTGGAATGAGACTATTGtTG-3′ and 5′-

CTTCCAGGACTGTCTTCAGATGG-3′) and ISG54K (5′- cggaaAGGCCAATGATAATCTCTTCcG-3′ and 5′-

CTGCGTCTTCATACTGATCTGC-3′), where the lowercase letters indicate non- homologous sequences. Amplified PCR products for each of the targets listed above were used in the generation of standard curves using the following primer 59

sets: UL122 (5′-GACAACCCACTCTTGCAGCG-3′ and 5′-

ATTGCGCACCTTCTCGTTGT-3′), UL138 (5′-ATGGACGATCTGCCGCTGAA-3′ and 5′-TCACGTGTATTCTTGATGATAA-3′), Viperin (5′-

GGGGGGATCCGCTCTGCTCCAGGCATCTGCC-3′ and 5′-

GGGGGTTTAAACATTACAGGCGTGACGCACTGTGCCATGAGGATTAGTAAA

GTGC-3′), IRF7 (5′-GGGGGGATCCTTATAACACCTGACCGCCACCTAACTGC-

3′ and 5′-GGGGGTTTAAACTGGAGTTCTTTTTATTAGACTGGGCGGC-3′),

IFNβ (5′-GGGGGGATCCACATTCTAACTGCAACCTTTCG-3′ and 5′-

GGGGGTTTAAACTGACTTTTGCACCAAAAATAATTTATTTTCC-3′), ISG54K

(5′-GGGGGGATCCATATATAGGTCTCTTCAGCATTTATTGG-3′ and 5′-

GGGGGTTTAACTTCTAGCTATGTATCTTCCTGTTCTCTGC-3′). Real-time

PCRs were run on an ABI 7300 real-time PCR system (Applied Biosystems) using Platinum PCR SuperMix-UDG with ROX (Invitrogen). Reaction conditions were as follows: 50°C for 2 min; 95°C for 10 min; 60 cycles of 95°C for 15 s and

58°C for 30 s, melting curve analysis consisting of 95°C for 15 s and 60°C for 30 s, and a final step at 95°C for 15 s. β-Actin expression was measured using certified Lux primers (Invitrogen) and used as a control for template loading in the reactions. Data were analyzed using the 7300 sequence detection software v1.3.1 (Applied Biosystems).

60

2.12 Interferon Manipulation

To determine IFN resistance to HCMV infection we added increasing amounts of recombinant IFNα (IFNα-2A) (PBL Interferon Source) to MRC5 cell plated on 96 well plates and GFP+ wells were counted at the indicated times post infection. CD34+ cells were pre-treated with anti-IFNα antibody (PBL Interferon

Source) for one day before infection at a concentration of 10 µg/ml and kept in media containing anti-IFNα antibody through the course of the experiment.

Table 1. Primers used for cloning expression vectors Construct Name Orientation Sequence (5´ to 3´)

FWD ggtaccTGACTCGGCAGCCACTGTA ICS-1 REV ggatccTTTGTTCTTCAGTTTACCCTTCGGTTT

FWD ggtaccGGGAATTTTTCAAAGATTCCATAATC ICS-2 REV ggatccAAGTCGGAAAATTCCTTATCACA

FWD ggtaccAAACCGAAGGGTAAACTGAAGAACAAA ICS-3 REV ggatccATGATCTTCTCTCCTGCTTGGAAT

FWD ggtaccGGACTGGCTTCCCTGGT ICS-4 REV ggatccCTTTTCCAAGAGTTCCTCGATGT

FWD ggtaccTCGTCGAGCTCTTTGTCGTCCT ICS-5 REV ggatccAAAACGGCACAGATAACGTGAAA

FWD ggtaccATGGTCGGACATCGAGGAACT ICS-6 REV ggatccTAATGACAGCCGCAAAATAGATCG

FWD ggtaccGTACAGTTGTATAGCAGCACACG ICS-7 REV ggatccTACGAGTCGCGGATGATGTTA

FWD ggtaccTTCACGTTATCTGTGCCGTTTTGC ICS-8 REV ggatccACCGTCCTCTGTCCGGATCTAC

FWD ggtaccGGTGTACGTCGCTCTACATAGGAG ICS-9 REV ggatccTCTCGTCTCGTCCACCATTT

FWD ggtaccTCCTCGTCGAGCTCTTTGTC ICS-10 REV ggatccTACGAGTCGCGGATGATGTTA

FWD ggtaccTAAGAACCTGAGCACGCCGC ICS-11 REV ggatccTCTCGTCTCGTCCACCATTT

FWD ggtaccGGGCTGGAGAATCTCTCGAA ICS-12 REV ggatccTCTCGTCTCGTCCACCATTT

FWD ggtaccGGACTTGGACGTTGGAAATAAAT ICS-13 REV ggatccTCTCTCTTTTGTACAGCACTCGC 61

FWD ggtaccGTCATTATCCTGAGAAGTGCCG ICS-14 REV ggatccTCTCTCTTTTGTACAGCACTCGC

FWD ggtaccTACGAGTCGCGGATGATGTTA ICS7-RC REV ggatccGTACAGTTGTATAGCAGCACACG

FWD ggtaccGTACAGTTGTATAGCAGCACACG ICS-7A REV ggatccTAATGACAGCCGCAAAATAGATCG

FWD ggtaccGTACAGTTGTATAGCAGCACACG ICS-7B REV ggatccATTTCCACCGTGTCGAAGC

FWD ggtaccTGAGTCGCATTCACCGGTTCT ICS-7C REV ggatccTACGAGTCGCGGATGATGTTA

FWD ggtaccATCTATTTTGCGGCTGTCATTATCC ICS-7D REV ggatccTACGAGTCGCGGATGATGTTA

FWD ggtaccTTGGACAGACTCCTACTTATAAAGC KSHV-IRES REV ggatccTTGTATATGTGAAGGCACCGAT

FWD ggtaccTTGTATATGTGAAGGCACCGATGT KSHV-IRES-RC REV ggatccTTGGACAGACTCCTACTTATAAAGCAGG

FWD ggtaccTTGGACAGACTCCTACTTATAAAGC KSHV-IRES-pGL3 REV aagcttTTGTATATGTGAAGGCACCGATGTG

FWD ggtaccGGACTGGCTTCCCTGGT ICS-4-pGL3 REV aagcttGCCTTTTCCAAGAGTTCCTCGATGT

FWD ggtaccGTACAGTTGTATAGCAGCACACG ICS-7-pGL3 REV aagcttTACGAGTCGCGGATGATGTTA

FWD ggtaccTACGAGTCGCGGATGATGTTA ICS-7-RC-pGL3 REV aagcttTACGAGTCGCGGATGATGTTA

FWD gtttaaacTTACACGGCGATCTTTCCGC FL REV tctagaATGGAAGACGCCAAAAACATAAAGAAAGGC

FWD tctagaTTGGCCTTTATGAGGATCTCTCTGATTTTTCTTG FL-RC REV gtttaaacATGGAAGACGCCAAAAACATAAAGAAAGGC

FWD gatatcGCAGTAGCGATGGGTTGCGACG FIX-3.6 FWD tctagaGCAGTAGCGATGGGTTGCGACG

FWD gatatcCGGAGCTTGAGATTCCAAGCAGG FIX-2.7 FWD tctagaCGGAGCTTGAGATTCCAAGCAGG

FWD gatatcGAGACATGCTCCACGATCTATTTT FIX-1.4 FWD tctagaGAGACATGCTCCACGATCTATTTT

REV gtttaaacTTACTGACTCATGTGAAAAGTGTGCTTTTTATTAACAGAGCAGAGGG FIX-T-3′ * REV gcgatcgcCATGTGAAAAGTGTGCTTTTTATTAACAGAGCAGAG

UL133 FWD ATGGGTTGCGACGTGCACG

UL134 FWD ATGGCCAGGACCAGGG

UL135 FWD ATGTCCGTACACCGGCC

UL136 FWD ATGTCAGTCAAGGGCGTGGAG

UL137 FWD ATGGCCACGATCAGCACGAG

ORF-Myc # REV gatatcGCTCACAGATCCTCTTCTGAGATGAGTT

UL133-pGemT Easy FWD ATGGGTTGCGACGTGCACGATC 62

REV TTACGTTCCGGTCTGATGCTGC

FWD ATGTCCGTACACCGGCCCTTC UL135-pGemT Easy REV TCAGGTCATCTGCATCGACTCG

FWD ATGTCAGTCAAGGGCGTGGAG UL136-pGemT Easy REV TTACGTAGCGGGAGATACGGC

FWD ATGGACGATCTGCCGCTGAACG UL138-pGemT Easy REV TCACGTGTATTCTTGATGATAATG

FWD AAAAAGAGGGGAGCGGATCG UL138b-pGemT Easy REV ATCACCGCCACCATTACCAC

FWD ATGCTGTGGATATTAATTTTATTTGC UL139-pGemT Easy REV TCACCGAGGCGGAGGTGGAAA

FWD GGATCCATGGGATTCGGACACCAAAACAAA PV 2A Protease REV GAATTCTTACTATTGTTCCATGGCTTCTTCTT Note: Enzyme sites are in lowercase. * used for all three cDNAs # used for all ORFs

Table 2 Primers used for recombinant virus generation Primer Name Sequence (5´ to 3´)

FIX-LUC∆-UL138-Fwd TGTACAAAAGAGAGAGACTGGGACGTAGATCCGGACAGAGGACGGTCACCATGGAAGACGCCAAAAACA

FIX-LUC∆-UL138-Rev GTCAAAACGACATTACCGCGATCCGCTCCCCTCTTTTTTCTTTTTCTCATTTACACGGCGATCTTTCCGC

FIX-UL133myc-GalK-Fwd CCGTGGCCGCGGCTCTACAACAACAGCAGCAGCAGCTACAGACCGGAACGCCTGTTGACAATTAATCATCGGCA

FIX-UL133myc-GalK-Rev GGAGATGTGGGCCAAGTCGGAAAATTCCTTATCACACCGGGGGCGGGTTATCAGCACTGTCCTGCTCCTT

FIX-UL133myc-Myc-Fwd GTGGCCGCGGCTCTACAACAACAGCAGCAGCAGCATCAGACCGGAACGGAACAAAAACTCATCTCAGAAGAGGATCTG

FIX-UL133myc-Myc-Rev AGATGTGGGCCAAGTCGGAAAATTCCTTATCACACCGGGGGCGGGTTACAGATCCTCTTCTGAGATGAGTTTTTGTTC

FIX-UL135myc-GalK-Fwd GGAAAAGGCGGTGCAGAGCGTCATGAAGGACGCCGAGTCAATGCAGATGACCCCTGTTGACAATTAATCATCGGC

FIX-UL135myc-GalK-Rev GAGGGAAGGCGTGTGCTGCTATACAACTGTACAACGGACGCGCTCGCTGTTTCGGTCTCAGCACTGTCCTGCTCCTT

FIX-UL135myc-Myc-Fwd GGAAAAGGCGGTGCAGAGCGTCATGAAGGACGCCGAGTCAATGCAGATGACCGAACAAAAACTCATCTCAGAAGAGGATCTG

FIX-UL135myc-Myc-Rev GGCGTGTGCTGCTATACAACTGTACAACGGACGCGCTCGCTGTTTCGGTCCAGATCCTCTTCTGAGATGAGTTTTTGTTC

FIX-UL136myc-GalK-Fwd CGACCGTGCAAGTAAGTGGCCCGCGGGAGAACGCCGTATCTCCCGCTACGCCTGTTGACAATTAATCATCGGCA

FIX-UL136myc-GalK-Rev TCTCTCTTTTGTACAGCACTCGCGCGGGAACGGCCCCCTCAACCCTCTTATCAGCACTGTCCTGCTCCTT

FIX-UL136myc-Myc-Fwd ACCGTGCAAGTAAGTGGCCCGCGGGAGAACGCCGTATCTCCCGCTACGGAACAAAAACTCATCTCAGAAGAGGATCTG

FIX-UL136myc-Myc-Rev TCTCTTTTGTACAGCACTCGCGCGGGAACGGCCCCCTCAACCCTCTTACAGATCCTCTTTGA

63

CHAPTER 3

UL138 transcript expression during productive and latent Human cytomegalovirus infection

3.1 Introduction

Little is known regarding the contribution of viral determinants to HCMV latency. Although HCMV latency associated MIEP products, a viral IL-10 homologue, and UL81-82 antisense transcripts have been identified in seropositive donors (12, 77, 86, 89), the putative UL138 ORF contained in the

ULb′ region, remains the only viral factor that has been directly associated with the establishment of HCMV latency using an in vitro latency model system (51).

Viral factors important to herpesvirus latency include the activity of both viral transcripts and viral proteins. An example of transcript-mediated latency includes the latency-associated transcripts (LATs) expressed during HSV latency. The

LATs promote HSV latency through miRNA silencing of lytic ICP0 protein expression and also the assembly of heterochromatin on viral lytic gene promoters (145, 176). Alternatively, the KSHV herpesviruses rely predominantly on protein expression for the establishment of latency. For example, protein expression from the KSHV ORF73 encodes a latency-associated nuclear antigen

(LANA) that functions to maintain the latent viral genome in infected cells (10,

35). In addition, KSHV latency proteins v-cyclin and v-flip function to promote cell growth and inhibit apoptosis, respectively (47). The presence of both viral transcript and protein mediated induced herpesvirus latency programs prompted 64

further investigation into the contribution of UL138 transcripts and/or protein products for the establishment of HCMV latency.

Our initial characterization of the UL138 transcripts determined that UL138 is encoded on two transcripts of approximately 3.6- and 2.7-kb in length (Fig. 6)

(124). Interestingly, both transcripts were detected as early as 6 hpi and accumulated throughout the time course of infection. Further, UL138 expression kinetics revealed that UL138 expression in fibroblasts is dependent on the synthesis of the IE proteins and is enhanced by, but not dependent on viral DNA synthesis (124). Overall, these results demonstrate that UL138 is expressed with early-late gene kinetics during productive infection in fibroblasts.

We were surprised to detect multiple transcripts encoding UL138. However, multiple transcripts have been reported for other HCMV ORFs (52, 56, 146, 181).

The size of the UL138 transcripts was also surprising especially because the

UL138 ORF is only 510 nucleotides in length. Multiple long transcripts necessitate further characterization of the UL138 transcripts. In addition, we wanted to determine the kinetics of UL138 transcript expression in CD34+ HPCs.

This work will contribute to a better understanding of the role of the UL138 transcripts in the establishment of HCMV latency.

65

3.2 UL138 is encoded on multiple transcripts in fibroblasts

and hematopoietic progenitor cells

We used Northern blotting to map the UL138 transcripts using radiolabeled antisense RNA probes to three ORFs upstream of UL138 (UL133,

UL135 and UL136) and two ORFs downstream (newORF11) (112), now referred to as UL138b, and UL139. Our analysis from both total and poly(A)+ purified

RNA from FIXwt infected fibroblasts, determined that the UL138 transcripts likely share a common 3′ end, including the sequences in UL138b. The 5′ ends of the

3.6- and 2.7-kb transcripts include ORFs as far upstream as UL133 and UL135, respectively, and antisense ribonucleotide probes to the ORFs upstream and downstream of UL138 did not identify any other transcripts in addition to the 3.6- and 2.7-kb transcripts (Fig. 9A). In addition, UL138 specific transcripts were not detected using a sense probe to the UL138 coding sequence indicating that transcription originates from a single strand in this region of the genome (Fig.

9B).

Northern blotting identified two UL138 specific transcripts of 3.6- and 2.7-kb in length, however we wanted to map the 5′ and 3′ ends of the UL138 transcripts during productive infection in fibroblasts and also during the establishment of

HCMV latency in CD34+ HPCs. To do this we used rapid amplification of cDNA ends (RACE) (Fig. 9C). Our analysis of cDNA derived from mock- or FIXwt infected fibroblast or CD34+ HPC poly(A)+ RNA at 24 hpi, determined that the 5′ end of the 3.6-kb transcript is at nucleotide position 190476, 8 nucleotides 66

upstream of the predicted start of UL133. The 5′ end of the 2.7-kb transcript starts at nucleotide position 189551, 24 nucleotides downstream of the predicted start of UL135. In addition, we identified a third transcript corresponding to 1.4- kb. The 5′ start of this transcript maps to nucleotide position 188206, 292 nucleotides downstream of the predicted start of UL136. The 3.6-, 2.7- and 1.4- kb transcripts share a common 3′ end at nucleotide position 186833, which corresponds to a polyadenylation site just downstream of the putative UL138b

ORF (previously ORF11). It is not clear why the 1.4-kb transcript was not identified in our original studies, as all RACE methods should favor the amplification of the smaller transcript. One possibility is that the 1.4-kb transcript is present in lower abundance during infection or that the RNA ligase mediated rapid amplification of cDNA ends (RLM-RACE) technology used in this study provides enhanced sensitivity. We also failed to detect this transcript by Northern blotting. The presence of this band on Northern blots may have been obscured, as it runs at the same molecular mass as 18S rRNA. Contiguous alignment of the

UL138 cDNAs with the HCMV genome suggests that splicing events are not involved in generation of the UL138 transcripts (data not shown). All three 5′

RACE products were detected in fibroblasts and CD34+ HPCs indicating that these transcripts are expressed in multiple contexts of infection.

67

68

Figure 9. UL138 is encoded on 3.6-, 2.7- and 1.4-kb transcripts in MRC5 fibroblasts. (A) Total, poly(A)+ and Mock infected RNA was isolated from FIXwt infected (MOI = 2) MRC5 fibroblasts at 24 hpi and analyzed by Northern blotting using a radiolabeled antisense probes specific to the ULb′ ORFs indicated. (B) Similarly poly(A)+ RNA from FIXwt infected MRC5 fibroblasts was harvested at 24 hpi and analyzed by Northern blotting using radiolabeled antisense probes specific to each of the ULb′ ORFs indicated or a sense probe specific to UL138. The positions of the 3- and 4-kb RNA markers are shown to the left of each blot. (C) Schematic representation of the 3.6-, 2.7- and 1.4kb UL138 transcripts including 5′ and 3′ RACE results [performed by Michael Rak]. Gene specific primers used for UL133, UL135 and UL138 5′ RACE and individual ORF gene specific primers for 3′ RACE are shown by open arrows. The results shown in each panel are representative of at least three experiments.

3.3 UL138 transcripts are specific and are present in

FIX∆UL138STOP infected fibroblasts

Next, we wanted to confirm the absence or presence of the UL138 transcripts from two FIX recombinant viruses FIX∆UL138, which is lacking the

UL138 ORF, and FIX∆UL138STOP, which was engineered with a stop codon

(TAA) at codon 15 (ATG) of the UL138 coding sequence, respectively. We compared these results to UL138 transcripts generated from FIXwt infection.

Comparison of these viruses in the in vitro latency model system will allow us to determine the contribution of the UL138 transcripts compared to UL138 protein production in the establishment of HCMV latency. Previous work determined that both FIX∆UL138 and FIX∆UL138STOP were not able to express the UL138 protein

(pUL138) after infection of MRC5 fibroblasts (124). However, we needed to confirm that the absence of pUL138 expression from the FIX∆UL138STOP virus 69

infection was due to the stop mutation and not due to a lack of UL138 specific transcripts. To do this, we isolated total RNA from infection with FIXwt,

FIX∆UL138, FIX∆UL138STOP viruses and analyzed using Northern blotting with antisense probes directed against UL138 (Fig. 10). We were unable to detect any

UL138 transcripts from infection with the FIX∆UL138 virus and both the 3.6- and

2.7-kb transcripts were present after infection with FIX∆UL138STOP. These results confirm that the UL138 coding sequence was absent from the FIX∆UL138 recombinant virus and also that the loss of pUL138 expression from the

FIX∆UL138STOP recombinant virus is due to the stop mutation and not the lack of

UL138 specific transcripts.

Figure 10. UL138 transcripts are specific and are present in FIX∆UL138STOP infected MRC5 fibroblasts. Total RNA isolated from MRC5 fibroblast cells infected (MIO = 2) with FIXwt, FIX∆UL138 and FIX∆UL138STOP viruses was analyzed by Northern blotting using an antisense probe to UL138.

70

3.4 The UL138 transcripts in FIX∆UL138STOP

infected cells are stable

Although we were able to confirm the presence of both the 3.6- and 2.7-kb

UL138 transcripts in the FIX∆UL138STOP virus infection in MRC5 cells, we needed to confirm that these transcripts were stably synthesized. To do this we analyzed total RNA derived from MRC5 cells infected with FIXwt or

FIX∆UL138STOP and treated with actinomycin D at 16 hpi. We determined that the

FIX∆UL138STOP virus synthesized both UL138 transcripts comparable to that of the FIXwt infection (Fig. 11). These results suggest that the 3.6- and 2.7-kb transcripts are stable in FIX∆UL138STOP virus infection and do not exhibit significantly different rates of decay that could abrogate protein production.

These results indicate that the UL138 transcripts synthesized during

FIX∆UL138STOP infection, while stable, do not serve as templates for pUL138 synthesis.

71

Figure 11. The UL138 transcripts in FIX∆UL138STOP infected cells are stable. MRC5 cells infected (MOI = 2) with FIX or FIX∆UL138STOP (STOP) were treated with 4 µg/ml actinomycin D at 16 hpi. Total RNA was isolated over a time course and analyzed by Northern blotting using a radiolabeled antisense probe to UL138. 18S and 28S rRNAs for each blot are shown and were stained with methylene blue to serve as a loading control (lower two panels).

3.5 Kinetics of UL138 expression in fibroblasts

and CD34+ HPCs

In order to better understand the contribution of UL138 to the establishment of HCMV latency we wanted to compare the UL138 transcript expression kinetics in MRC5 fibroblasts to expression in CD34+ HPCs. To do this, we used qRT-PCR to measure the expression of polyadenylated UL138 transcripts in MRC5 fibroblasts and CD34+ HPCs infected with FIXwt.

Transcripts were quantitated using a standard curve derived from the PCR product for an immediate early gene, UL122, compared to UL138. β-actin expression levels were used as a control for template loading. In productively infected MRC5 cells, UL138 was expressed starting at early times post infection.

The UL138 transcripts accumulated as the infection progressed to levels exceeding those of UL122 transcripts encoding the IE2 86-kDa protein (Fig.

12A).

The trend of UL138 expression differed in CD34+ HPCs infected with

FIXwt (Fig. 12B). UL138 transcripts were detected at maximal levels immediately following infection but steadily decreased during the time course of infection, 72

whereas UL122 transcripts were detected at lower levels during the early points of the time course compared to UL138. We have detected both UL122 and

UL123 in infected HPCs previously using microarray analysis (50, 51). It is not known if UL122 and UL138 are expressed in the same cell for if UL122 and

UL138 are expressed in mutually exclusive sub-populations of infected cells.

However, we have not detected IE1 72-kDa or IE2 86-kDa proteins in HPCs by either immunoblotting (124) or immunofluroesence (data not shown). Over the time course, both UL122 and UL138 transcripts decreased to levels that were undetectable by 15 days postinfection. These data are consistent with the previously observed loss of HCMV gene expression in HPCs infected in vitro

(49). The loss of detectable UL138 expression may reflect a global silencing of

HCMV transcription that is proposed to occur during the establishment of latency

(71, 114, 140). Alternatively, the loss of UL138 transcripts could also reflect a loss of viral genomes during cellular proliferation, a limitation of in vitro culture of

CD34+ HPCs (49). Nevertheless, UL138 was detected in CD34+ cells as well as

CD14+ monocytes derived from healthy seropositive individuals (51).

73

Figure 12. Kinetics of UL138 expression. Total RNA was isolated from MRC5 cells (A) or CD34+ HPCs (B) infected (MOI = 2) with FIXwt over a time course and analyzed using qRT-PCR. The green and blue bars represent data for UL138 and UL122, respectively. Amplification of the UL138- and UL122-specific D-Lux primers was quantified by comparison to standard curves generated from PCR-amplified UL138 or UL122 products. Reactions using RNA derived from mock-infected cells or infected cells where the reverse transcriptase was omitted (RT-) from the cDNA synthesis served as negative controls. The insets show real time ∆Rn curves for β-actin at each time point to demonstrate equivalent template loading into each reaction mixture. The bars represent averages from two experiments, performed in triplicate, for panel A and the averages from two experiments for panel B. Standard deviations are shown for the replicates in panel A. The brackets in panel B represent the range of the data points in the two experiments performed. [Experiment performed with Alex Petrucelli] 74

3.6 Discussion and Future Directions

UL138 is the first viral determinant demonstrated to function in promoting

HCMV latency (51). This work represents the initial characterization of the transcripts that encode UL138. UL138 is encoded on three transcripts 3.6-, 2.7- and 1.4-kb in length that have a common 3′ end at a polyadenylation site just downstream of UL138b. The region upstream of UL138 on the 3.6-kb transcript includes ORFs UL133, UL135 and UL136, whereas the majority of UL135 is contained on the 2.7-kb transcript as well as UL136 (Fig. 9A-B). UL138 is the first predicted cistron on the 1.4-kb transcript. Interestingly, we were not able to detect any evidence for splicing events in the generation of either of the transcripts (Fig. 9C), this suggests that the UL138 transcripts are co-linear.

Our work also focused on verifying the presence (Fig. 10) and stability

(Fig. 11) of the UL138 transcripts after infection with the FIX∆UL138STOP recombinant virus compared to FIXwt infection. This analysis was critical to accurately assess the contribution of either the UL138 transcripts or the UL138 protein to the establishment of HCMV latency. Subsequent work in this area has determined that the UL138 protein is required for the establishment of HCMV latency (Fig. 5B) (124). However, this experiment highlighted the ability of the

FIX∆2 virus to exhibit a complete loss of latency whereas FIX∆UL138 and

FIX∆UL138STOP exhibited partial loss of latency phenotypes. This work suggests that pUL138 is required for the establishment of latency although it allows for the 75

possibility for the contribution of the UL138 transcripts or other ORFs in this region to the latent infection of HCMV in vitro.

pUL138 is expressed in both productive infection in MRC5 fibroblasts and also during latent infection in CD34+ HPCs (124). UL138 expression was found to be dependent on IE gene expression in MRC5 fibroblasts however pUL138 was detected in CD34+ HPCs in the absence of detectable IE protein expression

(124) suggesting a context-specific aspect to the outcome of HCMV infection.

This concept reflects our results showing differential kinetics of UL138 expression in MRC5 fibroblasts compared to expression in CD34+ HPCs (Fig.

12). Our analysis included expression of all UL138 specific transcripts in both contexts of infection; however, it is interesting to speculate that differential UL138 expression from the 3.6-, 2.7- or 1.4-kb transcripts in various cellular contexts may also contribute to HCMV latency. 76

CHAPTER 4

Protein coding capacity of UL138 transcripts

4.1 Introduction

The protein coding capacity of many of HCMV ORFs is unknown. Putative

HCMV ORFs were originally identified through sequence analysis in which in frame start-to-stop codons were screened for the potential to encode at least 80 amino acids. Sequences that met these criteria were considered to be potential coding sequences (23, 38, 112). We were able to determine the coding potential of the UL138 ORF by testing protein lysates from recombinant viruses for UL138 protein expression using rabbit polyclonal antiserum generated against a C- terminal UL138 peptide. FIXwt infection showed a UL138 reactive band at approximately 21 kDa (124) that was confirmed through testing protein lysates from a recombinant virus with an in-frame, C-terminal Myc epitope tag (data not shown). Further, UL138 protein expression was not detected in recombinant viruses lacking the UL138 coding sequence (FIX∆UL138) or pUL138 expression due to a premature in-frame stop codon (FIX∆UL138STOP). Besides pUL138, the protein coding potential of the ORFs contained on the 3.6-, 2.7-kb and 1.4-kb

UL138 transcripts is unknown.

One possibility for the extensive sequence upstream of UL138 is that it may function as a 5′ untranslated region (UTR). The length of the UL138 putative

5′ UTR would be 2.8- and 1.9-kb for the 3.6- and 2.7-kb transcripts, respectively. 77

This is much larger than the average human 5′ UTR that is approximately 210 nt in length (25). Instead, the UL138 5′ UTR from the 3.6-kb transcript would rival the longest known human 5′ UTR that is contained on the Tre oncogene mRNA

(2858 nt) (107). However, transcripts as large as 10.5-kb have been identified during HCMV infection that also include multiple putative ORFs (181). In fact, viral mechanisms of transcription and translation often fall outside the bounds of cellular restraints, and HCMV is no exception. If the region upstream of UL138 functions as a 5′ UTR the translation efficiency of pUL138 may be adversely affected by the extensive sequence upstream of UL138 especially if this region is highly structured. However we detected pUL138 expression as early as 6 hpi after infection of MRC5 fibroblasts (124).

Alternatively, the region upstream of UL138 may contain sequences/structures that could function to promote the translation of pUL138.

Interestingly, translation from internal ribosome entry sites (IRESs) on polycistronic transcripts have been identified in the expression of herpesvirus latency associated factors in KSHV and Marek’s Disease herpesviruses (16, 54,

170). If the multiple ORFs upstream of UL138 encode proteins, then there is a possibility that pUL138 may be contained on a polycistronic transcripts. If this is true, then pUL138 expression from the 3.6- and 2.7-kb transcripts may involve alternative mechanisms of translation initiation.

78

4.2 Putative ORFs upstream of UL138 encode proteins

during HCMV infection

Our transcript mapping studies demonstrate extensive sequence (~0.5-2.5- kb) upstream of UL138 with the potential to encode multiple ORFs. The presence of putative ORFs upstream of UL138 raises the possibility that these transcripts encode more than one protein or that they contain a substantial 5′ UTR. To distinguish between these possibilities, we cloned each individual ORF into the pCIG2-IRES-EGFP expression vector with an in-frame C-terminal myc epitope tag. HEK-293 cells were transiently transfected with each construct and expression of the UL133, UL134, UL135, UL136, UL137, and UL138 ORFs was analyzed by immunoblotting using a mouse monoclonal antibody to the myc epitope. Only UL133, UL135, and UL136 exhibited evidence of protein coding capacity (Fig. 13A). Excluding the myc epitope tag, UL133, UL135, and UL136 in addition to UL138 coded for proteins of approximately 28-, 36-, 27-, and 21-kDa, respectively. The molecular masses of pUL133, pUL135 and pUL136 myc- tagged proteins are approximately 10 kDa greater than their predicted molecular mass based on sequence annotation, suggesting that these proteins are post- translationally modified. While we detected multiple bands for some proteins, the significance of these bands is not known. Protein products from ORFs UL134 and UL137 were not detected in this assay, suggesting that these ORFs may not encode stable proteins in a transient expression assay. It should be noted that both UL134 and UL137 are transcribed from the opposite DNA strand relative to 79

the other putative ORFs analyzed.

To determine if UL133, UL135, and UL136 encode proteins during HCMV infection, we constructed recombinant viruses with in-frame myc epitope tags inserted at the C-terminus of UL133, UL135, or UL136 of the FIX strain BAC.

Infection of MRC5 fibroblasts with the recombinant viruses termed FIX-UL133myc,

FIX-UL135myc and FIX-UL136myc resulted in protein production from each ORF at

48 hpi as detected by immunoblotting with a myc monoclonal antibody (Fig. 13B).

Each virus produced a single protein product, except for FIX-UL136myc, which produced two major protein species during infection. The larger molecular mass pUL136 band corresponds to the protein species produced following transfection of the pUL136 expression plasmid encoding the full length UL136 ORF (Fig 2A).

The smaller molecular mass species may represent a truncated form of pUL136.

HCMV IE proteins, IE1 and IE2, show similar levels of infection between the recombinant viruses such that the observed levels of pUL133, pUL135, pUL136 and pUL138 reflect their relative levels of expression during infection. These results are the first to demonstrate the coding potential of the ULb´ ORFs UL133,

UL135, and UL136.

80

Figure 13. UL133, UL135 and UL136 encode proteins. (A) Protein lysates derived from transfected HEK-293 cells expressing UL133, UL134, UL135, UL136, UL137, UL138 myc-tagged ORFs or empty vector were analyzed by immunoblotting with the rabbit anti-myc (71D10) and mouse anti-α tubulin (DM1A) antibodies. (B) Protein lysates from fibroblasts infected with FIX- UL133myc, FIX-UL135myc, FIX-UL136myc and FIX-UL138myc (MOI 0.5) recombinant viruses were analyzed at 48 hpi by immunoblotting with mouse anti-myc (9E10) and mouse anti-IE1/2 (3H4) antibodies. α-tubulin serves as a loading control. 81

4.3 pUL138 is expressed from polycistronic mRNAs

The existence of bona fide ORFs upstream of UL138 suggests that the

UL138 transcripts are polycistronic. To explore this possibility, cDNAs generated for each transcript were cloned into pCIG-IGW and molar equivalents of each were transfected in HEK-293 cells. Protein lysates were harvested 48 h post- transfection and analyzed by immunoblotting using rabbit polyclonal antisera generated against peptides specific to pUL133, pUL135, pUL136 or pUL138 coding sequences (Fig. 14A). Lysates from cells transfected with an expression construct encoding the myc tagged version of each individual ORF (as described for experiments in Fig. 13) were used as a positive controls.

The 3.6-kb cDNA expressed a 36-kDa protein corresponding to pUL133 in addition to a low intensity band corresponding to pUL138. The 2.7-kb cDNA resulted in reactive bands corresponding to pUL135 in addition to a protein specific for pUL138. The expression of pUL135 from the 2.7-kb cDNA was surprising as the 5′-end of the transcript does not include the predicted start codon for pUL135 (Fig. 9). The expression of this protein from the 2.7-kb cDNA suggests that the translation start for pUL135 may correspond with an in-frame

AUG 28-nt downstream of the 5′ end of the 2.7-kb transcript (Fig. 15). The 1.4-kb cDNA expressed two proteins, one corresponding to pUL138 and the other corresponding to a smaller molecular mass pUL136. This result was unexpected since the 5′ start of the 1.4-kb transcript start maps downstream of the predicted

AUG of UL136 (Fig. 15). This band most likely corresponds to the smaller 82

molecular mass species of pUL136 observed during virus infection (Fig. 13B).

Full-length pUL136 was not expressed from either the 3.6- or the 2.7-kb cDNAs during transfection, although it was detected during infection (Fig. 13B), suggesting that expression of full-length pUL136 may be specific for the context of viral infection. Importantly, pUL138 was expressed from each of the three cDNAs, with the greatest expression resulting from the transfection of the 1.4-kb cDNA (Fig. 14B).

Our results suggest that UL138 transcripts are polycistronic. To rule out the possibility that internal promoter activity or splicing could account for pUL138 expression we analyzed the 5′ and 3′ ends of the RNAs following transfection of the cDNA constructs by qRT-PCR. Equivalent amplification of 5′ and 3′ ends would argue against the presence of smaller distinct RNAs that may arise from internal promoter activity or splicing events since it is unlikely that distinct RNAs would be present at equivalent levels. These results were evaluated against comparable probe amplification of cDNA generated from full-length, in vitro transcribed RNA (iv-RNA) thus eliminating any possibility of mammalian promoter or splicing activity. Our results indicate equivalent amplification (within a

2 fold confidence level) of the 5′ and 3′ transcript ends derived from the transfected cells relative to iv-RNAs (Fig. 14C). We interpret this result to indicate that the transcripts supporting pUL138 expression are full-length and that monocistronic UL138 transcripts cannot be responsible for pUL138 protein production seen in Fig. 14A. In addition, transcript numbers from each cDNA 83

transfection were comparable indicating that levels of pUL138 expression cannot be explained by differences in transfection efficiency.

84

Figure 14. UL138 transcripts are polycistronic. (A) HEK-293 cells were transfected with cDNAs representing the 3.6-, 2.7- and 1.4-kb UL138 transcripts or constructs encoding individual myc-tagged ORFs and analyzed by immunoblotting using polyclonal rabbit antisera specific to pUL133, pUL135, pUL136 and pUL138 or a monoclonal antibody specific to tubulin as a loading control. (B) The intensity of pUL138 band was quantitated with a Li-Cor Odyssey imaging system and are represented graphically. (C) RNA isolated from lysates described in panel A was analyzed by qRT-PCR. Inverse crossing points (1/Ct) were determined using the Roche Light Cycler 480 Software with the second derivative max method (graphed). Fold change compares amplification of a 5´- and 3´- ends of experimental cDNAs isolated from transfected cells to cDNA derived from full-length iv-RNA. Transcripts with fold changes ≤ 2 compared to iv- RNA were considered to be full-length.

4.4 Model of the coding potential of the polycistronic

UL138 transcripts

Our data suggest that all three UL138 transcripts are polycistronic. We have generated a model to depict pUL133, pUL135, pUL136 and pUL138 expression from the 3.6-, 2.7- and 1.4-kb UL138 transcripts (Fig. 15). This model depicts the sequence surrounding the 5′ end of each transcript including the transcript start sites (vertical arrows) and the experimentally determined coding sequence for pUL133, pUL135 and the truncated form of pUL136 (underlined).

The predicted start codon for pUL135 expression as determined through a bioinformatic screen of the coding potential of the HCMV genome is shown as an asterisk (*) (23, 38, 112). Taken together, this model depicts our understanding of the coding potential from each of the UL138 transcripts.

85

Figure 15. Model of the coding potential of the polycistronic UL138 transcripts. RLM RACE data was used in conjunction with experimental data from cDNA construct transfections that depict protein expression from the 3.6-, 2.7- and 1.4-kb UL138 transcripts. The 5′ end sequence of each transcript start is shown below the transcript. Closed vertical arrows indicate the transcript start site as determined by RACE and underlined sequence indicates the experimentally proposed, translation start sites for the most 5′ ORF of each transcript. In the case of the 2.7-kb transcript, the mapped 5′ end of the transcript does not include the annotated translation start for UL135 (112) which is marked by an asterisk.

4.5 Discussion and Future Directions

Genes in the ULb′ region likely represent viral adaptations responsible for latency and persistence. The coding potential of the ULb′ region unique to clinical isolates of HCMV has been explored for only a small number of ORFs, as this region is considered non-essential for lytic virus replication. This is the first account of the protein coding capability of the proteins pUL133, pUL135 and 86

pUL136 during in expression constructs and also during HCMV infection (Fig.

13A-B). In addition, we have demonstrated that these proteins are expressed from the 3.6-, 2.7- and 1.4-kb UL138 transcripts, respectively (Fig 14A-B). These results suggest that the UL138 transcripts are polycistronic. To confirm this possibility, we developed a qRT-PCR methodology to assay the 3.6-, 2.7- and

1.4-kb transcripts for smaller monocistronic transcripts that could be generated through alternative splicing or promoter activity. We used in vitro transcribed

RNA (iv-RNA) as a control for full length RNA as bacterial in vitro transcription systems do not entail eukaryotic splicing or promoter activity. Our results determined that each transfected cDNA construct generated transcripts with approximately equal number of 5′ and 3′ ends suggesting that the 3.6-, 2.7- and also the 1.4-kb UL138 transcripts are indeed polycistronic (Fig. 14C).

Our data indicate that pUL136 may be expressed in two forms, full-length and truncated, depending on the translation template. While both forms of pUL136 are expressed during infection (Fig. 13B) only the truncated form can be detected during transfection of the cDNAs containing UL136 (Fig. 13A and 14A).

This finding suggests that infection-specific factors may be required for full-length pUL136 expression. Strikingly, full-length UL136 is not a 5′ cistron on any of the transcripts characterized. The mechanisms allowing for expression of full-length pUL136 are unknown. It is tempting to suggest a role for the truncated pUL136 in regulating the expression or activity of full-length pUL136 or contributing to

HCMV latency. Ongoing work is focused on understanding the role of each 87

UL138 transcript and the different forms of pUL136 in regulating the outcome of

HCMV infection.

We have determined that the additional ORFs present on the UL138 transcripts do encode proteins in addition to pUL138 and that these transcripts are polycistronic. These results present a significant challenge toward understanding the expression of pUL138. It is clear that the sequence upstream of UL138 does not function as a 5′ UTR, but instead pUL138 expression likely requires alternative mechanisms of translation initiation. 88

CHAPTER 5

pUL138 is expressed by alternative mechanisms of translation initiation

5.1 Introduction

Canonical eukaryotic translation initiation involves the recognition of the mRNA 5′ m7G-cap structure by the translation initiation complex and subsequent downstream scanning along the 5′ UTR to the initiation codon where the 60S ribosomal subunit joins the complex and translation ensues (64, 123). However, alternative mechanisms of translation initiation are often rapidly initiated in response to changes in the environment and are common during responses to cellular stress, apoptosis or viral infection. Our analysis of the UL138 transcripts revealed that the 3.6-, 2.7- and 1.4-kb UL138 transcripts are polycistronic and pUL138 expression likely requires alternative mechanisms of translation initiation for expression. Examples of alternative mechanisms of translation initiation that may be involved in pUL138 expression include (i) translation of a short upstream

ORF (uORF) (83, 84, 187) that regulates expression of a downstream cistron, (ii) ribosome shunting from the 5′ cap to the start codon (147) or the use of an (iii) internal ribosomal entry site (IRES) in which RNA secondary structure guides translation machinery to a site either upstream of the start codon or directly to the start codon for expression of a downstream cistron (75, 120, 152, 153).

There is little information regarding alternative mechanisms of translation initiation during HCMV infection. Regulatory uORFs have been identified in the 89

transcripts encoding HCMV gp48 and IE1, however both reports associate translation from the uORF with inhibitory effects of the downstream cistron expression. Further, leaky scanning was attributed to the expression of gp48 and the significance of the uORF in IE1 mRNA was not determined (20, 85).

Polycistronic transcripts, similar to the UL138 transcripts, have been described in

HCMV infection, however, the exact mechanisms of translation initiation involved in expression of the downstream cistrons have not been determined (52, 56, 146,

181).

Recent work highlights the involvement of alternative mechanisms of translation initiation in the expression of latency-associated proteins in other herpesvirus infections. For example, internal ribosome entry sites have been described for v-flip expression during KSHV infection, EBNA1 expression during

EBV latency, MK3 expression from murine gammaherpesvirus 68 and two IRESs have been identified in Marek’s disease virus within latency associated transcripts (16, 29, 54, 72, 97, 169, 170). Alternatively, uORF and ribosome shunting have not been identified in other models of herpesvirus latency, nor have they been attributed to an increase downstream cistron expression during

HCMV infection. Given the number of IRESs found within latency-associated regions in other herpesviruses, we wanted to explore the possibility that an IRES may mediate pUL138 expression.

IRESs were first identified in poliovirus (PV) and encephalomyocarditis virus

(EMCV) in which infections IRES mediated translation from viral RNAs allowed 90

for expression of proteins independent of a 5′ cap structure (75, 120). This pioneering work led the way for the identification of additional IRES sequences in viral and cellular transcripts (42, 63, 102, 161). Many of the IRES sequences identified to date have been screened using a bicistronic luciferase vector that allows for a quantitative means to assay for downstream cistron expression

(131). Renilla luciferase (RL) is the first cistron that is translated by canonical cap-dependent manner followed by firefly luciferase (FL) as the second cistron. If the intercistronic region contains an IRES sequence FL is expressed, however if the intercistronic region does not contain an IRES FL is not expressed.

Quantitative biochemical analysis of the relative amounts of RL and FL expression allows for the identification of IRES activity. This is typically expressed as a ratio of FL to RL in comparison to controls lacking the intercistronic region or sequences that do not contain an IRES, where FL is not expressed. This tool has been instrumental in the identification of many IRESs.

The major criticism of the bicistronic vector is that cryptic promoter activity and/or splicing events could account for monocistronic transcripts coding for FL expression. Consequently, use of the bicistronic vector necessarily requires extensive measures to eliminate these possibilities. Transcript length analysis and in-vitro transcribed RNA are typically used to eliminate alternative means of

FL expression. The benefit of using in vitro transcribed RNA from the bicistronic constructs is that splicing and mammalian promoter activity are not active in in vitro systems. Because IRES activity is inherent to the RNA secondary structure 91

of the IRES, the activity of the IRES is retained in this system, however it is know that some IRESs need IRES transactivating factors (ITAFs) for optimal expression from the IRESs. Conscious of the limitations of the bicistronic luciferase construct, we applied this approach to determine if the UL138 transcripts have IRES activity that may mediate the expression of pUL138.

5.2 A sequence overlapping the 5′ end of the UL136 ORF

promotes translation of a downstream cistron

The expression of multiple proteins from a single transcript necessitates alternative mechanisms of translation initiation. We hypothesize that pUL138 translation is mediated by upstream sequences that facilitate translation initiation.

Given the extensive sequence (~0.5-2.5-kb) 5′ of UL138, we favored the possibility that an internal ribosome entry site (IRES) was responsible for pUL138 expression. To explore this possibility, we cloned sequences 5′ of UL138 into a bicistronic reporter construct where the 5′ cistron encodes renilla luciferase (RL) and the 3′ cistron encodes firefly luciferase (FL) (Fig. 16A). The cloned sequences were designated IRES candidate sequences (ICS). A construct containing the poliovirus IRES (PV-IRES) or the Kaposi’s sarcoma associated herpesvirus IRES (KSHV-IRES) served as positive controls. A construct lacking an ICS or IRES was used as a negative control (Empty). The RL and FL activity of each construct were measured 48 hr post-transfection of HEK-293 cells and 92

IRES activity was expressed as a ratio of FL to RL. Negligible FL expression was measured in the Empty control, whereas the PV or KSHV-IRES resulted in substantial expression of the downstream FL cistron. Of the ICS sequences analyzed, only ICS-7 FL expression was comparable to the PV-IRES and to a greater extent than the KSHV-IRES (Fig. 16B). ICS-7 is a 663-nt sequence that starts 65-nt upstream of the start for UL136. ICS-5, -6, and -8 overlapped with

ICS-7 however, these sequences had moderate to low expression of FL. These results suggest that a sequence element upstream of UL138 has the potential to mediate translation of pUL138.

To further define the sequences required for maximal IRES activity, we cloned fragments extending from the 5′ and 3′ ends of ICS-7 (Fig. 16C).

Interestingly, none of these sequences showed FL activity beyond that seen with

ICS-7. ICS-10 and ICS-12 exhibited 52.7% and 47.2% of the activity of ICS-7, respectively. However, ICS-9, ICS-11, ICS-13, and ICS-14 exhibited less than

10% of ICS-7. These data indicate that additional sequence around ICS-7 is not required for activity and that the context of the sequence is likely critical for proper folding of the RNA sequence into a functional element.

Next, we sought to determine the minimal ICS-7 sequence required for

IRES function. Shorter fragments of ICS-7 as well as the reverse complement of

ICS-7 (ICS-7RC) were cloned into the bi-cistronic dual luciferase reporter construct and analyzed for FL expression (Fig. 16D). None of the sequences had comparable FL expression to ICS-7. Importantly, the reverse complement of ICS- 93

7 (ICS-7RC) exhibited diminished activity relative to ICS-7. These data are similar to those of the reverse complement of the KSHV-IRES (KSHV-RC). Our results indicate that ICS-7 is the optimal sequence element required for expression of the downstream cistron. While our results do not exclude other mechanisms of alternative translation initiation, they are consistent with this element functioning as an IRES.

94

Figure 16. A 663-nt sequence overlapping UL136 supports expression of a downstream cistron. (A) Schematic of the bicistronic luciferase construct used to screen sequences 5´ of UL138 designated as IRES candidate sequences (ICS). Known viral IRESs (v-IRES) or a construct with an empty intercistronic region (Empty) were used as positive and negative controls respectively. RL and FL luciferase activity was quantified 48 h following transfection of equal molar quantities of each construct into HEK-293 cells. The activity of each intercistronic sequence is expressed as a ratio of FL to RL activity. (B) Analysis of ICS sequences 5´ of UL138 relative to poliovirus IRES (PV), KSHV-IRES (KSHV) or empty controls. (C) Analysis of sequences proximal to and including ICS-7 to determine sequences required for maximal activity. (D) Analysis of portions of ICS-7 to determine the minimal sequence required for ICS-7 activity. The reverse complement of KSHV-IRES (KSHV-RC) and ICS-7 were also analyzed. The activity of each construct relative to ICS-7 is expressed to the right of the graphs in panels C and D.

5.3 Transcripts originating from the dual luciferase vector are full length

and ICS-7 do not have inherent promoter activity

Possible alternatives of our interpretation that downstream cistron expression results from internal ribosome entry include the presence of monocistronic UL138 transcripts resulting from (i) alternative splicing or RNA fragmentation or (ii) cryptic promoter activity. To rule out these possibilities, we examined the length of transcripts generated in HEK-293 cells transfected with the bicistronic luciferase vector by Northern blotting. A radiolabeled anti-sense

RNA probe specific to FL was hybridized to total RNA isolated from HEK-293 cells transfected with the empty bicistronic luciferase construct or the constructs containing PV-IRES, ICS-7, or ICS-7RC. RNA from untransfected cells was used as a negative control (NC). We detected a predominant reactive transcript corresponding to the predicted size of each full-length bicistronic luciferase 95

transcript (Fig. 17A). Smaller transcripts unique to the ICS-7 construct that could independently result in FL activity were not detected. These results were confirmed using qRT-PCR to demonstrate equivalent levels of RL and FL transcripts for each of the bicistronic constructs compared to cDNA derived from in vitro transcribed RNA (data not shown), suggesting that monocistronic transcripts are not responsible for expression of the FL cistron.

Previously, in vitro transcription and translation assays have been used to rule out the possibility of promoter activity in testing the existence of an IRES. We analyzed the ability of full-length bicistronic luciferase iv-RNA containing ICS-7 as a template for expression of the downstream FL cistron. The empty vector or two

ICS sequences that were either determined to be negative (ICS-9) or positive

(ICS-10) for IRES activity during our initial IRES screen were used as controls.

Transcripts were translated in vitro using rabbit reticulocyte lysates (RRL).

Similar to the IRES activity measured by DNA transfection (Fig. 16), ICS-7 and

ICS-10 supported expression of the downstream cistron, whereas only modest levels of FL were supported by ICS-9 (Fig. 17B). These data suggest that cryptic promoter activity or splicing cannot explain FL expression observed in Fig 16.

We also tested ICS-7 directly for promoter activity by cloning it upstream of

FL in a promoterless plasmid relative to the HCMV major immediate early promoter (MIEP1400). The empty plasmid or a plasmid containing the KSHV-IRES were used as negative controls. In addition to ICS-7, we included ICS-4 and ICS-

7RC in our analysis since these sequences did not stimulate expression of the 96

downstream cistron, as shown in Figure 16. FL activity was measured 48 hours after transfection of HEK-293 cells. Relative to the MIEP1400, ICS-7 did not promote significant FL activity (Fig. 17C). Instead, ICS-7 FL activity was similar to the KSHV-IRES and HCMV sequences that did not exhibit IRES activity in bicistronic assays. The FL activity from each of these constructs was 0.02-0.05% of MIEP1400 activity. It is therefore, unlikely that ICS-7 facilitates gene expression through promoter function. 97

98

Figure 17. Downstream cistron expression is not due to the presence of monocistronic transcripts specific to ICS-7. Total RNA isolated from HEK-293 cells untransfected (NC) or transfected with the indicated bicistronic constructs were analyzed by Northern blotting using a radiolabeled antisense probe specific to FL (upper panel). The expected transcript sizes were Empty 2.7-kb; PV-IRES 3.4-kb; ICS-7 3.4-kb; and ICS-7RC 3.4-kb. 18S and 28S rRNAs were stained with methylene blue and served as a loading control (lower panel). (B) iv-RNAs generated from linearized bicistronic constructs were translated in vitro using RRL. IRES activity is expressed as the ratio of FL to RL. (C) A promoterless expression plasmid (pGL3-Basic), depicted in the schematic, was used to assay for promoter activity of ICS compared to the HCMV major immediate early promoter (MIEP1400), the KSHV-IRES, or an empty vector. FL activity was measured and expression relative to the MIEP1400 is indicated below the graph.

5.4 ICS-7 Mfold prediction suggests extensive

RNA secondary structure

IRES sequences have extensive RNA secondary structures that allow for the recruitment of the translation machinery. We analyzed the RNA secondary structure of ICS-7 using Mfold, a computer based RNA modeling program (Fig.

18) (192). We used the p-num function to identify highly favorable structures. The shaded stem-loops represent predicted highly stable structures (-17 Kcal/mol) with p-num scores that correspond to 95.8-99.8% probability that these bases will pair only with each other. The likelihood of these structures forming was considerably higher compared to other structures within the same molecule independent of whether only ICS-7 or the entire 3.6- or 2.7-kb transcripts were folded (data not shown). These results support the idea that the hairpins are structurally conserved in the mRNA molecule and may be involved in IRES function (193). Our primary sequence analysis also revealed a region containing 99

a 15 base polypyrimidine tract (underlined) with a polypyrimidine tract binding protein (PTB) consensus sequence (bold) (100, 122). Polypyrimidine tracts have been found in many IRES elements may facilitate PTB binding to stabilize the mRNA and then recruit the translation machinery (79). This region is relatively unstable (-7.6 Kcal/mol) compared to the conserved hairpins. The presence of highly stable, conserved hairpins, combined with a polypyrimidine tract supports our hypothesis that this region functions as an IRES.

Figure 18. Predicted RNA secondary structure of HCMV the IRES. Secondary structure predictions were created using Mfold. The Mfold p-num option was used to identify highly favorable structures. The free energies of the three highly predictable (shaded) hairpins are shown. Each of the featured hairpins (i-iii) are enlarged. Hairpin (i) contains a polypyrimidine tract (underlined) with a conserved PTB binding consensus sequence (bold-red) (100, 122). Hairpin (iii) contains a predicted pseudoknot between nucleotide bases at the base of the hairpin (bold) with those in the loop of the hairpin. Start codons for full length and truncated versions of pUL136 are indicated with by (*) and (◊), respectively.

100

5.5 ICS-7 IRES activity is dependent on eIF4G

Many IRES structures mediate internal ribosome entry independent of translation initiation factors that associate at the 5′ cap structure or the 5′ cap itself. However some IRESs require translation initiation factors or ITAFs for full functionality (156). For example, IRES mediated translation of the poliovirus (PV) polyprotein allows for the expression of the PV 2A protease. This protease cleaves the eIF4G translation initiation factor that (i) inhibits global host cell translation and (ii) the eIF4G cleavage product facilitates expression from the PV

IRES (13). We wanted to explore the dependence of ICS-7 activity on eIF4G. To do this we transfected HEK-293 cells with the bicistronic luciferase construct containing either the PV-IRES or the ICS-7 sequence compared to a bicistronic construct lacking an intercistronic sequence (Empty). Our results indicate ICS-7

IRES activity was decreased in cells co-transfected with ICS-7 and an expression construct containing the PV 2A protease (Fig. 19). It is likely that activity from the

PV IRES is increased by co-transfection with the PV 2A protease through an increase in available ribosomes for internal translation. These results indicate that ICS-7 IRES activity is dependent on intact eIF4G. In addition, these results suggest that other translation initiation factors and/or ITAFs may be necessary for ICS-7 IRES activity.

101

Figure 19. ICS-7 IRES activity is dependent on eIF4G. Bicistronic luciferase constructs either lacking an intercistronic IRES (Empty) or containing either the PV-IRES or ICS-7 were transfected into HEK-293 cells either alone (untreated) or co-transfected with a pEF1α construct containing the PV 2A protease coding sequence (+ 2A Protease). Samples were assayed for luciferase activity at 24 h post-transfection. The activity of each intercistronic sequence is expressed as a ratio of FL to RL activity.

5.6 Discussion and Future Directions

Our work has identified a 663-nt sequence (ICS-7) within the 5′ end of

UL136 that facilitates the translation of a downstream cistron (Fig. 16). This sequence does not produce monocistronic UL138 transcripts through either promoter or splicing activity (Fig. 17). In addition, ICS-7 displays many features characteristic of IRESs including stable secondary structure and a polypyrimidine tract that could facilitate ITAF binding and translation initiation from this sequence 102

(Fig. 18). The cap-dependency of ICS-7 containing constructs was measured using the poliovirus 2A protease that cleaves the translation initiation factor 4G.

We determined that ICS-7 required eIF4G for optimal translation initiation (Fig.

19). Research into the role of other translation initiation factors involved in ICS-7 function is currently underway. The function of ICS-7 most closely resembles that of an IRES. ICS-7 would be the first determination of IRES mediated protein expression from HCMV transcripts.

Classifying an RNA sequence as an IRES must satisfy several experimental criteria to rule out other possible reasons for expression of a downstream cistron.

We have carefully evaluated transcripts generated from our bicistronic luciferase constructs to rule out the existence of monocistronic RNAs resulting from (i) a cryptic promoter or (ii) cryptic splicing events or fragmentation that could result in protein translation even in the absence of a 5′ cap. ICS-7 did not support FL expression in a promoterless vector relative to a bona-fide promoter (Fig 17C).

Consistent with this data, and also providing evidence against cryptic splicing or

RNA fragmentation, we were not able to detect smaller monocistronic transcripts originating from ICS-7 through northern blotting (Fig 17A). This data is further supported by qRT-PCR transcript analysis from previous transfections of UL138 cDNA constructs (Fig. 14C). Further, iv-RNAs containing ICS-7 supported expression of the downstream cistron in in vitro translation assays (Fig. 17B).

Taken together, these results support the conclusion that ICS-7 functions as an

IRES to mediate protein expression through internal initiation of translation. 103

However, in transfected cells compromised for cap-dependent translation by co- transfection with PV 2A protease, the expression from ICS-7 was reduced (Fig.

19). Thus, while ICS-7 exhibits IRES-like activity in all the assays reported, it more closely resembles a cellular IRES rather than the original activity described for RNA viruses such as the picornaviruses.

The HCMV IRES we describe here most closely resembles the KSHV IRES that is responsible for v-flip expression. The KSHV IRES is contained within the upstream coding sequence of v-cyclin (15, 54) similar to the IRES contained within the UL136 ORF on the 3.6- and 2.7-kb UL138 transcripts. However, unlike the KSHV IRES, the HCMV IRES likely represents only one mechanism by which pUL138 can be expressed. Our data suggest that although ICS-7 may guide pUL138 expression from the 3.6- and 2.7-kb UL138 transcripts, other mechanisms of translation initiation must be responsible for pUL138 expression from the 1.4-kb transcript. Further, we have determined that the entire ICS-7 region is necessary for IRES function. Consequently, pUL138 expression from the 1.4-kb transcript likely results from leaky scanning or readthrough. Taken together, our results suggest a dynamic interplay between the IRES function and protein coding capacities of the UL136 ORF, as we have also confirmed the coding potential of this sequence. The interplay between pUL136 expression and

IRES activity may constitute a molecular switch that may be play a role in HCMV latency.

104

CHAPTER 6

pUL138 expression from the 3.6- and 2.7-kb transcripts is stress inducible

6.1 Introduction

We have confirmed IRES activity from ICS-7 within the context of the bicistronic luciferase construct, however we would like to determine if this IRES is able to directly mediate pUL138 expression from each of the individual UL138 transcripts. We hypothesize that pUL138 expression from the 3.6- and 2.7-kb transcripts will be mediated by ICS-7 IRES activity, however, we do not expect

IRES mediated pUL138 expression from the 1.4-kb transcript as our previous

ICS screen determined that the entire ICS-7 sequence was necessary for full

IRES activity (Fig. 16). Although, we have not ruled out other means of translation initiation for pUL138 from this transcript, as it is polycistronic. Thus, we will explore the mechanisms of translation initiation that may be necessary for pUL138 expression from the 1.4-kb transcript.

IRES elements allow for regulated protein expression during cellular contexts when the canonical cap-dependent translation profile has been inhibited from hypoxia, mitosis, heat shock, differentiation or apoptosis (162). Many cellular proteins expressed during these contexts are either needed for the stress response or are involved in the recovery from the stress response (162). IRES mediated protein expression is not necessarily dependent on a cellular stress, instead, many IRES elements are unaffected by stresses and subsequent protein 105

expression is unaffected by the cellular context. However, IRES mediated translation initiation is a dynamic process and some IRESs are inducible under certain cellular contexts. An interesting example is the adaptive nutrient stress response of the arginine/lysine transporter cat-1 mRNA in which translation of an upstream uORF alters the secondary structure of the mRNA leader sequence to allow for IRES mediated expression of the downstream cistron (187). Thus, we want to determine if ICS-7 mediated expression is affected during specific cellular contexts.

Besides conformational regulation of IRES activity, specific expression and sub-cellular location of IRES trans-activating factors (ITAFs) have been implicated in regulating the function of numerous cellular IRESs (162). While the polypyrimidine tract binding protein (PTB), hnRNPs and the autoantigen La all have roles outside the realm of translation initiation, each has been found to be associated with IRES mediated expression. ITAFs can either activate or repress expression from the IRES. A unique example is expression of the X-linked inhibitor of apoptosis protein (XIAP) in which relocalization of hnRNP A1 from the nucleus to the cytoplasm during cellular stress results in inhibition of IRES mediated translation of XIAP (93). Overall, the contexts in which IRES mediated translation initiation can occur are varied and allow for additional regulation of protein expression during important cellular events.

106

6.2 ICS-7 is a stress-inducible IRES

Alternative mechanisms of translation initiation are often active during cellular stress when canonical translation is repressed. This allows for a select subset of mRNAs to be expressed during stress. Therefore, we investigated the ability of ICS-7 to maintain translation in the presence of stress in a cell type that is permissive for HCMV infection. Bicistronic constructs containing either no

IRES (Empty), the PV-IRES or ICS-7 were nucleofected into MRC5 fibroblasts that were fed under normal growth conditions or serum stressed for 18 hr prior to nucleofection and luciferase levels were quantitated 6 hr later. ICS-7 IRES activity was increased approximately 2-fold under serum stressed conditions relative to fed conditions whereas the PV-IRES was unaffected by serum stress

(Fig. 20). These results suggest that ICS-7 may act as an inducible IRES in a context specific manner.

Figure 20. ICS-7 is stress inducible. Lysates derived from MRC5 fibroblasts grown under either fed or serum stressed conditions were nucleofected with the bicistronic constructs described in Fig. 4 and analyzed for downstream FL cistron expression. IRES activity is expressed as a ratio of FL:RL activity.

107

6.3 UL138 cistron expression from the 3.6- and 2.7-kb transcripts, but not

the 1.4-kb transcript, is stress-inducible

Next, we analyzed the impact of serum stress on UL138 cistron expression from the 3.6-, 2.7-, or 1.4-kb cDNAs where we substituted UL138 with FL as a reporter gene (Fig. 21A). MRC5 fibroblasts were cultured under fed or serum stressed conditions for 18 hr and then nucleofected with each pFL-cDNA construct. Lysates were analyzed for FL expression 6 hr later and FL expression was normalized to FL expression under fed conditions. Consistent with the ability of ICS-7 to induce expression of a downstream cistron (Fig. 20) we demonstrated enhanced FL expression from the pFL-3.6 and pFL-2.7 cDNAs. FL expression was induced 2.14- and 9.43-fold from the 3.6- and 2.7-kb cDNAs, respectively, whereas FL expression from the pFL-1.4 cDNA decreased ~2-fold in serum stressed cells (Fig. 21B). Similar to the pFL1.4-kb cDNA FL expression was decreased (~2-fold) from a monocistronic construct encoding only FL under serum starved conditions, indicating that cap-dependent translation initiation mechanisms were inhibited under stress. From these data, we conclude that FL is expressed by a canonical mechanism from the pFL1.4-kb transcript, whereas

FL is expressed by alternative IRES-like mechanisms from the pFL-3.6 and pFL-

2.7 cDNAs. Taken together these results suggest that pUL138 expression is ensured under a variety of cellular contexts and UL138 cistron expression from the pFL-3.6 and pFL-2.7 cDNAs is inducible under stress.

In parallel, we isolated RNA from cells nucleofected under serum stressed 108

or fed conditions and used qRT-PCR to confirm that the increase in FL expression measured from the pFL-3.6 and pFL-2.7 cDNAs in Fig. 21B was not due to (i) increased transcript levels or (ii) shorter, monocistronic transcripts specific to serum stress. We quantitated the 5′ and 3′ ends of transcripts generated after nucleofection using qRT-PCR as we did for the experiment shown in Fig. 14C. Our results demonstrate equivalent levels of 5′ and 3′ ends under fed or stressed conditions for each pFL-cDNA (eg. Stressed/Fed) (Fig.

21C) suggesting that increased FL expression from pFL-3.6 and pFL-2.7 is not due to an increase in FL transcripts in the stressed condition. Additionally, we measured ≤ 2 fold change in amplification of 5′ and 3′ transcript ends relative to cDNA derived from iv-RNA for all samples (eg. 135/FL), with the exception of pFL-3.6 (stressed) and pFL-1.4 (fed) (Fig. 21C). These data suggest that most transcripts are full length independent of cellular context. The two points outside the 2-fold confidence interval indicate that there are a greater number of 5′ ends relative to 3′ ends and therefore cannot account for the increased FL expression.

Importantly, equivalent 5′ and 3′ transcript ends in the pFL-2.7 fed and stressed samples indicate that the 9-fold induction of FL expression (Fig. 21B) is not due to changes in transcription.

To confirm that our cDNA nucleofection results are reflective of bona fide

IRES activity, we analyzed full-length RNAs transcribed in vitro for their ability to support expression from the downstream UL138 cistron. RNA was transcribed in vitro using linearized pFL-cDNA (Fig. 21A) as template and then equal molar 109

equivalents were either translated in RRL (Fig. 21D) or transfected into HEK-293 cells (Fig. 21E). FL expression from the pFL-2.7 and pFL-1.4 cDNAs in the RRL in vitro translation system were comparable (Fig. 21D). In contrast, FL expression from the pFL-3.6 cDNA was minimal, consistent with previous experiments. Similarly, when the pFL-2.7 and pFL-1.4 iv-RNAs were transfected into cells, similar levels of FL expression were measured (Fig. 21E). Taken together, these results suggest that the pFL-2.7- and pFL-1.4 cDNAs function efficiently as templates for translation. These data further support our interpretation that translation of the UL138 cistron from the pFL-3.6- or pFL-2.7 cDNAs requires alternative mechanisms of translation initiation and cannot be explained by the presence of monocistronic RNAs encoding the UL138 cistron. 110

Figure 21. IRES mediated expression from the UL138 cistron is induced during stress. (A) Schematic for the generation of UL138 cDNAs with FL substituted as a marker for UL138 expression. cDNA constructs for each transcript were cloned into pEF1/myc-His-B (pFL-3.6, pFL-2.7, pFL-1.4). The location of the IRES is marked. (B) MRC5 cells were cultured for 18 h in fed or stressed conditions and nucleofected with molar equivalents of pFL-cDNA constructs and analyzed for FL expression 6 h post nucleofection. FL expression during stressed conditions were normalized to expression during fed conditions. (C) qRT-PCR analysis of RNA isolated from lysates described in panel B. Inverse crossing points (1/Ct) were determined using the second derivative max method (graphed). Fold change compares amplification of the 5′ and 3′ ends of experimental cDNAs isolated from transfected cells to cDNA derived from full- 111

length iv-RNA. Transcripts with fold changes ≤ 2 compared to iv-RNA were considered to be full-length. Similarly, the 5′ and 3′ ends of transcripts derived from fed and stressed cells were compared for each cDNA construct. Values ≤ 2 indicate that equal numbers of 5′ and 3′ transcript ends were present under fed and stressed conditions. (D) Linearized pFL-3.6, pFL-2.7, pFL-1.4 cDNAs or a vector encoding FL alone (pFL) were transcribed and translated in vitro. The resulting FL was quantified. (E) Linearized pFL-2.7 and pFL-1.4 or empty vector were transcribed in vitro and RNA was transfected into HEK-293 cells. Protein lysates were analyzed 6 h later for FL expression.

6.4 IRES mediated UL138 cistron expression from

the 2.7-kb transcript is dependent on eIF4G

Our previous results demonstrated that the translation initiation factor eIF4G was necessary for the IRES activity of ICS-7 in the bicistronic luciferase vector (Fig 19). Similarly, we wanted to determine if eIF4G was involved in

UL138 cistron expression from the pFL-2.7 cDNA in both fed and serum stressed conditions. To do this, we cultured MRC5 fibroblasts under fed or serum stressed conditions for 18 hr and then nucleofected the cells with either firefly luciferase

(pFL) or with the pFL-2.7 cDNA construct either alone or in combination with a construct containing the PV 2A protease. Our results indicate that the PV 2A protease was able to inhibit cap-dependent translation initiation of the pFL construct to the same extent as serum starvation (Fig. 22A). These results confirm that PV 2A protease cleavage of eIF4G can inhibit cap-dependent translation in our system. Interestingly, co-nucleofection of pFL with PV 2A was not able to inhibit FL expression beyond the level measured after serum stress 112

alone suggesting that the activity of PV 2A under stressed conditions may be compromised.

Similarly, we analyzed the effect of PV 2A protease on UL138 cistron expression from the pFL-2.7 cDNA construct. Our results show that nucleofection of pFL-2.7 with PV 2A decreased UL138 cistron expression by approximately 2 fold independent of whether the cells were fed or stressed (Fig. 22B). These results are similar to those measured after nucleofection of pFL suggesting that eIF4G is involved in UL138 cistron expression in both cellular contexts. Taken together our data suggest that IRES mediated UL138 cistron expression from the

2.7-kb transcript is dependent on eIF4G. This data is consistent with ICS-7 function in the bicistronic luciferase construct and further supports that UL138 cistron expression is mediated from the ICS-7 element found in the 3.6- and 2.7- kb UL138 transcripts. 113

Figure 22. IRES-mediated expression of the UL138 cistron is dependent on eIF4G. (A) Firefly luciferase expression from MRC5 fibroblasts cultured in either fed or serum stressed conditions for 18 hr and then nucleofected with the pFL construct either alone (untreated) or together with a construct expressing the PV 2A protease (+ 2A Protease). (B) Similarly, firefly luciferase expression was measured from MRC5 fibroblasts that were cultured in either fed or serum stressed conditions and nucleofected with the pFL-2.7 cDNA construct either alone or with the PV-2A construct.

6.5 UL138 protein expression is stress-inducible from

the 3.6- and 2.7-kb UL138 transcripts,

but not the 1.4-kb transcript

Next, we wanted to confirm our previous results showing inducible UL138 cistron expression from the polycistronic UL138 cDNA constructs using constructs expressing pUL138 instead of FL. In order to determine the impact of 114

stress on expression of pUL133, pUL135, pUL136 and pUL138 from UL138, polycistronic transcripts we nucleofected MRC5 fibroblasts cultured under fed or starved conditions for 18 hr with the 3.6-, 2.7-, or 1.4-kb cDNAs. Lysates were analyzed 24 hr later by immunoblotting using polyclonal rabbit antisera specific to pUL133, pUL135, pUL136, or pUL138. We were able to detect pUL133, pUL135, truncated pUL136 and pUL138 expression in both fed and serum stressed conditions (Fig. 23A). Under serum-starved conditions, the expression of each protein encoded, as a first cistron was slightly decreased relative to fed conditions. Similar to previous results, we were not able to detect full-length pUL136 from the 3.6- or 2.7-kb transcripts. These results confirm the polycistronic nature of the UL138 transcripts in MRC5 fibroblasts under both cellular contexts.

Similar to our results that showed inducible UL138 cistron expression from the 3.6- and 2.7-kb cDNA constructs, we measured an induction of pUL138 expression from the 3.6- and 2.7-kb cDNA constructs under serum stressed conditions (Fig. 23B). We measured an induction of pUL138 expression from the

3.6- (57 fold) and 2.7-kb (4 fold) cDNA under serum stressed conditions compared to fed conditions. In addition, pUL138 expression from the 1.4-kb cDNA construct was decreased under serum stressed conditions (0.7 fold) suggesting that pUL138 expression from this transcript is translated canonically.

Taken together these results suggest that the UL138 transcripts are polycistronic in a cell type more relevant to HCMV infection and that pUL138 expression is 115

inducible from the 3.6- and 2.7-kb transcripts under serum stressed conditions.

Figure 23. IRES mediated expression of pUL138 from the 3.6- and 2.7-kb transcripts is induced during stress. (A) MRC5 fibroblasts were cultured in either fed or serum stressed conditions and nucleofected with cDNA constructs representing the 3.6-, 2.7- and 1.4-kb UL138 transcripts. Lysates were analyzed by immunoblotting using rabbit polyclonal antisera specific to pUL133, pUL135, pUL136 and pUL138 or a monoclonal antibody specific to tubulin as a loading control. Lysates expressing individual myc-tagged ORFs were used as positive controls. (B) The intensity of pUL138 reactive bands were quantitated using a Li- Cor Odyssey imaging system the fold change pUL138 expression over empty pUL138 expression are represented graphically.

116

6.6 Discussion and Future Directions

Our results indicate that the ICS-7 sequence functions to support stress inducible IRES mediated translation initiation of pUL138. We determined that

IRES mediated translation from ICS-7 was inducible under cellular stress in both the bicistronic luciferase construct (Fig. 20) and that this function was also supported by stress inducible UL138 cistron and protein expression from the 3.6- and 2.7-kb cDNA constructs (Fig. 21B and 23A-B).

In contrast, UL138 cistron and protein expression from the 1.4-kb transcript was inhibited (Fig. 21B and 23A-B) under serum stress suggesting that pUL138 expression from this transcript is cap-dependent. We were able to verify that

UL138 cistron expression was regulated at the level of translation as we were not able to detect smaller monocistronic transcripts that could account for the ~10 fold induction of UL138 cistron expression (Fig 21C). Further, iv-RNAs containing

ICS-7 supported expression of the downstream cistron in in vitro translation assays (Fig. 21D) and when transfected into cells (Fig. 21E) suggesting that ICS-

7 does not contain promoter or splicing activity in the context of the native UL138 transcripts. We also determined that eIF4G is necessary for IRES mediated expression of the UL138 cistron (Fig. 22). Taken together these results suggest that the UL138 transcripts coordinate the expression of multiple proteins

(pUL133, pUL135, pUL136) in addition to pUL138 and that the expression of pUL138 is ensured under multiple cellular contexts. The coordinated expression of these proteins likely has important implications for the function of pUL138 in 117

regulating HCMV latent infection.

Our data suggests that the HCMV IRES is induced under serum stress to allow for increased pUL138 expression during this cellular context from the 3.6- and 2.7-kb transcripts. Induction of the IRES may involve the action of ITAFs that, despite having other cellular roles, provide an additional level of IRES mediated regulation of downstream cistron expression (91, 162). In support of this hypothesis some mRNAs may require a ‘nuclear experience’ to acquire association with particular nuclear ITAFs required for IRES activity and some

IRESs require supplementation with cellular factors for IRES activity in vitro

(156). The most compelling evidence in support of the ‘nuclear experience’ is that after transfection of IRES containing mRNAs into the cytoplasm, IRESs that have been shown to not require association with ITAFs are more likely to have increased expression from the downstream cistron, compared to IRESs that require the action of ITAFs (42, 166). We hypothesize that there may be an association between cell type specific ITAFs that may regulate HCMV IRES mediated expression of pUL138 during the context of HCMV latency.

Our work represents the first demonstration for IRES mediated alternative translation initiation for the expression of an HCMV protein, pUL138 that is important to latency. We hypothesize that the stress inducible nature of the

HCMV IRES may be suggestive of another cellular context during which pUL138 expression is more relevant. For example, IRES mediated pUL138 expression from the 3.6- and 2.7-kb transcripts may be a way in which HCMV is able to 118

overcome the transcriptionally repressive environment of CD34+ HPCs during the establishment of HCMV latency. Taken together, the existence of alternative translation initiation mechanisms may ensure the expression of pUL138 in multiple contexts. It is likely that differential utilization of these transcripts and their encoded proteins may regulate the outcome of viral infection in a cell type or cell context dependent manner. The interaction of these proteins during HCMV latency is the focus of ongoing research. 119

CHAPTER 7

The role of type I IFN in human cytomegalovirus latency

7.1 Introduction

The type I IFN response in hematopoietic progenitor cells in response to

HCMV infection is unknown. Further any association between the type I IFN response and the establishment of HCMV latency has not been explored to date.

However, the type I IFN response has been demonstrated to contribute to latency in other herpesviruses including herpes simplex virus (HSV) Epstein Barr virus (EBV) and kaposi sarcoma-associated herpesviruses (KSHV). For example,

HSV encodes ICP0, which is a protein that antagonizes the JAK-STAT pathway to ensure viral replication during productive infection. However, during HSV latency, ICP0 levels are decreased due to LAT interference resulting in higher levels of type I IFN signaling. This is suggestive that the type I IFN response can suppress HSV replication into a latent state (58). During the onset of EBV latency, EBNA1 transcription is activated by interferon response factor 7 (IRF-7)

(191). In addition, KSHV encodes viral homologues of IRF-1 through IRF-4 that regulate the mechanisms of cellular IRF proteins and ISGs, thus modulating the type I IFN response during KSHV latency (45, 98, 143). Taken together, these data suggest that the herpesvirus family of viruses have evolved mechanisms to modulate the antiviral state of the type I interferon system to promote latency. 120

Most work regarding the type I IFN response has focused on productive

HCMV infection in MRC5 fibroblasts. For example, microarray data comparing productive HCMV infection to cells infected with a UV inactivated virus demonstrated that the number of IFN-inducible genes was significantly decreased, suggesting that viral gene expression is needed for the inhibition of the type I IFN response (18). HCMV is also able to inhibit 2’5’-oligoadenylate synthetase, and subsequent MHC class I upregulation by JAK1 and IRF9 expression in IFNα treated cells (108). Collectively, these results suggest that during a productive HCMV infection the type I IFN response is inhibited to facilitate HCMV replication. However, during HCMV latency in hematopoietic progenitor cells, viral gene expression is suppressed and the virus does not replicate. We would like to determine if viral mutants that are not able to establish a latent infection in the in vitro latency model system have differences in their ability to inhibit the type I IFN response compared to wt viruses during infection in fibroblasts and physiologically relevant CD34+ hematopoietic progenitor cells.

This information would link the ULb′ region with HCMV latency and modulation of the type I IFN response.

The mechanisms by which HCMV is able to inhibit the type I IFN response are unclear, however, the immediate early (IE) gene products IE1 and IE2 have been found to block the IFN induced JAK-STAT pathway (118) and the induction of IFNβ, respectively (172). Work in our laboratory has demonstrated that the IE proteins are not expressed during HCMV latency (124), suggesting that the type I 121

IFN response may be differentially regulated in this cell type or that additional viral factors may be involved in regulating the type I IFN response during this context of infection. The interplay between the host’s innate immune response and HCMV infection is likely to be important for the outcome of HCMV infection in physiologically relevant hematopoietic progenitor cells. We would like to understand if the viral determinants encoded within the ULb′ region contribute to

HCMV latency and also any association with a differential type I IFN response during HCMV latency.

7.2 The ULb′ ORFs UL136-UL142 are associated with decreased

resistance to the type I IFN response

in MRC5 fibroblasts

In order to determine if ULb′ sequences play a role in modulating the type

I IFN response we infected MRC5 fibroblasts with either a wt clinical strain (FIX) that contains the complete ULb′ sequence or a recombinant virus lacking ORFs

UL136-UL142 (FIX∆2) and measured the expression of type I IFN and interferon stimulated gene (ISG) effector expression using qRT-PCR. The FIX∆2 virus is not able to establish a latent infection in CD34+ hematopoietic progenitor cells, however both viruses replicate productively in MRC5 fibroblasts (51, 124). Thus, we first infected MRC5 fibroblasts with a UV inactivated counterpart to each virus to determine the type I IFN response profile as a result of virus binding and entry in the absence of viral gene expression. Our results determined UV inactivated 122

FIX had an earlier onset of IFNβ gene expression compared to UV inactivated

FIXΔ2 over a time course (Fig. 24A). Similarly, viperin and ISG54K gene expression were enhanced after infection with UV inactivated FIX compared to

UV inactivated FIXΔ2 (Fig. 24B-C). These results suggest that the components of the FIX virion are more likely to induce a type I IFN response in MRC5 fibroblasts than the FIXΔ2 virion, since viral gene expression was inhibited by UV inactivation of these viruses. However, these results may alternatively reflect differences in virus preparation, for example, as repeated experiments showed a similar trend, more work needs to be done to address the significance of these conclusions.

We also wanted to compare the ability of FIX and FIXΔ2 viruses to inhibit the type I IFN response after infection. To do this we again measured IFNβ, viperin and ISG54K gene expression over a time course using qRT-PCR (Fig.

24D-F). Whereas expression of IFNβ, viperin and ISG54K was decreased after infection with both viruses compared to infection with the UV inactivated viruses,

IFNβ gene expression at the 6 hr and 20 hr time points was increased in FIXwt infection. Similarly, viperin gene expression was elevated in FIXwt infected cells compared to FIXΔ2 infection, suggesting that the FIXwt virus was not able to inhibit the type I IFN response as effectively as FIXΔ2 in MRC5 fibroblasts.

Interestingly, ISG54K gene expression was similar in both infections. Taken together, these results suggest that the UL136-UL142 region of the ULb′ confers a slight decrease in resistance to the type I IFN response in MRC5 fibroblasts. 123

Figure 24. ULb′ ORFs UL136-UL142 are associated with decreased resistance to the type I IFN response in MRC5 fibroblasts. (A-C) Total RNA was isolated from MRC5 fibroblasts infected with UV inactivated FIX or FIX∆2 (MOI-2) over a time course and analyzed using qRT-PCR using primers specific for IFNβ (A) and the interferon stimulated genes (ISGs) viperin (B) and ISG54K (C). (D-F) Total RNA was isolated from MRC5 fibroblasts infected with FIX or FIX∆2 (MOI-2) over a time course and analyzed using qRT-PCR using primers specific for IFNβ (D) and ISGs viperin (E) and ISG54K (F)

124

7.3 The ULb′ ORFs UL136-UL142 are associated with decreased

resistance to IFNα challenge

Next we wanted to test the replication ability of FIX and FIXΔ2 in the presence of IFNα. This analysis challenged the ability of virus to not only inhibit the type I IFN in response to viral infection but also continuous exogenous IFN stimulation. Increasing amounts of recombinant IFNα were added to MRC5 fibroblast cultured cells and viral replication was measured by calculating the percentage of wells containing viral plaques as measured by GFP+ MRC5 fibroblasts over a time course. Consistent with our previous results, we determined that the FIXΔ2 virus was more resistant to the effects of type I IFN as measured by increased viral replication in the presence of recombinant IFNα

(Fig. 25A-C). This result was more pronounced at the 3- and 6 dpi time points, however, by 8 dpi both FIX and FIXΔ2 had similar number of GFP+ wells indicating that although FIX is less resistant to type I IFN, replication of this virus at later times of infection is not inhibited in MRC5 fibroblasts. Taken together, our data suggests that the ULb′ ORFs UL136-UL142 are dispensable for growth in

MRC5 fibroblasts and that these genes may confer a slight disadvantage for viral replication at early times of infection. 125

126

Figure 25. ULb′ ORFs UL136-UL142 are associated with decreased resistance to IFN challenge. (A-C) MRC5 fibroblasts were infected in a 96 well plate with FIX or FIX∆2 in the presence of increasing concentrations (units) of recombinant IFNα at an MOI of 0.01. The percentage of GFP+ wells, indicating HCMV infection, at each concentration of IFNα, was determined at 3- (A), 6- (B) and 8 dpi (C).

7.4 The ULb′ ORFs UL136-UL142 are associated with increased

resistance to the type I IFN response in CD34+ HPCs

To determine if the ULb′ region, necessary for the establishment of HCMV latency, is also associated with a decrease in resistance to the type I IFN response in CD34+ HPCs as determined in MRC5 fibroblasts we first infected with UV inactivated FIX and FIXΔ2 and measured ISG gene expression using qRT-PCR over a time course. Our results showed a slight increase in viperin,

ISG54K and IRF7 gene expression with UV inactivated FIX compared to UV inactivated FIXΔ2 at 10-20 hpi (Fig. 26A-C). It should be noted that the ISG gene expression in response to UV inactivated HCMV infection is significantly decreased in this cell type (~20 fold) compared to MRC5 fibroblasts (~200,000-

20,000 fold). These results suggest that the virion components of FIX may slightly increase the type I IFN response compared to the virion components of

FIXΔ2. However, because the overall ISG gene expression was decreased in this cell type it is difficult to make any firm conclusions in determining the role of the

ULb′ UL136-UL142 region in relation to the type I IFN response in CD34+ HPCs. 127

We also wanted to compare the ability of FIX and FIXΔ2 viruses to inhibit the type I IFN response after infection in CD34+ HPCs. To do this we infected

CD34+ HPCs with FIX and FIXΔ2 and measured ISG gene expression at 10- and

24-hpi using qRT-PCR. Given the low ISG expression from UV inactivated virus, we were surprised that viperin gene expression was increased after infection with both FIX and FIXΔ2 (Fig. 26D). These results suggest that viperin gene expression may not limit viral replication in CD34+ HPCs. In addition, FIXΔ2 showed a slight increase in viperin and IRF7 gene expression compared to FIX at both early and later times of infection (Fig 26D and F), although the significance of this data questionable as more experiments need to be done to repeat this result. However, ISG54K gene expression was elevated in FIXΔ2 infection compared to FIX especially at the 24 hpi time point (Fig. 26E). Taken together these results suggest that the FIX ULb′ ORFs UL136-UL142 have a slight resistance to the type I IFN response in CD34+ HPCs as determined by decreased ISG gene expression compared to FIXΔ2. These results suggest that the UL136-UL142 sequences are advantageous during infection in CD34+ HPCs however we would expect that an increase in ISG gene expression from FIXΔ2 would be associated with a decrease in viral replication in this cell type. However, our previous results using the in vitro latency model system, suggest that the

FIXΔ2 replicates productively in CD34+ HPCs. In addition, our results in MRC5 fibroblasts (Fig. 24 and 25) suggest that FIX has decreased resistance to the type I IFN response, which contradicts our results in CD34+ HPCs (Fig. 26). The 128

significance of increased resistance of FIX compared to FIXΔ2 in CD34+ HPCs is currently under investigation and likely represents a cell type specific factor that is active during HCMV latency.

Figure 26. ULb′ ORFs UL136-UL142 are associated with increased resistance to the type I IFN response in CD34+ HPCs. (A-C) Total RNA was isolated from CD34+ HPCs infected with UV inactivated FIX or FIX∆2 (MOI-2) 129

over a time course and analyzed using qRT-PCR using primers specific for interferon stimulated genes (ISGs) viperin (A) ISG54K (B) and IRF7 (C). (D-F) Total RNA was isolated from CD34+ HPCs infected with FIX or FIX∆2 (MOI-2) over a time course and analyzed using qRT-PCR using primers specific for and the ISGs viperin (D), ISG54K (E) and IRF7 (F).

7.5 ORFs UL136-UL142 are necessary for FIX to inhibit ISG expression in

CD34+ HPCs

We wanted to determine the extent that FIX and FIXΔ2 were able to inhibit the type I IFN response in CD34+ HPCs. In order to do this we infected CD34+

HPCs in the presence of neutralizing antibodies against IFNαβ and measured

ISG gene expression using qRT-PCR with the rationale that the effect of the neutralizing antibodies would be an indirect measure of the amount of type I IFN produced during infection. For example, if the addition of neutralizing antibodies did not have an inhibitory effect on ISG expression then we could conclude that the virus was able to inhibit IFN production. In contrast, if the addition of neutralizing antibodies were able to inhibit ISG gene expression, it would suggest that more IFN was produced during the course of infection than could be neutralized. Our results indicate that the addition of neutralizing antibodies against IFNα during infection with FIX and FIXΔ2 did not affect viperin, ISG54K and IRF7 gene expression in FIX infected cells however infection with FIXΔ2 had greater inhibition of viperin, ISG54K and IRF gene expression (Fig. 27A-C).

These results suggest that FIX infection results in less type I IFN production, as a 130

result of viral control, compared to FIXΔ2 which corroborates our data in Figure

26 where we show an increase in ISG gene expression during FIXΔ2 infection.

Collectively, our results indicate that the ULb′ ORFs UL136-UL142 are associated with slightly increased resistance against the effects of type I IFN mediated inhibition of viral replication. However, these results are do not follow with our original hypothesis that associates a decrease in viral replication with an inability to modulate the type I IFN response. However, future work will elucidate the significance of the increased resistance to type I IFN of FIX infection during

HCMV latency.

131

Figure 27. ORFs UL136-UL142 are necessary for FIX to inhibit ISG expression in CD34+ HPCs. (A-C) CD34+ cells were infected with FIX or FIX∆2 at an MOI of 2 in the presence of neutralizing antibody to IFNα and total RNA was isolated at 24 hpi and analyzed for ISG expression using specific primers for viperin (A), ISG54K (B) and IRF7 (C). The degree of IFN neutralization is represented as a fold change of ISG expression of antibody treated over untreated ISG expression.

132

7.6 Discussion and Future Directions

Our results represent a considerable amount of preliminary data that will be used to understand the relationship of the type I IFN response and the establishment of HCMV latency. We were able to determine that depending on the cellular context the ULb′ ORFs UL136-UL142 either provided slight sensitivity or slight resistance against the type I IFN response in MRC5 fibroblasts or

CD34+ HPCs, respectively. This conclusion was based on increased IFNβ, viperin and ISG54K gene expression measured after infection in MRC5 fibroblasts with FIX in both the UV inactivated and native form compared to infection with FIXΔ2 (Fig. 24A-F). These results suggest that the FIXΔ2 virus is better able to inhibit the type I IFN response in MRC5 fibroblasts. Our conclusion is supported by increased replication after challenge with recombinant IFNα in

FIXΔ2 infection in MRC5 fibroblasts. In contrast, our results in CD34+ HPCs associate the ULb′ ORFs UL136-UL142 with increased resistance to the type I

IFN response as measured by decreased ISG gene expression (Fig. 26D-F) and futile effects of type I IFN neutralizing antibodies (Fig. 27A-C) after infection with

FIX compared to FIXΔ2. Collectively, these results do not support an association between the ability of the type I IFN response to inhibit viral replication and decreased viral replication during HCMV latency. However, we did identify cell type specific modulation of the type I IFN response by the ULb′ ORFs UL136-

UL142 that may be associated with alternative roles during the establishment of

HCMV latency. 133

Previous work that was done to develop an HCMV vaccine relied on extensive serial passage in MRC5 fibroblasts to attenuate the virus, and consequently resulted in HCMV strains that lacked the ULb′ region (113). The resulting laboratory strains of HCMV replicated better in culture and although these strains did not succeed as vaccine candidates, they have served as a negative control for the establishment of HCMV latency, as the ULb′ region is necessary for HCMV latency in CD34+ HPCs (51). Given our results, it is possible that the type I IFN response may play a role in maintenance of the ULb′ region, allowing for better fitness, during the establishment of HCMV latency in the context of CD34+ HPCs. In contrast the ULb′ region is associated with decreased replication, and possibly fitness, in a productive cellular context in

MRC5 fibroblasts and thus the ULb′ region confers a disadvantage in this cell type, possibly due to the increased type I IFN response after infection with clinical strains of HCMV. We were not able to detect significant differences in the ability of FIX and FIXΔ2 to replicate in MRC5 cells (124) however, challenge with recombinant IFNα suggested that the FIX ULb′ ORFs UL136-UL142 were associated with decreased replication ability. It is likely that this disadvantage would be selectively inhibited during serial passage of the virus in MRC5 fibroblasts.

The type I IFN response in CD34+ cells was slightly reduced compared to

MRC5 fibroblasts as measured by interferon related gene expression during infection with UV inactivated virus (Fig. 24A-C and Fig. 26A-C). Recent work 134

suggests that HPCs have functional toll-like receptors and the ability to elicit an innate immune response (115) however it is not known how the type I IFN response in CD34+ HPCs impacts HCMV infection. Our work suggests that the

ULb′ ORFs UL136-UL142 can possibly inhibit the production of type I IFN (Fig.

27A-C) and also ISG gene expression (Fig. 26D-F) during infection in CD34+

HPCs. Considering that we were not able to associate an increase in viral replication from FIX infected CD34+ cells compared to FIXΔ2, it is possible that the type I IFN response may be inhibited for other means during this context of infection.

Further work needs to be done to advance understanding of the contribution of the ULb′ ORFs UL136-UL142 to HCMV infection with respect to the type I IFN response. Infection with UV inactivated FIX resulted in increased

IFNβ, viperin and ISG54K gene expression over time (Fig. 24A-C) suggesting that the FIX virion is fundamentally different than that of the FIXΔ2 virus. We have associated the UL136 ORF with IRES mediated expression of pUL138 and preliminary results show that pUL136 localizes to the golgi apparatus (data not shown) similar to pUL138, suggesting a role for pUL136 in HCMV latency. It is unclear if the UL139 and UL140 ORFs encode proteins, however the UL139 sequence has been found to be polymorphic (101), whereas the UL140 sequence is conserved among clinical strains of HCMV (136). UL141 and UL142 encode proteins during HCMV infection and are involved in immune evasion of

NK cells (7, 24, 133, 179, 180). Our results support a role for UL141 and UL142 135

with the innate immune response during HCMV infection. Further work will elucidate which ORFs are involved in the context specific modulation of the type I

IFN response during HCMV infection in MRC5 fibroblasts and CD34+ HPCs.

The major limitation of this work included variation in ISG expression and

IFN resistance among FIX isolates. This is likely due to the virus preparation in which slight differences in virus propagation may affect the extent the type I IFN response after infection. For example, we had to rely on numerous biological replicates to have even slight confidence in our results. Variation could have arisen from inadequate qRT-PCR software analysis, differences in the virus preparations or limited template in CD34+ HPC analysis. Advancements in our laboratory will likely eliminate some of these concerns for future investigations.

Future work to understand the role of the type I IFN response in contributing to HCMV latency will likely include a thorough analysis of the exact components of the HCMV virion that contribute to the type I IFN response or inhibition thereof. In addition, we would like to determine if type I IFN neutralizing antibodies can abrogate the ability of clinical strains of HCMV to establish a latent infection in CD34+ HPCs using the in vitro latency model system. This body of work is the first account of the type I IFN response in CD34+ HPCs from

HCMV infection and the differential regulation of the type I IFN response associated with the ULb′ ORFs UL136-UL142. Our results provide a framework for future investigations into the type I IFN response with respect to HCMV latency. 136

CHAPTER 8

Discussion

Mechanisms involved in the establishment of HCMV latency are poorly understood, however, work in our laboratory has demonstrated the ULb′ encoded protein, pUL138, as the first viral determinant to function in the establishment of

HCMV latency (51, 124). The work described here characterizes the transcripts that encode pUL138. pUL138 is encoded on three polycistronic transcripts of 3.6-

, 2.7- and 1.4-kb in length. The ULb′ proteins pUL133, pUL135 and truncated pUL136, are expressed on the 3.6-, 2.7- and 1.4-kb transcripts, respectively, in addition to pUL138. We also the first to describe the novel coding potential of the

ULb′ proteins, pUL133, pUL135, and pUL136 that may have important implications for HCMV latency. In addition, our work represents the first demonstration of IRES mediated alternative translation initiation in the expression of pUL138, a latency factor in HCMV. We demonstrate that pUL138 expression is inducible from the IRES on the 3.6- and 2.7-kb transcripts under conditions of cellular stress. Differential utilization of the UL138 transcripts and their respective encoded proteins may regulate the outcome of viral infection in a cell type or cell context dependent manner. The interaction of these proteins during HCMV latency is the focus of ongoing research. In addition, we have generated preliminary data regarding the type I IFN response directed toward 137

HCMV during productive infection in MRC5 fibroblasts and during the establishment of HCMV latency in CD34+ HPCs.

The transcripts encoding pUL138 were identified in both productive infection and during the establishment of HCMV latency suggesting that UL138 transcript expression alone is not responsible for the establishment of HCMV latency. This is in accordance with latency-associated gene expression in other herpesviruses.

For example, HSV LATs are expressed during both the lytic and latent infections, but a role for the LATs during productive infection has not been described.

Similarly, UL138 is dispensable for viral replication in fibroblasts (124). Our results do not exclude a role of the UL138 transcripts for the establishment of

HCMV latency. Consequently, we assessed the contribution of either the UL138 transcripts or pUL138 to the establishment of HCMV latency by analyzing the pUL138 expression and UL138 transcript stability after infection with FIXwt compared to infection with the recombinant viruses FIX∆UL138 and

FIX∆UL138STOP. Our results clearly demonstrate that although pUL138 is expressed during both productive and latent HCMV infection, pUL138 is required for the establishment of HCMV latency in CD34+ HPCs suggesting that the outcome of infection likely depends on the context-dependent balance of competing viral and cellular factors that either promote productive viral replication or latency.

Many viruses have multiple promoters to optimize gene expression and

HCMV is no exception. Interestingly, transcripts that encode pp28, a viral 138

tegument protein, are contained in a family of 3′ co-terminal transcripts that have the potential to encode 8 additional proteins (181). 3′ co-terminal transcript families have been described for other HCMV transcripts including the transcripts encoding the gp47 and gp52 family, US18, US19 and US20 family and pp65 and pp71 family of tegument proteins (52, 56, 146). Transcription through the gene families is variable in transcript abundance during different times of infection and transcription start sites in these families tend to correspond to the start of the 5′ cistrons. These results likely represent a combination of unique transcription regulation, the potential post-transcriptional gene expression, and possible regulation through polyadenylation signals, of which there are very few in the

HCMV genome (109). These proteins, unlike pUL138, have not been associated with HCMV latency and the coding potential of upstream ORFs has not been fully determined. It is tempting to speculate that alternative mechanisms of translation initiation that may allow for the 3′ cistron expression from these transcripts, similar to pUL138 expression. Interestingly, polycistronic transcripts generated from latency-associated regions of herpesviruses are not uncommon, and may represent a conserved strategy for the regulation of protein expression during latency (16, 29, 54, 72, 97, 169, 170). For example, KSHV encodes a cluster of genes, LANA (ORF73), v-cyclin (ORF72) and v-flip (ORF71) that are expressed on 3′ co-terminal transcripts that are expressed during KSHV latency (4, 119).

Similar to pUL138 expression, v-flip protein expression is mediated by an

IRES element that overlaps the upstream v-cyclin ORF (16, 54, 97). The HCMV 139

IRES we describe overlaps the UL136 ORF upstream of UL138. The overlap between these ORFs and IRES sequences is not surprising given the limited intergenic space in the herpesviral genomes and the need to maximize viral genome coding capacity (112). The Epstein-Barr virus (EBV) nuclear antigen-1

(EBNA1), the only nuclear protein expressed during all types of EBV latency, is also expressed via an IRES which is present in the U leader exon inherent to all

EBNA1 mRNA despite which promoter is utilized (72). Further, IRES-mediated translation of immediate-early proteins RLORF9 (170) and pp14 (169) in Marek’s disease virus and the expression of the immune evasion ORF, MK3 in murine gammaherpesvirus 68 are more examples of IRES-mediated protein expression related to herpesvirus latency (29). Collectively, these findings demonstrate an attractive mechanism by which herpesviruses tightly coordinate the expression of multiple proteins required for complex host-virus interactions that impact the outcome of infection.

Consistent with this speculation, it is tempting to suggest a role for pUL133, pUL135 and pUL136 in the establishment of HCMV latency. Preliminary results show that each of these proteins, in addition to pUL138 localize to the golgi apparatus during productive infection and that these proteins may form a complex (data not shown). Interestingly, in other models of herpesvirus infection, latency associated factors map to specific regions of the genome that also encode factors for viral reactivation (82). For example, during HSV infection the

LATs, expressed during HSV latency, share sequence with viral factors that 140

support productive HSV infection such as ICP0. The 8.3-kb LAT RNA is antisense to the ICP0 mRNA, which is thought to regulate ICP0 gene expression

(26). The role of pUL133, pUL135 and pUL136 on the establishment of HCMV latency is currently being investigated.

Of particular interest is the differential expression of pUL136. Our data indicate that pUL136 is expressed in two forms detectable after later times in infection of MRC5 fibroblasts, although we could not detect expression of the full length pUL136 after transfection with either the 3.6- or 2.7-kb cDNAs in either

HEK-293 or MRC5 fibroblasts. However, we did detect the truncated version of pUL136 after transfection with the 1.4-kb cDNA, suggesting that infection-specific factors may be required for the expression of full-length pUL136. It is tempting to speculate a role for the two forms of pUL136 in the expression of other ULb′ proteins or during the establishment of HCMV latency.

Similar examples of IRES mediated regulation of alternative forms of a protein suggest that pUL136 may play a regulatory role. For example, lymphoid enhancer factor-1 (LEF-1) is translated in two forms, a full-length protein translated via IRES-mediated cap-independent mechanisms and a truncated form translated from a shorter transcript by canonical cap-dependent mechanisms (78). The truncated form of LEF-1 inhibits the growth-promoting activity of full-length LEF-1 (70). Similar to the two forms of pUL136 expression,

US3 during HSV infection is expressed in both a full length and a truncated version (US3.5) from two co-linear and 3′ co-terminal transcripts (128, 144). The 141

larger transcript codes for full length US3 that functions as a protein kinase to phosphorylate HDAC1 to prevent apoptosis, however, while US3.5 can still phosphorylate HDAC1 it cannot prevent apoptosis suggesting an interesting regulatory dynamic between the two forms of the protein (128, 130). Interestingly, both forms of US3 are associated with promoting productive infection through blocking histone deacetylation independent of ICP0 function (129). Ongoing work is focused on understanding the role of each form of pUL136 in regulating the outcome of HCMV infection.

Many IRES elements are identified by elimination of other means by which downstream cistron expression could occur. For example, thorough analysis to eliminate promoter and/or splicing events is a key component of IRES research.

With this in mind, we have devoted a substantial amount of work to verifying the polycistronic nature of the 3.6-, 2.7-, and 1.4-kb transcripts. Although, the 1.4-kb transcript is most likely not translated via an IRES, it was influential in our development of transcript length analysis, in which we determined that all of the

UL138 transcripts were indeed polycistronic. In addition our in vitro work that showed comparable IRES activity from sequences containing ICS-7 in the bicistronic reporter as well as from the UL138 transcripts themselves, was further evidence against the possibility that promoter and/or splicing events were responsible for the HCMV IRES activity. Furthermore, the ICS-7 sequence was not able to promote expression of a downstream cistron in a promoterless construct, suggesting that the HCMV IRES does not have promoter activity. It 142

should be noted that our analysis did not include characterization of the promoters that drive the expression of the UL138 transcripts, however, our qRT-

PCR analysis suggests that cis-acting promoter elements are not involved in the expression of the shorter 2.7- and 1.4-kb transcripts, as we would have detected them after transfection with the 3.6-kb cDNA construct. Taken together, our results suggest that the ICS-7 region is an IRES.

The activity of cellular IRESs is functionally less defined compared to the activity associated with poliovirus and encephalomyocarditis virus (EMCV)

IRESs. Cellular IRESs have been identified in a number of transcripts and have been associated with protein expression during contexts of cellular stress including mitosis, cellular differentiation and apoptosis (63, 102, 161). Our results have determined that an IRES is responsible for the translation of pUL138 under serum stressed conditions from the 3.6- and 2.7-kb transcripts. One of the defining characteristics of IRESs was the ability to mediate cap-independent translation of a cistron. While this unique aspect of alternative translation initiation holds true for RNA viruses that do not have cap structures or mRNA processing similar to cellular transcripts, the definition of cap-independence is unclear for cellular IRESs. This distinction is confounded by the fact that many

IRESs, including many DNA viral IRESs, require a ‘nuclear experience’ for optimal expression from the IRES (42, 156). The exact reason as to why some

IRES elements require nuclear processing is unclear, however, recent work 143

suggests that ITAF association with the IRES in the nucleus, likely plays a critical role in IRES mediated expression (156).

An intriguing finding of our studies is the stress-inducible nature of the

HCMV IRES. Alternative mechanisms of translation of cellular transcripts are typically active during cellular stress (40, 48, 60, 69) mitosis (33, 135, 154), and apoptosis (68, 69, 92), when canonical cap-dependent translation initiation is inhibited (48). Our studies demonstrate that the HCMV IRES mediates expression of the downstream cistron in the context of bicistronic luciferase construct as well as from the native UL138 transcripts. Additionally, pUL138 from the HCMV IRES from the 3.6- and 2.7-kb transcripts is enhanced under stress conditions, whereas pUL138 expression from the 1.4-kb transcript was inhibited within the same cellular context. This result is interesting given the fact that the

1.4-kb transcript encoded a truncated form of pUL136 in addition to pUL138, indicating that this transcript is also polycistronic. Our data suggests that translation of pUL138 from the 1.4-kb transcript occurs through canonical mechanisms that are inhibited by stress. This may require translation reinitiation or readthrough. The fact that the UL138 cistron is optimally expressed from different transcripts depending on the cellular context implies that the multiple polycistronic transcripts may represent a viral strategy to ensure the expression of pUL138.

We determined that IRES mediated pUL138 expression was inducible under serum stressed conditions suggesting that the HCMV IRES is cap- 144

independent, as serum starvation is an effective means to inhibit global cellular cap-dependent translation. In accordance with this statement, our results determined that expression of a monocistronic message (FL) was inhibited during serum starvation. This inhibition was comparable to cap-dependent translation inhibition by co-transfection with PV-2A protease, suggesting that cap- dependent translation is inhibited in both contexts. Despite the means of cap- dependent inhibition of translation initiation, we downstream cistron expression was still induced. However, it should be noted that the means by which cap- dependent translation is being inhibited in each circumstance are likely different.

Serum starvation promotes the phosphorylation of eIF2α by a family of kinases that are activated under stress that blocks translation initiation by inhibiting the

met eIF2/GTP/tRNAi ternary complex (188). This is in contrast to the action of the

PV 2A protease that inhibits cap-dependent translation through the cleavage of eIF4G that is needed to bridge the 5′ cap and eIF4E during translation initiation

(155). Our results determined that downstream cistron expression from the

HCMV IRES, although inducible under serum starved conditions, was still dependent on eIF4G as determined through co-transfection with PV-2A protease, suggesting a role of eIF4G in IRES mediated expression in both normal and serum starved conditions. Its possible that off target effects of PV-2A such as disruption of the nuclear pore complex, may have also had an effect during normal and serum stressed conditions (21). A time course of expression would highlight the exact effects of PV 2A that are active in this system. Collectively, 145

our data suggest using in vitro transcribed RNA from either the bicistronic vector and the UL138 transcripts suggest that pUL138 expression from the HCMV IRES is cap-independent and this IRES is more similar to the type IV IRESs that requires supplementation with cellular translation initiation factors for full expression from the IRES.

Cellular IRES sequences are often activated under conditions of stress due to interactions with ITAFs (65, 92, 100, 162). ITAFs provide an additional level of regulation for IRES- mediated expression through specific ITAF relocalization

(91, 162) or cell type specificity (36, 106, 126). The context specificity of the

HCMV IRES is consistent with the hypothesis that the HCMV IRES requires association with a stress-associated ITAF. Preliminary data suggests that in vitro derived RNA of bicistronic constructs containing ICS-7 show decreased IRES mediated FL expression compared to the bicistronic vector containing the poliovirus KSHV IRES (data not shown), however, IRES activity measured from

ICS sequences corresponded to IRES activity as determined in the initial ICS

IRES screen suggesting IRES mediated expression in vitro is inefficient and may require an ITAF for optimal expression. In support of this conclusion, we noticed a band of higher molecular mass than FL on gels of in vitro translated lysates that was only present from constructs that contained the ICS-7 sequence and showed positive activity after the initial ICS screen (data not shown). The identity of this band is unknown, however, it is possible that it represents a product of inefficient internal translation initiation. 146

Much of our work suggests that IRES dependent translation of pUL138 may require ITAFs for optimal expression from the IRES. Early work demonstrated that poliovirus IRES required cellular factors for full IRES activity using in vitro rabbit reticulolysate extracts and was restored after supplementation with Hela

S10 extract (39). Consequently, Hela S10 extract was added to ICS-7 in-vitro translation reactions, however, FL activity was not affected (data not shown).

These results suggest a yet unknown ITAF may be needed for optimal alternative translation initiation from ICS-7. Consequently we wanted to determine if HCMV

IRES functionality could be enhanced by supplementation with the S10 extract, however, we did not measure any change in expression from the IRES. A similar result was obtained after in vitro translation of RNA containing the murine gammaherpesvirus 68, MK3 coding sequence (29). In contrast, the KSHV IRES shows high levels of expression in rabbit reticulolysate systems, indicating that not all of the herpesvirus IRESs have the same mechanism of downstream cistron expression.

Polypyrimidine tract binding protein (PTB) is an attractive ITAF that may regulate the HCMV IRES function, as a PTB binding consensus sequence exists within the IRES. In addition, we identified two stable hairpins within the IRES sequence and potential pseudoknot formation that may function to recruit PTB or other ITAFs for IRES functionality. Interestingly, recent work demonstrates that

PTB localization is temporally regulated during HCMV infection (46) and that

HCMV major immediate early gene expression is inhibited by PTB through 147

regulating alternative splicing of the MIEP IE products (34). Taken together these data suggest a possible role for PTB in the establishment of HCMV latency through promoting IRES mediated translation of pUL138 and concurrent inhibition of the MIEP. Further work needs to be done to determine the validity of this hypothesis.

The cellular stress response has evolved over time to preserve cellular survival during conditions of nutrient depletion, hypoxia and ER stress. Similarly, viral infection elicits a cellular stress response that many viruses have evolved mechanisms to manipulate in order to ensure viral replication. The most common mechanism by which the cell responds to cellular stress is by inhibiting protein translation (67). Inhibition of this energy intensive process decreases the cellular demand for ATP/GTP and lower levels of ER protein processing further conserves cellular energy (5). In addition, if the cell is infected with a virus that relies on classical cap-dependent translation mechanisms, translation of viral proteins is also inhibited. This allows the cell to combat viral infection until the innate and adaptive immune system is utilized to clear the infection. The benefit of inhibiting cellular translation in response to stress is that this process is fast and reversible through the alteration of the phosporylation state of translation regulatory proteins (5). If cellular stress is not overcome the cell induces apoptosis. Recent data suggests that HCMV may be able to manipulate stress responses to maintain viral replication. For example, during productive viral infection, recent work suggests that HCMV maintains cap-dependent translation 148

initiation by modulation of the phosphatidylinositol-3′ kinase (PI3K)-Akt-tuberous sclerosis complex (TSC)-mammalian target of rapamycin (mTOR) signaling pathway (5). The mechanism by which HCMV is able to overcome stress induced translation inhibition is mainly through the action of pUL38 that inhibits cellular stress by antagonizing the TSC complex (111) and targeting ATF4 expression and JNK activation to inhibit viral induction of ER stress (185). It is interesting to speculate that the stress inducible nature of IRES mediated pUL138 expression from the 3.6- and 2.7-kb transcripts reflects a mechanism by which HCMV uses cellular stress, upon viral infection, for the expression of pUL138. Furthermore, pUL138 expression from the 1.4-kb transcript may be initiated once HCMV has restored cap-dependent translation through the actions of pUL38, since we show that pUL138 expression from the 1.4-kb transcript is decreased in cellular contexts when cap-dependent translation mechanisms are inhibited.

UL138 transcript abundance may influence the means by which pUL138 expression occurs during infection. This may be regulated in either a temporal or cellular context. For example, if the 3.6- and 2.7-kb transcripts were in higher abundance at any time during infection, or in either cell type, we would assume that IRES mediated translation of pUL138 accounts for the majority of pUL138 expression. Similarly, varied transcript abundance of monocistronic and polycistronic transcripts encoding v-flip during KSHV latency, suggest that the

IRES is active in a context dependent manner (54). It is reasonable to assume that the transcript profile could change temporally or in response to a stress to 149

allow for differential pUL138 expression from the three UL138 transcripts.

Experiments are ongoing to further elucidate if UL138 transcript abundance can help determine IRES usage during the context of infection. Preliminary work has determined that we can extend the qRT-PCR assay we developed to determine transcript length to further determine the relative transcript abundance after infection. We can do this by calculating the ratios of the unique 5′ targets to the common 3′ target for each transcript. This analysis will be essential for transcript analysis from infection of CD34+ cells where template is limiting.

Overall, our work represents the first demonstration for alternative translation initiation in the expression of an HCMV protein important to latency.

Further, our work demonstrates the novel coding potential of three additional

ULb′ proteins pUL133, pUL135, and pUL136. The existence of multiple polycistronic transcripts and the possibility of multiple mechanisms to ensure the expression of pUL138 have important implications for latency. Differential utilization of these transcripts and their encoded proteins may regulate the outcome of viral infection in a cell type or cell context dependent manner. The interaction of these proteins during HCMV latency is the focus of ongoing research.

We have also focused attention to the type I IFN response generated against HCMV infection in both MRC5 fibroblasts and CD34+ HPCs in order to determine if the type I IFN response is associated with HCMV latency. Although, our results are preliminary, we have determined that the ULb′ ORFs UL136- 150

UL142, that are necessary for the establishment of HCMV latency, are associated with a slight increase in sensitivity to the type I IFN response after productive infection of MRC5 fibroblasts, compared to recombinant virus lacking this region. However, infection of CD34+ HPCs, a cell type that supports HCMV latency, supported an alternative role for the ORFs UL136-UL142 that conferred slight resistance to the type I IFN response. Collectively, these results do not necessarily support an association between the ability of the type I IFN response to inhibit viral replication and decreased viral replication during HCMV latency.

However, we have not ruled out other associations between the UL136-UL142 region and other aspects of HCMV biology, such as viral fitness in CD34+ HPCs, the significance of which has yet to be determined.

Examples of associations between decreased viral replication and the type I IFN response during herpesvirus latency have been determined for HSV,

EBV and KSHV latent infections. However, our data suggests that in a cellular context that can support a latent infection, decreased viral replication was associated with an increase in the ability to inhibit the type I IFN response.

Although preliminary, these results suggest that inhibition of the type I IFN response may play a role in other mechanisms that contribute to the establishment of HCMV latency besides viral replication. This is in contrast to constitutive expression of virally encoded IRF3 (vIRF3) during KSHV latent infection that supports IFNα production (98, 143). Given these discrepancies we 151

would like to further explore the role of the ULb′ region in the establishment of

HCMV latency.

We determined ISG expression, as analyzed by qRT-PCR, as a measure of the type I IFN response generated after infection with FIX and FIXΔ2. We were expecting that if the ULb′ UL136-UL142 ORFs that are necessary for HCMV latency were also involved in the type I IFN response, then we would measure differential ISG expression between the two viruses. We chose to analyze viperin

(virus inhibitory protein, endoplasmic reticulum associated, interferon-γ-inducible) expression as it is one of the few interferon inducible cellular proteins found to be associated with decreased HCMV replication and it localizes to the ER/Golgi apparatus, similar to pUL138 that is localizes to the Golgi (28, 124). Similarly, we were interested in IRF7 expression after HCMV infection because during the establishment of EBV latency IRF7 is involved in expression of the viral latency factor, EBNA1 (191). We also monitored expression of ISG54K because previous work suggested that it is an ISG that is significantly inhibited upon HCMV infection (18). Unfortunately, we were not able to significantly associate expression of these ISGs with infection of MRC5 fibroblasts or CD34+ cells after infection with FIX and FIXΔ2, although notable differences in ISG expression included increased viperin expression in MRC5 fibroblasts during FIX infection and increased ISG54K expression after infection of CD34+ cells with FIXΔ2.

Interestingly, recent work has associated ISG54K function to inhibition of eIF3 during EBV infection (55). This connection, likely corresponds to increased 152

viral replication; however, it is interesting that translation of viral proteins is not inhibited by this function of ISG54K (REF). These data suggest that EBV may be using the type I IFN response to alter translation initiation which may hold greater implications in the possible role of the HCMV IRES during the establishment of

HCMV latency. Other connections between IRES mediated translation initiation and the type I IFN response include IRES mediated expression of IRF2, which is a negative regulator of IFN induced transcription that is involved in repression of the IFN stimulated genes (37). Alternatively, IFNα was found to inhibit IRES mediated expression during hepatitis C virus infection (61) and IFNα induced

MyD88 (myeloid differentiation factor 88) activation inhibited expression of PTB during hepatitis B infection (94). Taken together these data suggest that a complex interplay between alterations of translation initiation and subsequent cellular and viral IRES expression that may contribute to the outcome of infection.

This work provides an initial characterization of the coding potential of the

UL138 transcripts resulting in the identification of three novel HCMV proteins pUL133, pUL135 and pUL136. It also includes the first identification of IRES mediated pUL138 expression during cellular stress that may play a role in HCMV latency. Further, we compared the type I IFN response in MRC5 fibroblasts and

CD34+ HPCs in response to HCMV infection. Future work will elucidate the mechanism by which pUL138 contributes to HCMV latency. It will be interesting to understand the possible significance of IRES mediated pUL138 expression 153

during the establishment of latency in addition to the potential role of type I IFN modulation by the ULb′ ORFs UL136-UL142 during HCMV infection in CD34+

HPCs. 154

REFERENCES

1. Abbas, A. a. L., Andrew. 2003. Cellular and Molecular Immunology, Fifth ed. Elsevier Science.

2. Adler, S. P., S. H. Hempfling, S. E. Starr, S. A. Plotkin, and S. Riddell. 1998. Safety and immunogenicity of the Towne strain cytomegalovirus vaccine. Pediatr Infect Dis J 17:200-6.

3. Aebi, M., J. Fah, N. Hurt, C. E. Samuel, D. Thomis, L. Bazzigher, J. Pavlovic, O. Haller, and P. Staeheli. 1989. cDNA structures and regulation of two interferon-induced human Mx proteins. Mol Cell Biol 9:5062-72.

4. Allen, R. D., 3rd, S. Dickerson, and S. H. Speck. 2006. Identification of spliced gammaherpesvirus 68 LANA and v-cyclin transcripts and analysis of their expression in vivo during latent infection. J Virol 80:2055-62.

5. Alwine, J. C. 2008. Modulation of host cell stress responses by human cytomegalovirus. Curr Top Microbiol Immunol 325:263-79.

6. Arvin, A. M., P. Fast, M. Myers, S. Plotkin, and R. Rabinovich. 2004. Vaccine development to prevent cytomegalovirus disease: report from the National Vaccine Advisory Committee. Clin Infect Dis 39:233-9.

7. Ashiru, O., N. J. Bennett, L. H. Boyle, M. Thomas, J. Trowsdale, and M. R. Wills. 2009. NKG2D ligand MICA is retained in the cis-Golgi apparatus by human cytomegalovirus protein UL142. J Virol 83:12345-54.

8. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1998. Current Protocols in Molecular Biology, vol. 1. Wiley and Sons Inc.

9. Baird, S. D., M. Turcotte, R. G. Korneluk, and M. Holcik. 2006. Searching for IRES. RNA 12:1755-85.

10. Ballestas, M. E., P. A. Chatis, and K. M. Kaye. 1999. Efficient persistence of extrachromosomal KSHV DNA mediated by latency-associated nuclear antigen. Science 284:641-4.

11. Barton, E. S., M. L. Lutzke, R. Rochford, and H. W. t. Virgin. 2005. Alpha/beta interferons regulate murine gammaherpesvirus latent gene expression and reactivation from latency. J Virol 79:14149-60. 155

12. Bego, M., J. Maciejewski, S. Khaiboullina, G. Pari, and S. St Jeor. 2005. Characterization of an antisense transcript spanning the UL81-82 locus of human cytomegalovirus. J Virol 79:11022-34.

13. Belsham, G. J., and N. Sonenberg. 1996. RNA-protein interactions in regulation of picornavirus RNA translation. Microbiol Rev 60:499-511.

14. Bevan, I. S., C. C. Sammons, and C. Sweet. 1996. Investigation of murine cytomegalovirus latency and reactivation in mice using viral mutants and the polymerase chain reaction. J Med Virol 48:308-20.

15. Bieleski, L., C. Hindley, and S. J. Talbot. 2004. A polypyrimidine tract facilitates the expression of Kaposi's sarcoma-associated herpesvirus vFLIP through an internal ribosome entry site. J Gen Virol 85:615-20.

16. Bieleski, L., and S. J. Talbot. 2001. Kaposi's sarcoma-associated herpesvirus vCyclin open reading frame contains an internal ribosome entry site. J Virol 75:1864-9.

17. Boehm, U., T. Klamp, M. Groot, and J. C. Howard. 1997. Cellular responses to interferon-gamma. Annu Rev Immunol 15:749-95.

18. Browne, E. P., B. Wing, D. Coleman, and T. Shenk. 2001. Altered cellular mRNA levels in human cytomegalovirus-infected fibroblasts: viral block to the accumulation of antiviral mRNAs. J Virol 75:12319-30.

19. Burny, W., C. Liesnard, C. Donner, and A. Marchant. 2004. Epidemiology, pathogenesis and prevention of congenital cytomegalovirus infection. Expert Rev Anti Infect Ther 2:881-94.

20. Cao, J., and A. P. Geballe. 1995. Translational inhibition by a human cytomegalovirus upstream open reading frame despite inefficient utilization of its AUG codon. J Virol 69:1030-6.

21. Castello, A., J. M. Izquierdo, E. Welnowska, and L. Carrasco. 2009. RNA nuclear export is blocked by poliovirus 2A protease and is concomitant with nucleoporin cleavage. J Cell Sci 122:3799-809.

22. Cebulla, C. M., D. M. Miller, and D. D. Sedmak. 1999. Viral inhibition of interferon signal transduction. Intervirology 42:325-30.

23. Cha, T. A., E. Tom, G. W. Kemble, G. M. Duke, E. S. Mocarski, and R. R. Spaete. 1996. Human cytomegalovirus clinical isolates carry at least 19 genes not found in laboratory strains. J Virol 70:78-83. 156

24. Chalupny, N. J., A. Rein-Weston, S. Dosch, and D. Cosman. 2006. Down- regulation of the NKG2D ligand MICA by the human cytomegalovirus glycoprotein UL142. Biochem Biophys Res Commun 346:175-81.

25. Chatterjee, S., and J. K. Pal. 2009. Role of 5'- and 3'-untranslated regions of mRNAs in human diseases. Biol Cell 101:251-62.

26. Chen, S. H., L. Y. Lee, D. A. Garber, P. A. Schaffer, D. M. Knipe, and D. M. Coen. 2002. Neither LAT nor open reading frame P mutations increase expression of spliced or intron-containing ICP0 transcripts in mouse ganglia latently infected with herpes simplex virus. J Virol 76:4764-72.

27. Cheung, A. K., A. Abendroth, A. L. Cunningham, and B. Slobedman. 2006. Viral gene expression during the establishment of human cytomegalovirus latent infection in myeloid progenitor cells. Blood 108:3691-9.

28. Chin, K. C., and P. Cresswell. 2001. Viperin (cig5), an IFN-inducible antiviral protein directly induced by human cytomegalovirus. Proc Natl Acad Sci U S A 98:15125-30.

29. Coleman, H. M., I. Brierley, and P. G. Stevenson. 2003. An internal ribosome entry site directs translation of the murine gammaherpesvirus 68 MK3 open reading frame. J Virol 77:13093-105.

30. Coll, O., G. Benoist, Y. Ville, L. E. Weisman, F. Botet, M. A. T. W. Maurizio, A. Greenough, R. S. Gibbs, and X. Carbonell-Estrany Coordinator. 2009. Guidelines on CMV congenital infection. J Perinat Med 37:433-45.

31. Compton, T., R. R. Nepomuceno, and D. M. Nowlin. 1992. Human cytomegalovirus penetrates host cells by pH-independent fusion at the cell surface. Virology 191:387-95.

32. Compton, T., D. M. Nowlin, and N. R. Cooper. 1993. Initiation of human cytomegalovirus infection requires initial interaction with cell surface heparan sulfate. Virology 193:834-41.

33. Cornelis, S., Y. Bruynooghe, G. Denecker, S. Van Huffel, S. Tinton, and R. Beyaert. 2000. Identification and characterization of a novel cell cycle- regulated internal ribosome entry site. Mol Cell 5:597-605.

34. Cosme, R. S., Y. Yamamura, and Q. Tang. 2009. Roles of polypyrimidine tract binding proteins in major immediate-early gene expression and viral replication of human cytomegalovirus. J Virol 83:2839-50. 157

35. Cotter, M. A., 2nd, C. Subramanian, and E. S. Robertson. 2001. The Kaposi's sarcoma-associated herpesvirus latency-associated nuclear antigen binds to specific sequences at the left end of the viral genome through its carboxy-terminus. Virology 291:241-59.

36. Creancier, L., D. Morello, P. Mercier, and A. C. Prats. 2000. Fibroblast growth factor 2 internal ribosome entry site (IRES) activity ex vivo and in transgenic mice reveals a stringent tissue-specific regulation. J Cell Biol 150:275-81.

37. Dhar, D., S. Roy, and S. Das. 2007. Translational control of the interferon regulatory factor 2 mRNA by IRES element. Nucleic Acids Res 35:5409- 21.

38. Dolan, A., C. Cunningham, R. D. Hector, A. F. Hassan-Walker, L. Lee, C. Addison, D. J. Dargan, D. J. McGeoch, D. Gatherer, V. C. Emery, P. D. Griffiths, C. Sinzger, B. P. McSharry, G. W. Wilkinson, and A. J. Davison. 2004. Genetic content of wild-type human cytomegalovirus. J Gen Virol 85:1301-12.

39. Dorner, A. J., B. L. Semler, R. J. Jackson, R. Hanecak, E. Duprey, and E. Wimmer. 1984. In vitro translation of poliovirus RNA: utilization of internal initiation sites in reticulocyte lysate. J Virol 50:507-14.

40. Fernandez, J., I. Yaman, R. Mishra, W. C. Merrick, M. D. Snider, W. H. Lamers, and M. Hatzoglou. 2001. Internal ribosome entry site-mediated translation of a mammalian mRNA is regulated by amino acid availability. J Biol Chem 276:12285-91.

41. Filbin, M. E., and J. S. Kieft. 2009. Toward a structural understanding of IRES RNA function. Curr Opin Struct Biol 19:267-76.

42. Fitzgerald, K. D., and B. L. Semler. 2009. Bridging IRES elements in mRNAs to the eukaryotic translation apparatus. Biochim Biophys Acta 1789:518-28.

43. Fortunato, E. A., and D. H. Spector. 1999. Regulation of human cytomegalovirus gene expression. Adv Virus Res 54:61-128.

44. Fraser, C. S., and J. A. Doudna. 2007. Structural and mechanistic insights into hepatitis C viral translation initiation. Nat Rev Microbiol 5:29-38.

45. Fuld, S., C. Cunningham, K. Klucher, A. J. Davison, and D. J. Blackbourn. 2006. Inhibition of interferon signaling by the Kaposi's sarcoma-associated 158

herpesvirus full-length viral interferon regulatory factor 2 protein. J Virol 80:3092-7.

46. Gaddy, C. E., D. S. Wong, A. Markowitz-Shulman, and A. M. Colberg- Poley. Regulation of the sub-cellular distribution of key cellular RNA processing factors during permissive human cytomegalovirus infection. J Gen Virol.

47. Ganem, D. 2007. Kaposi's Sarcoma-associated Herpesviruses, p. 2847- 2888. In D. M. a. H. Knipe, Peter M. (ed.), Fields Virology, vol. 2. Lippincott Williams and Wilkins, a Wolters Kluwer Business, Philadelphia.

48. Gilbert, W. V., K. Zhou, T. K. Butler, and J. A. Doudna. 2007. Cap- independent translation is required for starvation-induced differentiation in yeast. Science 317:1224-7.

49. Goodrum, F., C. T. Jordan, S. S. Terhune, K. High, and T. Shenk. 2004. Differential outcomes of human cytomegalovirus infection in primitive hematopoietic cell subpopulations. Blood 104:687-95.

50. Goodrum, F. D., C. T. Jordan, K. High, and T. Shenk. 2002. Human cytomegalovirus gene expression during infection of primary hematopoietic progenitor cells: a model for latency. Proc Natl Acad Sci U S A 99:16255-60.

51. Goodrum, F. D., M. Reeves, J. Sinclair, K. High, and T. Shenk. 2007. Human cytomegalovirus sequences expressed in latently infected individuals promote a latent infection in vitro. Blood.

52. Gretch, D. R., B. Kari, R. C. Gehrz, and M. F. Stinski. 1988. A multigene family encodes the human cytomegalovirus glycoprotein complex gcII (gp47-52 complex). J Virol 62:1956-62.

53. Griffiths, P. D. 2000. Cytomegalovirus, p. 79-116. In J. E. B. In A. J. Zuckerman, & J. R. Pattison (Eds.) (ed.), Prinicipals and Practice of Clinical Virology. Wiley and Sons, London.

54. Grundhoff, A., and D. Ganem. 2001. Mechanisms governing expression of the v-FLIP gene of Kaposi's sarcoma-associated herpesvirus. J Virol 75:1857-63.

55. Guo, J., D. J. Hui, W. C. Merrick, and G. C. Sen. 2000. A new pathway of translational regulation mediated by eukaryotic initiation factor 3. EMBO J 19:6891-9. 159

56. Guo, Y. W., and E. S. Huang. 1993. Characterization of a structurally tricistronic gene of human cytomegalovirus composed of U(s)18, U(s)19, and U(s)20. J Virol 67:2043-54.

57. Hahn, G., H. Khan, F. Baldanti, U. H. Koszinowski, M. G. Revello, and G. Gerna. 2002. The human cytomegalovirus ribonucleotide reductase homolog UL45 is dispensable for growth in endothelial cells, as determined by a BAC-cloned clinical isolate of human cytomegalovirus with preserved wild-type characteristics. J Virol 76:9551-5.

58. Halford, W. P., C. Weisend, J. Grace, M. Soboleski, D. J. Carr, J. W. Balliet, Y. Imai, T. P. Margolis, and B. M. Gebhardt. 2006. ICP0 antagonizes Stat 1-dependent repression of herpes simplex virus: implications for the regulation of viral latency. Virol J 3:44.

59. Haller, A. A., and B. L. Semler. 1992. Linker scanning mutagenesis of the internal ribosome entry site of poliovirus RNA. J Virol 66:5075-86.

60. Harding, H. P., I. Novoa, Y. Zhang, H. Zeng, R. Wek, M. Schapira, and D. Ron. 2000. Regulated translation initiation controls stress-induced gene expression in mammalian cells. Mol Cell 6:1099-108.

61. Hazari, S., A. Patil, V. Joshi, D. E. Sullivan, C. D. Fermin, R. F. Garry, R. M. Elliott, and S. Dash. 2005. Alpha interferon inhibits translation mediated by the internal ribosome entry site of six different hepatitis C virus genotypes. J Gen Virol 86:3047-53.

62. Heineman, T. C., M. Schleiss, D. I. Bernstein, R. R. Spaete, L. Yan, G. Duke, M. Prichard, Z. Wang, Q. Yan, M. A. Sharp, N. Klein, A. M. Arvin, and G. Kemble. 2006. A phase 1 study of 4 live, recombinant human cytomegalovirus Towne/Toledo chimeric vaccines. J Infect Dis 193:1350- 60.

63. Hellen, C. U., and P. Sarnow. 2001. Internal ribosome entry sites in eukaryotic mRNA molecules. Genes Dev 15:1593-612.

64. Hershey, J., and W. Merrick. 2000. The pathway and mechanism of initiation of protein synthesis. In Translation Control of Gene Expression. Cold Spring Harbor Laboratory Press, New York.

65. Holcik, M., B. W. Gordon, and R. G. Korneluk. 2003. The internal ribosome entry site-mediated translation of antiapoptotic protein XIAP is modulated by the heterogeneous nuclear ribonucleoproteins C1 and C2. Mol Cell Biol 23:280-8. 160

66. Holcik, M., C. Lefebvre, C. Yeh, T. Chow, and R. G. Korneluk. 1999. A new internal-ribosome-entry-site motif potentiates XIAP-mediated cytoprotection. Nat Cell Biol 1:190-2.

67. Holcik, M., and N. Sonenberg. 2005. Translational control in stress and apoptosis. Nat Rev Mol Cell Biol 6:318-27.

68. Holcik, M., N. Sonenberg, and R. G. Korneluk. 2000. Internal ribosome initiation of translation and the control of cell death. Trends Genet 16:469- 73.

69. Holcik, M., C. Yeh, R. G. Korneluk, and T. Chow. 2000. Translational upregulation of X-linked inhibitor of apoptosis (XIAP) increases resistance to radiation induced cell death. Oncogene 19:4174-7.

70. Hovanes, K., T. W. Li, J. E. Munguia, T. Truong, T. Milovanovic, J. Lawrence Marsh, R. F. Holcombe, and M. L. Waterman. 2001. Beta- catenin-sensitive isoforms of lymphoid enhancer factor-1 are selectively expressed in colon cancer. Nat Genet 28:53-7.

71. Ioudinkova, E., M. C. Arcangeletti, A. Rynditch, F. De Conto, F. Motta, S. Covan, F. Pinardi, S. V. Razin, and C. Chezzi. 2006. Control of human cytomegalovirus gene expression by differential histone modifications during lytic and latent infection of a monocytic cell line. Gene 384:120-8.

72. Isaksson, A., M. Berggren, and A. Ricksten. 2003. Epstein-Barr virus U leader exon contains an internal ribosome entry site. Oncogene 22:572- 81.

73. Jabbour, M., E. M. Campbell, H. Fares, and L. Lybarger. 2009. Discrete domains of MARCH1 mediate its localization, functional interactions, and posttranscriptional control of expression. J Immunol 183:6500-12.

74. Jackson, R. J., C. U. Hellen, and T. V. Pestova. The mechanism of eukaryotic translation initiation and principles of its regulation. Nat Rev Mol Cell Biol 11:113-27.

75. Jang, S. K., H. G. Krausslich, M. J. Nicklin, G. M. Duke, A. C. Palmenberg, and E. Wimmer. 1988. A segment of the 5' nontranslated region of encephalomyocarditis virus RNA directs internal entry of ribosomes during in vitro translation. J Virol 62:2636-43.

76. Jarvis, M. A., and J. A. Nelson. 2002. Mechanisms of human cytomegalovirus persistence and latency. Front Biosci 7:d1575-82. 161

77. Jenkins, C., A. Abendroth, and B. Slobedman. 2004. A novel viral transcript with homology to human interleukin-10 is expressed during latent human cytomegalovirus infection. J Virol 78:1440-7.

78. Jimenez, J., G. M. Jang, B. L. Semler, and M. L. Waterman. 2005. An internal ribosome entry site mediates translation of lymphoid enhancer factor-1. RNA 11:1385-99.

79. Kafasla, P., N. Morgner, T. A. Poyry, S. Curry, C. V. Robinson, and R. J. Jackson. 2009. Polypyrimidine tract binding protein stabilizes the encephalomyocarditis virus IRES structure via binding multiple sites in a unique orientation. Mol Cell 34:556-68.

80. Kerr, I. M., and R. E. Brown. 1978. pppA2'p5'A2'p5'A: an inhibitor of protein synthesis synthesized with an enzyme fraction from interferon- treated cells. Proc Natl Acad Sci U S A 75:256-60.

81. Kieft, J. S. 2008. Viral IRES RNA structures and ribosome interactions. Trends Biochem Sci 33:274-83.

82. Knipe, D. M. a. H., Peter M. 2007. Fields Virology, 5th ed, vol. 1. Lippincott Williams and Wilkins, a Wolters Kluwer Business.

83. Kochetov, A. V. 2008. Alternative translation start sites and hidden coding potential of eukaryotic mRNAs. Bioessays 30:683-91.

84. Kochetov, A. V., S. Ahmad, V. Ivanisenko, O. A. Volkova, N. A. Kolchanov, and A. Sarai. 2008. uORFs, reinitiation and alternative translation start sites in human mRNAs. FEBS Lett 582:1293-7.

85. Kondo, K., J. Sashihara, K. Shimada, M. Takemoto, K. Amo, H. Miyagawa, and K. Yamanishi. 2003. Recognition of a novel stage of betaherpesvirus latency in human herpesvirus 6. J Virol 77:2258-64.

86. Kondo, K., J. Xu, and E. S. Mocarski. 1996. Human cytomegalovirus latent gene expression in granulocyte-macrophage progenitors in culture and in seropositive individuals. Proc Natl Acad Sci U S A 93:11137-42.

87. Kozak, M. 2002. Pushing the limits of the scanning mechanism for initiation of translation. Gene 299:1-34.

88. Kozak, M. 1991. Structural features in eukaryotic mRNAs that modulate the initiation of translation. J Biol Chem 266:19867-70. 162

89. Landini, M. P., T. Lazzarotto, J. Xu, A. P. Geballe, and E. S. Mocarski. 2000. Humoral immune response to proteins of human cytomegalovirus latency-associated transcripts. Biol Blood Marrow Transplant 6:100-8.

90. Landolfo, S., M. Gariglio, G. Gribaudo, and D. Lembo. 2003. The human cytomegalovirus. Pharmacol Ther 98:269-97.

91. Lewis, S. M., and M. Holcik. 2008. For IRES trans-acting factors, it is all about location. Oncogene 27:1033-5.

92. Lewis, S. M., and M. Holcik. 2005. IRES in distress: translational regulation of the inhibitor of apoptosis proteins XIAP and HIAP2 during cell stress. Cell Death Differ 12:547-53.

93. Lewis, S. M., A. Veyrier, N. Hosszu Ungureanu, S. Bonnal, S. Vagner, and M. Holcik. 2007. Subcellular relocalization of a trans-acting factor regulates XIAP IRES-dependent translation. Mol Biol Cell 18:1302-11.

94. Li, J., S. Lin, Q. Chen, L. Peng, J. Zhai, Y. Liu, and Z. Yuan. 2010. Inhibition of hepatitis B virus replication by MyD88 involves accelerated degradation of pregenomic RNA and nuclear retention of preS/S RNAs. J Virol.

95. Liston, P., W. G. Fong, and R. G. Korneluk. 2003. The inhibitors of apoptosis: there is more to life than Bcl2. Oncogene 22:8568-80.

96. Liu, R., J. Baillie, J. G. Sissons, and J. H. Sinclair. 1994. The transcription factor YY1 binds to negative regulatory elements in the human cytomegalovirus major immediate early enhancer/promoter and mediates repression in non-permissive cells. Nucleic Acids Res 22:2453-9.

97. Low, W., M. Harries, H. Ye, M. Q. Du, C. Boshoff, and M. Collins. 2001. Internal ribosome entry site regulates translation of Kaposi's sarcoma- associated herpesvirus FLICE inhibitory protein. J Virol 75:2938-45.

98. Lubyova, B., M. J. Kellum, A. J. Frisancho, and P. M. Pitha. 2004. Kaposi's sarcoma-associated herpesvirus-encoded vIRF-3 stimulates the transcriptional activity of cellular IRF-3 and IRF-7. J Biol Chem 279:7643- 54.

99. Macejak, D. G., and P. Sarnow. 1991. Internal initiation of translation mediated by the 5' leader of a cellular mRNA. Nature 353:90-4.

100. Majumder, M., I. Yaman, F. Gaccioli, V. V. Zeenko, C. Wang, M. G. Caprara, R. C. Venema, A. A. Komar, M. D. Snider, and M. Hatzoglou. 163

2009. The hnRNA-binding proteins hnRNP L and PTB are required for efficient translation of the Cat-1 arginine/lysine transporter mRNA during amino acid starvation. Mol Cell Biol 29:2899-912.

101. Mao, Z. Q., Y. Huang, Q. Ruan, M. Sun, Y. Qi, and H. Gao. 2007. [Polymorphisms of human cytomegalovirus UL139 open reading frame in Hirschsprung's disease]. Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi 21:322-4.

102. Martinez-Salas, E., R. Ramos, E. Lafuente, and S. Lopez de Quinto. 2001. Functional interactions in internal translation initiation directed by viral and cellular IRES elements. J Gen Virol 82:973-84.

103. Massa, P. T., L. W. Whitney, C. Wu, S. L. Ropka, and K. W. Jarosinski. 1999. A mechanism for selective induction of 2'-5' oligoadenylate synthetase, anti-viral state, but not MHC class I genes by interferon-beta in neurons. J Neurovirol 5:161-71.

104. Meijer, H. A., and A. A. Thomas. 2003. Ribosomes stalling on uORF1 in the Xenopus Cx41 5' UTR inhibit downstream translation initiation. Nucleic Acids Res 31:3174-84.

105. Mendelson, M., S. Monard, P. Sissons, and J. Sinclair. 1996. Detection of endogenous human cytomegalovirus in CD34+ bone marrow progenitors. J Gen Virol 77 ( Pt 12):3099-102.

106. Merrill, M. K., E. Y. Dobrikova, and M. Gromeier. 2006. Cell-type-specific repression of internal ribosome entry site activity by double-stranded RNA- binding protein 76. J Virol 80:3147-56.

107. Mignone, F., C. Gissi, S. Liuni, and G. Pesole. 2002. Untranslated regions of mRNAs. Genome Biol 3:REVIEWS0004.

108. Miller, D. M., Y. Zhang, B. M. Rahill, W. J. Waldman, and D. D. Sedmak. 1999. Human cytomegalovirus inhibits IFN-alpha-stimulated antiviral and immunoregulatory responses by blocking multiple levels of IFN-alpha signal transduction. J Immunol 162:6107-13.

109. Mocarski, E. S., and Courcelle, C. T. . 2001. Cytomegalovirus and their replication, p. 2629-2673. In a. P. H. In D. Knipe (ed.), Fields Virology. Lippincott, Williams and Wilkins, Philadelphia.

110. Mocarski, E. S., T. Shenk, and R. F. Pass. 2007. Cytomegaloviruses. In D. M. K. a. P. M. Howley (ed.), Fields Virology, 5th ed, vol. 1. Lippincott, Williams & Wilkins, Philadelphia. 164

111. Moorman, N. J., I. M. Cristea, S. S. Terhune, M. P. Rout, B. T. Chait, and T. Shenk. 2008. Human cytomegalovirus protein UL38 inhibits host cell stress responses by antagonizing the tuberous sclerosis protein complex. Cell Host Microbe 3:253-62.

112. Murphy, E., D. Yu, J. Grimwood, J. Schmutz, M. Dickson, M. A. Jarvis, G. Hahn, J. A. Nelson, R. M. Myers, and T. E. Shenk. 2003. Coding potential of laboratory and clinical strains of human cytomegalovirus. Proc Natl Acad Sci U S A 100:14976-81.

113. Murphy, E. a. S., Thomas. 2008. Human Cytomegalovirus Genome, p. 1- 19. In T. a. S. Shenk, Mark (ed.), Human Cytomegalovirus, vol. 325. Springer, Verlag Berlin Heidelberg.

114. Murphy, J. C., W. Fischle, E. Verdin, and J. H. Sinclair. 2002. Control of cytomegalovirus lytic gene expression by histone acetylation. EMBO J 21:1112-20.

115. Nagai, Y., K. P. Garrett, S. Ohta, U. Bahrun, T. Kouro, S. Akira, K. Takatsu, and P. W. Kincade. 2006. Toll-like receptors on hematopoietic progenitor cells stimulate innate immune system replenishment. Immunity 24:801-12.

116. Pass, R. F. 2001. Cytomegalovirus, p. 2675-2705. In D. M. a. H. Knipe, P. M. (ed.), Fields Virology, vol. 4. Lippincot, Williams & Wilkins, Philadelphia.

117. Pass, R. F., A. M. Duliege, S. Boppana, R. Sekulovich, S. Percell, W. Britt, and R. L. Burke. 1999. A subunit cytomegalovirus vaccine based on recombinant envelope glycoprotein B and a new adjuvant. J Infect Dis 180:970-5.

118. Paulus, C., S. Krauss, and M. Nevels. 2006. A human cytomegalovirus antagonist of type I IFN-dependent signal transducer and activator of transcription signaling. Proc Natl Acad Sci U S A 103:3840-5.

119. Pearce, M., S. Matsumura, and A. C. Wilson. 2005. Transcripts encoding K12, v-FLIP, v-cyclin, and the microRNA cluster of Kaposi's sarcoma- associated herpesvirus originate from a common promoter. J Virol 79:14457-64.

120. Pelletier, J., and N. Sonenberg. 1988. Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature 334:320-5. 165

121. Pepperl-Klindworth, S., K. Besold, N. Frankenberg, M. Farkas, J. Kuball, M. Theobald, and B. Plachter. 2006. Cytomegalovirus interleukin-10 expression in infected cells does not impair MHC class I restricted Peptide presentation on bystanding antigen-presenting cells. Viral Immunol 19:92- 101.

122. Perez, I., C. H. Lin, J. G. McAfee, and J. G. Patton. 1997. Mutation of PTB binding sites causes misregulation of alternative 3' splice site selection in vivo. RNA 3:764-78.

123. Pestova, T. V., V. G. Kolupaeva, I. B. Lomakin, E. V. Pilipenko, I. N. Shatsky, V. I. Agol, and C. U. Hellen. 2001. Molecular mechanisms of translation initiation in eukaryotes. Proc Natl Acad Sci U S A 98:7029-36.

124. Petrucelli, A., M. Rak, L. Grainger, and F. Goodrum. 2009. Characterization of a novel Golgi apparatus-localized latency determinant encoded by human cytomegalovirus. J Virol 83:5615-29.

125. Pfaffl, M. W. 2001. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45.

126. Pilipenko, E. V., T. V. Pestova, V. G. Kolupaeva, E. V. Khitrina, A. N. Poperechnaya, V. I. Agol, and C. U. Hellen. 2000. A cell cycle-dependent protein serves as a template-specific translation initiation factor. Genes Dev 14:2028-45.

127. Poole, E., C. A. King, J. H. Sinclair, and A. Alcami. 2006. The UL144 gene product of human cytomegalovirus activates NFkappaB via a TRAF6- dependent mechanism. Embo J 25:4390-9.

128. Poon, A. P., L. Benetti, and B. Roizman. 2006. U(S)3 and U(S)3.5 protein kinases of herpes simplex virus 1 differ with respect to their functions in blocking apoptosis and in virion maturation and egress. J Virol 80:3752- 64.

129. Poon, A. P., H. Gu, and B. Roizman. 2006. ICP0 and the US3 protein kinase of herpes simplex virus 1 independently block histone deacetylation to enable gene expression. Proc Natl Acad Sci U S A 103:9993-8.

130. Poon, A. P., and B. Roizman. 2007. Mapping of key functions of the herpes simplex virus 1 U(S)3 protein kinase: the U(S)3 protein can form functional heteromultimeric structures derived from overlapping truncated polypeptides. J Virol 81:1980-9. 166

131. Poulin, F., A. C. Gingras, H. Olsen, S. Chevalier, and N. Sonenberg. 1998. 4E-BP3, a new member of the eukaryotic initiation factor 4E-binding protein family. J Biol Chem 273:14002-7.

132. Preston, C. M., and M. J. Nicholl. 2006. Role of the cellular protein hDaxx in human cytomegalovirus immediate-early gene expression. J Gen Virol 87:1113-21.

133. Prod'homme, V., D. M. Sugrue, R. J. Stanton, A. Nomoto, J. Davies, C. R. Rickards, D. Cochrane, M. Moore, G. W. Wilkinson, and P. Tomasec. Human Cytomegalovirus UL141 Promotes Efficient Downregulation of the Natural Killer cell Activating Ligand CD112. J Gen Virol.

134. Prosch, S., R. Wuttke, D. H. Kruger, and H. D. Volk. 2002. NF-kappaB--a potential therapeutic target for inhibition of human cytomegalovirus (re)activation? Biol Chem 383:1601-9.

135. Pyronnet, S., L. Pradayrol, and N. Sonenberg. 2000. A cell cycle- dependent internal ribosome entry site. Mol Cell 5:607-16.

136. Qi, Y., Y. P. Ma, R. He, Z. R. Sun, Y. H. Ji, Y. J. Huang, and Q. Ruan. 2008. Sequence conservation of human cytomegalovirus UL140 open reading frame in clinical strains. Chin Med Sci J 23:23-7.

137. Rawlinson, W. D., H. E. Farrell, and B. G. Barrell. 1996. Analysis of the complete DNA sequence of murine cytomegalovirus. J Virol 70:8833-49.

138. Reddehase, M. J. 2006. Cytomegaloviruses Molecular Biology and Immunology. Caister Academic Press, Norfolk, U.K.

139. Reeves, M. B., P. J. Lehner, J. G. Sissons, and J. H. Sinclair. 2005. An in vitro model for the regulation of human cytomegalovirus latency and reactivation in dendritic cells by chromatin remodelling. J Gen Virol 86:2949-54.

140. Reeves, M. B., P. A. MacAry, P. J. Lehner, J. G. Sissons, and J. H. Sinclair. 2005. Latency, chromatin remodeling, and reactivation of human cytomegalovirus in the dendritic cells of healthy carriers. Proc Natl Acad Sci U S A 102:4140-5.

141. Reeves, M. B., and J. H. Sinclair. 2010. Analysis of latent viral gene expression in natural and experimental latency models of human cytomegalovirus and its correlation with histone modifications at a latent promoter. J Gen Virol 91:599-604. 167

142. Reeves, M. B. a. S., John. 2008. Aspects of Human Cytomegalovirus Latency and Reactivation, p. 297-313. In T. a. S. Shenk, Mark (ed.), Human Cytomegalovirus, vol. 325. Springer, Verlag Berlin Heidelberg.

143. Rivas, C., A. E. Thlick, C. Parravicini, P. S. Moore, and Y. Chang. 2001. Kaposi's sarcoma-associated herpesvirus LANA2 is a B-cell-specific latent viral protein that inhibits p53. J Virol 75:429-38.

144. Rixon, F. J., and D. J. McGeoch. 1985. Detailed analysis of the mRNAs mapping in the short unique region of herpes simplex virus type 1. Nucleic Acids Res 13:953-73.

145. Roizman, B., Knipe, D. M., Whitley, R. J. 2007. Herpes simplex viruses, p. 2501-2601. In D. M. knipe, Howley, P. M. (ed.), Fields Virology, vol. 2. Lippincott Williams and Wilkins, a Wolters Kluwer Business, Philadephia.

146. Ruger, B., S. Klages, B. Walla, J. Albrecht, B. Fleckenstein, P. Tomlinson, and B. Barrell. 1987. Primary structure and transcription of the genes coding for the two virion phosphoproteins pp65 and pp71 of human cytomegalovirus. J Virol 61:446-53.

147. Ryabova, L. A., M. M. Pooggin, and T. Hohn. 2006. Translation reinitiation and leaky scanning in plant viruses. Virus Res 119:52-62.

148. Saffert, R. T., R. R. Penkert, and R. F. Kalejta. 2010. Cellular and Viral Control over the Initial Events of Human Cytomegalovirus Experimental Latency in CD34+ Cells. J Virol.

149. Samuel, C. E. 1991. Antiviral actions of interferon. Interferon-regulated cellular proteins and their surprisingly selective antiviral activities. Virology 183:1-11.

150. Samuel, C. E. 2001. Antiviral actions of interferons. Clin Microbiol Rev 14:778-809, table of contents.

151. Sarnow, P. 1989. Translation of glucose-regulated protein 78/immunoglobulin heavy-chain binding protein mRNA is increased in poliovirus-infected cells at a time when cap-dependent translation of cellular mRNAs is inhibited. Proc Natl Acad Sci U S A 86:5795-9.

152. Sarnow, P. 2003. Viral internal ribosome entry site elements: novel ribosome-RNA complexes and roles in viral pathogenesis. J Virol 77:2801- 6. 168

153. Sarnow, P., R. C. Cevallos, and E. Jan. 2005. Takeover of host ribosomes by divergent IRES elements. Biochem Soc Trans 33:1479-82.

154. Schepens, B., S. A. Tinton, Y. Bruynooghe, E. Parthoens, M. Haegman, R. Beyaert, and S. Cornelis. 2007. A role for hnRNP C1/C2 and Unr in internal initiation of translation during mitosis. EMBO J 26:158-69.

155. Seipelt, J., A. Guarne, E. Bergmann, M. James, W. Sommergruber, I. Fita, and T. Skern. 1999. The structures of picornaviral proteinases. Virus Res 62:159-68.

156. Semler, B. L., and M. L. Waterman. 2008. IRES-mediated pathways to polysomes: nuclear versus cytoplasmic routes. Trends Microbiol 16:1-5.

157. Sindre, H., G. E. Tjoonnfjord, H. Rollag, T. Ranneberg-Nilsen, O. P. Veiby, S. Beck, M. Degre, and K. Hestdal. 1996. Human cytomegalovirus suppression of and latency in early hematopoietic progenitor cells. Blood 88:4526-33.

158. Soderberg-Naucler, C. 2006. Does cytomegalovirus play a causative role in the development of various inflammatory diseases and cancer? J Intern Med 259:219-46.

159. Soderberg-Naucler, C., K. N. Fish, and J. A. Nelson. 1997. Reactivation of latent human cytomegalovirus by allogeneic stimulation of blood cells from healthy donors. Cell 91:119-26.

160. Soderberg-Naucler, C., D. N. Streblow, K. N. Fish, J. Allan-Yorke, P. P. Smith, and J. A. Nelson. 2001. Reactivation of latent human cytomegalovirus in CD14(+) monocytes is differentiation dependent. J Virol 75:7543-54.

161. Sonenberg, N., and A. G. Hinnebusch. 2007. New modes of translational control in development, behavior, and disease. Mol Cell 28:721-9.

162. Spriggs, K. A., M. Stoneley, M. Bushell, and A. E. Willis. 2008. Re- programming of translation following cell stress allows IRES-mediated translation to predominate. Biol Cell 100:27-38.

163. Stagno, S., R. F. Pass, G. Cloud, W. J. Britt, R. E. Henderson, P. D. Walton, D. A. Veren, F. Page, and C. A. Alford. 1986. Primary cytomegalovirus infection in pregnancy. Incidence, transmission to fetus, and clinical outcome. JAMA 256:1904-8. 169

164. Stark, G. R., I. M. Kerr, B. R. Williams, R. H. Silverman, and R. D. Schreiber. 1998. How cells respond to interferons. Annu Rev Biochem 67:227-64.

165. Stewart, S. R., and B. L. Semler. 1998. RNA structure adjacent to the attenuation determinant in the 5'-non-coding region influences poliovirus viability. Nucleic Acids Res 26:5318-26.

166. Stoneley, M., T. Subkhankulova, J. P. Le Quesne, M. J. Coldwell, C. L. Jopling, G. J. Belsham, and A. E. Willis. 2000. Analysis of the c-myc IRES; a potential role for cell-type specific trans-acting factors and the nuclear compartment. Nucleic Acids Res 28:687-94.

167. Svitkin, Y. V., T. V. Pestova, S. V. Maslova, and V. I. Agol. 1988. Point mutations modify the response of poliovirus RNA to a translation initiation factor: a comparison of neurovirulent and attenuated strains. Virology 166:394-404.

168. Sylwester, A. W., B. L. Mitchell, J. B. Edgar, C. Taormina, C. Pelte, F. Ruchti, P. R. Sleath, K. H. Grabstein, N. A. Hosken, F. Kern, J. A. Nelson, and L. J. Picker. 2005. Broadly targeted human cytomegalovirus-specific CD4+ and CD8+ T cells dominate the memory compartments of exposed subjects. J Exp Med 202:673-85.

169. Tahiri-Alaoui, A., D. Matsuda, H. Xu, P. Panagiotis, L. Burman, L. S. Lambeth, L. Petherbridge, W. James, V. Mauro, and V. Nair. 2009. The 5' leader of the mRNA encoding the marek's disease virus serotype 1 pp14 protein contains an intronic internal ribosome entry site with allosteric properties. J Virol 83:12769-78.

170. Tahiri-Alaoui, A., L. P. Smith, S. Baigent, L. Kgosana, L. J. Petherbridge, L. S. Lambeth, W. James, and V. Nair. 2009. Identification of an intercistronic internal ribosome entry site in a Marek's disease virus immediate-early gene. J Virol 83:5846-53.

171. Tavalai, N., P. Papior, S. Rechter, M. Leis, and T. Stamminger. 2006. Evidence for a role of the cellular ND10 protein PML in mediating intrinsic immunity against human cytomegalovirus infections. J Virol 80:8006-18.

172. Taylor, R. T., and W. A. Bresnahan. 2005. Human cytomegalovirus immediate-early 2 gene expression blocks virus-induced beta interferon production. J Virol 79:3873-7.

173. Taylor-Wiedeman, J., J. G. Sissons, L. K. Borysiewicz, and J. H. Sinclair. 1991. Monocytes are a major site of persistence of human 170

cytomegalovirus in peripheral blood mononuclear cells. J Gen Virol 72 ( Pt 9):2059-64.

174. Taylor-Wiedeman, J., P. Sissons, and J. Sinclair. 1994. Induction of endogenous human cytomegalovirus gene expression after differentiation of monocytes from healthy carriers. J Virol 68:1597-604.

175. Tomasec, P., E. C. Wang, A. J. Davison, B. Vojtesek, M. Armstrong, C. Griffin, B. P. McSharry, R. J. Morris, S. Llewellyn-Lacey, C. Rickards, A. Nomoto, C. Sinzger, and G. W. Wilkinson. 2005. Downregulation of natural killer cell-activating ligand CD155 by human cytomegalovirus UL141. Nat Immunol 6:181-8.

176. Wang, Q. Y., C. Zhou, K. E. Johnson, R. C. Colgrove, D. M. Coen, and D. M. Knipe. 2005. Herpesviral latency-associated transcript gene promotes assembly of heterochromatin on viral lytic-gene promoters in latent infection. Proc Natl Acad Sci U S A 102:16055-9.

177. Wang, X. Q., and J. A. Rothnagel. 2004. 5'-untranslated regions with multiple upstream AUG codons can support low-level translation via leaky scanning and reinitiation. Nucleic Acids Res 32:1382-91.

178. Warming, S., N. Costantino, D. L. Court, N. A. Jenkins, and N. G. Copeland. 2005. Simple and highly efficient BAC recombineering using galK selection. Nucleic Acids Res 33:e36.

179. Wilkinson, G. W., P. Tomasec, R. J. Stanton, M. Armstrong, V. Prod'homme, R. Aicheler, B. P. McSharry, C. R. Rickards, D. Cochrane, S. Llewellyn-Lacey, E. C. Wang, C. A. Griffin, and A. J. Davison. 2008. Modulation of natural killer cells by human cytomegalovirus. J Clin Virol 41:206-12.

180. Wills, M. R., O. Ashiru, M. B. Reeves, G. Okecha, J. Trowsdale, P. Tomasec, G. W. Wilkinson, J. Sinclair, and J. G. Sissons. 2005. Human cytomegalovirus encodes an MHC class I-like molecule (UL142) that functions to inhibit NK cell lysis. J Immunol 175:7457-65.

181. Wing, B. A., and E. S. Huang. 1995. Analysis and mapping of a family of 3'-coterminal transcripts containing coding sequences for human cytomegalovirus open reading frames UL93 through UL99. J Virol 69:1521-31.

182. Woodhall, D. L., I. J. Groves, M. B. Reeves, G. Wilkinson, and J. H. Sinclair. 2006. Human Daxx-mediated repression of human cytomegalovirus gene expression correlates with a repressive chromatin 171

structure around the major immediate early promoter. J Biol Chem 281:37652-60.

183. Wright, E., M. Bain, L. Teague, J. Murphy, and J. Sinclair. 2005. Ets-2 repressor factor recruits histone deacetylase to silence human cytomegalovirus immediate-early gene expression in non-permissive cells. J Gen Virol 86:535-44.

184. Xia, X., and M. Holcik. 2009. Strong eukaryotic IRESs have weak secondary structure. PLoS One 4:e4136.

185. Xuan, B., Z. Qian, E. Torigoi, and D. Yu. 2009. Human cytomegalovirus protein pUL38 induces ATF4 expression, inhibits persistent JNK phosphorylation, and suppresses endoplasmic reticulum stress-induced cell death. J Virol 83:3463-74.

186. Yamagiwa, Y., C. Marienfeld, F. Meng, M. Holcik, and T. Patel. 2004. Translational regulation of x-linked inhibitor of apoptosis protein by interleukin-6: a novel mechanism of tumor cell survival. Cancer Res 64:1293-8.

187. Yaman, I., J. Fernandez, H. Liu, M. Caprara, A. A. Komar, A. E. Koromilas, L. Zhou, M. D. Snider, D. Scheuner, R. J. Kaufman, and M. Hatzoglou. 2003. The zipper model of translational control: a small upstream ORF is the switch that controls structural remodeling of an mRNA leader. Cell 113:519-31.

188. Yamasaki, S., and P. Anderson. 2008. Reprogramming mRNA translation during stress. Curr Opin Cell Biol 20:222-6.

189. Yu, D., H. M. Ellis, E. C. Lee, N. A. Jenkins, N. G. Copeland, and D. L. Court. 2000. An efficient recombination system for chromosome engineering in Escherichia coli. Proc Natl Acad Sci U S A 97:5978-83.

190. Zaia, J. A., J. G. Sissons, S. Riddell, D. J. Diamond, M. R. Wills, A. J. Carmichael, M. P. Weekes, M. Gandhi, C. La Rosa, M. Villacres, S. Lacey, S. Markel, and J. Sun. 2000. Status of Cytomegalovirus Prevention and Treatment in 2000. Hematology (Am Soc Hematol Educ Program):339-355.

191. Zhang, L., and J. S. Pagano. 1997. IRF-7, a new interferon regulatory factor associated with Epstein-Barr virus latency. Mol Cell Biol 17:5748- 57. 172

192. Zuker, M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31:3406-15.

193. Zuker, M., and A. B. Jacobson. 1995. "Well-determined" regions in RNA secondary structure prediction: analysis of small subunit ribosomal RNA. Nucleic Acids Res 23:2791-8.