<%BANNER%>

Pathogenesis of Vaccinia Virus Attributed to Viral Genes E3L and K3L in the Control of the Host Interferon Response Gene...

Permanent Link: http://ufdc.ufl.edu/UFE0021991/00001

Material Information

Title: Pathogenesis of Vaccinia Virus Attributed to Viral Genes E3L and K3L in the Control of the Host Interferon Response Genes PKR and RNaseL
Physical Description: 1 online resource (181 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: e3l, host, interaction, k3l, pathogen, pkr, poxvirus, rnasel, vaccinia
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The interactions between the vaccinia virus (VV) genes E3L and K3L and their predicted host immune modulator targets, PKR (RNA dependent protein kinase) and RNaseL, have been investigated using both knockout mice and deletion viruses. The importance of the OAS/RNaseL and PKR pathways in response to double stranded RNA (dsRNA) within a host has been well documented. The virally encoded dsRNA binding protein, E3L, and PKR pseudo substrate, K3L, have been reported to target these host pathways specifically and reported to prevent the induction of the dsRNA induced interferon response. To determine the importance of these host pathways in controlling VV infection, single mouse knockouts of RNaseL and PKR and a double knockout of PKR/RNaseL (DKO) were studied using an intratracheal inoculation of VV. Animals were examined for clinical symptomology, virus dissemination, and histological findings of the lung, liver and spleen. VV caused lethal disease in all the mouse constructs. The single knockout animals were 10 times more susceptible while the DKO mice were 100 times more susceptible. VV?E3L was determined to be nonlethal in wild type mice, suggesting that E3L plays a critical role in controlling the host immune response. Lethal disease was however observed in DKO mice inoculated with 108 pfu, exhibiting a distinct pathology from that seen with a wild type VV infection. Wild type and the RNaseL single knockout mice did not exhibit severe disease while 20% of the PKR single knockout mice exhibited lethal disease at a dose of 108 pfu. VV?K3L exhibited no differences in virulence among any of the mouse constructs, suggesting that PKR is not the exclusive target of K3L. It was concluded that K3L was involved in virus spread from the lung based the on the lack of virus dissemination and histological findings.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Moyer, Richard W.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021991:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021991/00001

Material Information

Title: Pathogenesis of Vaccinia Virus Attributed to Viral Genes E3L and K3L in the Control of the Host Interferon Response Genes PKR and RNaseL
Physical Description: 1 online resource (181 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: e3l, host, interaction, k3l, pathogen, pkr, poxvirus, rnasel, vaccinia
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The interactions between the vaccinia virus (VV) genes E3L and K3L and their predicted host immune modulator targets, PKR (RNA dependent protein kinase) and RNaseL, have been investigated using both knockout mice and deletion viruses. The importance of the OAS/RNaseL and PKR pathways in response to double stranded RNA (dsRNA) within a host has been well documented. The virally encoded dsRNA binding protein, E3L, and PKR pseudo substrate, K3L, have been reported to target these host pathways specifically and reported to prevent the induction of the dsRNA induced interferon response. To determine the importance of these host pathways in controlling VV infection, single mouse knockouts of RNaseL and PKR and a double knockout of PKR/RNaseL (DKO) were studied using an intratracheal inoculation of VV. Animals were examined for clinical symptomology, virus dissemination, and histological findings of the lung, liver and spleen. VV caused lethal disease in all the mouse constructs. The single knockout animals were 10 times more susceptible while the DKO mice were 100 times more susceptible. VV?E3L was determined to be nonlethal in wild type mice, suggesting that E3L plays a critical role in controlling the host immune response. Lethal disease was however observed in DKO mice inoculated with 108 pfu, exhibiting a distinct pathology from that seen with a wild type VV infection. Wild type and the RNaseL single knockout mice did not exhibit severe disease while 20% of the PKR single knockout mice exhibited lethal disease at a dose of 108 pfu. VV?K3L exhibited no differences in virulence among any of the mouse constructs, suggesting that PKR is not the exclusive target of K3L. It was concluded that K3L was involved in virus spread from the lung based the on the lack of virus dissemination and histological findings.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Moyer, Richard W.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021991:00001


This item has the following downloads:


Full Text

PAGE 1

1 PATHOGENESIS OF VACCINIA VIRUS ATTR IBUTED TO VIRAL GENES E3L AND K3L IN THE CONTROL OF THE HOST INTERFE RON RESPONSE GENES PKR AND RNASEL By AMANDA DAWN RICE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Amanda Dawn Rice

PAGE 3

3 To my family, thank you for your unending support

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my me ntor, Dr. Richar d Moyer for guidance and support over the last five years. He has provided me with a unique environment in which to learn and work that continuously challenged me scientifically. I also would like to thank my committee members Dr. David Bloom, Dr. Mavis McKe nna, and Dr. Lyle Moldawer, who with Dr. Moyer, have ensured my success as a researcher and o ffered guidance at every step of the way. The members of the Moyer lab have made coming to work each day an adventure. I can honestly say that I have learned something from every person that I have had the pleasure of knowing. The largest thank you goe s to Michael Duke, Dorothy Smith, Andrew Smith, JoAnne Anderson, Tommie Albright, and Elizabeth White th at have helped with the mouse infections and sample processing in this study. My mouse breeding technical support from Dorothy Smith and Andrew Smith was invaluable Michael Duke has tittered mouse tissues in the thousands and even came out of retirement to help get th em all donethank you just does not suffice. All the technical support I have received throughout my research has been truly amazing and made the long days go much faster and smoother. The members of the departmental administra tive and fiscal staff have always provided excellent support to ensure that science can progress with as few administrative problems as possible. A special thank you goes to Michele Ramsey and Joyce Conners whose friendships and support have made the process of science an d life as a graduate student easier. Connie Philebaum or my filler mom has helped me to maintain sanity for the last five years and deserves a huge thank you. The microarray portion of my dissertation, alth ough not large in size took many people to complete. Dr. Moldawer, Mathew Delano, Cynthia Tannahill, Dr. Henry Baker and Cecilia Lopez all worked on the microarray section of th is study from sample preparation to teaching me

PAGE 5

5 to analyze the data. Jennifer Embry, DVM, not only taught me to read my histology slides but provided superior help whenever needed. Dr. Robert Silverman and Dr. Bryan Williams provided the foundation breeding mice utilized for this experiment. Dr. Silverman also aided in the early experimental design and provided invaluable background on the mice constructs. Finally, the biggest thank you goe s to my family. They have supported me in more ways than I can count or remember, from staying wi th my daughter for a week during large animal experiments to reminding me that I would even tually finish. Thanks Mom and Dad. My husband, Brian, has been my source for une nding support and motivation. My daughter, Jasmine, reminds me daily of what in life is r eally important. Graduate school would not have been the same without Brian and Jasmine; they have made this accomplishment mean that much more.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................9LIST OF FIGURES .......................................................................................................................10ABSTRACT ...................................................................................................................... .............16 CHAPTER 1 INTRODUCTION AND BACKGROUND ...........................................................................18Introduction .................................................................................................................. ...........18Interferon .................................................................................................................... ............19Types of Interferon ..........................................................................................................19Type I Interferon Receptor Signaling .............................................................................. 20Activation of Type I Interferons ......................................................................................20RNA Dependent Protein Kinase (PKR) ..........................................................................222-Oligoadenylate Synthetase (OAS) and RNaseL ...................................................... 24Poxviruses .................................................................................................................... ...........25Genomic Structure ...........................................................................................................26Poxvirus Life Cycle .........................................................................................................27Vaccinia Virus Infection of Mice .................................................................................... 29Control of Host Immune Responses by Poxviruses ........................................................ 32Vaccinia virus control of innate immune response ............................................... 32Vaccinia virus control of interferon ....................................................................... 33Vaccinia Virus Gene E3L ................................................................................................34Vaccinia Virus Gene K3L ...............................................................................................35Study Objectives .....................................................................................................................362 MATERIALS AND METHODS ...........................................................................................47Tissue Culture and Viro logical Techniques ........................................................................... 47Tissue Culture ..................................................................................................................47Growth of Virus Stocks for Animal Injections ................................................................ 47Plaquing of Virus .............................................................................................................48Generation of Virus Mutants and Rec onstructed Wild type Revertants ......................... 49DNA Techniques ................................................................................................................ ....50Cloning of Wild Type E3L and K3L Genes .................................................................... 50Cloning of E3L::gfp Deletion Fragment .........................................................................50Cloning of K3L::gfp Deletion Fragment .........................................................................52Purification of Viral DNA ............................................................................................... 53Sequencing of Viruses ..................................................................................................... 53PCR ..................................................................................................................................53

PAGE 7

7 RNA Techniques ................................................................................................................ ....54RNA Isolation from Lung Tissue ....................................................................................54RNA Processing and Microarrays ................................................................................... 55Microarray Data Analysis ................................................................................................ 55Animal Techniques .................................................................................................................56Mouse Breeding and Line Maintenance ..........................................................................56Mouse Infections .............................................................................................................57Monitoring of Infected Animals ...................................................................................... 57Processing of Tissues for Titering ...................................................................................58Interferon Beta ELISA Assays ........................................................................................58Histological Methods ..............................................................................................................59Tissue Processing ............................................................................................................59H&E Staining ..................................................................................................................59Anti-gfp Immunostaining ................................................................................................59Anti-STAT-1 Immunostaining ........................................................................................60Analysis of Slides and Photography ................................................................................ 613 PATHOLOGY OF VACCINIA VIRU S IN WILD TYPE MICE .......................................... 65Introduction .................................................................................................................. ...........65Results .....................................................................................................................................66Survival Studies of Wild Type Mice Infected with Vaccinia Virus ................................ 66Clinical Symptoms in Wild Type Mice Infected with Vaccinia Virus ...........................67Induction of Interferon Synthesis in Infected Lung Tissue ..........................................68Virus Dissemination from the Site of Inoculation ...........................................................69Microarray Analysis of Vaccini a Virus Infected Lung Tissue ........................................ 70STAT-1 Staining of Lung Tissue ....................................................................................74Histopathology of Infected Mice .....................................................................................75Discussion .................................................................................................................... ...........764 PATHOLOGY OF VACCINIA VI RUS IN KNOCKOUT MICE ......................................... 94Introduction .................................................................................................................. ...........94Results .....................................................................................................................................95Survival Studies of Knockout Mice Infected with Vaccinia Virus ................................. 95Clinical Symptoms in Knockout Mi ce Infected with Vaccinia Virus ............................. 96Induction of Interferon Synthesis in Infected Lung Tissue ..........................................97Virus Dissemination from the Site of Inoculation ...........................................................98STAT-1 Staining of Lung Tissue ....................................................................................99Histopathology of Infected Mice ...................................................................................100Discussion .................................................................................................................... .........1025 PATHOLOGY OF VV E3L::GFP IN WILD TYPE AND KNOCKOUT MICE ............... 115Introduction .................................................................................................................. .........115Results ...................................................................................................................................116In Vitro Data of VV E3L .............................................................................................. 116

PAGE 8

8 Rescue of VV E3L::gfp ................................................................................................ 116Survival Studies of Mice Infected with VV E3L::gfp ................................................. 117Clinical Symptoms in Mice Infected with VV E3L::gfp ............................................. 119Virus Dissemination from the Site of Inoculation ......................................................... 120STAT-1 Staining of Lung Tissue ..................................................................................121Histopathology of Infected Mice ...................................................................................123Discussion .................................................................................................................... .........1266 PATHOLOGY OF VV K3L::GFP IN WILD TYPE AND KNOCKOUT MICE .............. 144Introduction .................................................................................................................. .........144Results ...................................................................................................................................144In Vitro Data .................................................................................................................. 144Survival Studies of Mice Infected with VV K3L::gfp ................................................. 145Clinical Symptoms in Mice Infected with VV K3L::gfp ............................................. 146Virus Dissemination from the Site of Inoculation ......................................................... 148STAT-1 Staining of Lung Tissue ..................................................................................150Histopathology .............................................................................................................. 150Discussion .................................................................................................................... .........1537 CONCLUSIONS AND DISCUSSION ................................................................................ 171Overall Conclusions ..............................................................................................................171Final Thoughts ......................................................................................................................172LIST OF REFERENCES .............................................................................................................174BIOGRAPHICAL SKETCH .......................................................................................................181

PAGE 9

9 LIST OF TABLES Table page 1-1 Poxvirus classifications ......................................................................................................431-2 Vaccinia virus encoded genes that control the host immune system ................................. 463-2 Summary of Survival Data for Wild Type Mice Infected with Vaccinia Virus ................793-1 Clinical scoring criteria for poxvirus infected mice. ......................................................... 823-3 Virus spread of VV in wild type mice ...............................................................................843-4 Expression values for differentially expre ssed genes in the IFN activation pathways. ..... 894-1 Virus spread of VV in all mouse constructs. ................................................................... 1095-1 In vitro titer data VV E3L::gfp host range ..................................................................... 1305-2 Tissue titers of mice infected with VV E3L::gfp ........................................................... 1365-3 Tissue titers for DKO mice infected with 109 pfu VV E3L::gfp ....................................1376-1 In vitro titer data VV K3L::gfp host range .....................................................................1576-2 Titer of tissues from all mous e constructs infected with VV LD80 doses of VV K3L::gfp. .................................................................................................................. 1636-3 Titer of tissues for all mous e constructs infected with VV K3L::gfp 106 pfu ................ 164

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Interferon signaling through the Type I IFN receptor. ...................................................... 391-2 Response to dsRNA within a cell ......................................................................................401-3 PKR activation ............................................................................................................ .......411-4 OAS/ RNaseL activation ....................................................................................................421-5 Poxvirus life cycle..............................................................................................................441-6 Infection of Mice. ........................................................................................................ .......452-1 Generation of VV E3L::gfp .............................................................................................. 622-2 Generation of VV K3L::gfp ............................................................................................. 632-3 Mouse genotyping. .............................................................................................................643-1 Survival curves for wild type mice following infection with VV .....................................783-2 Average body temperature and weight loss curves of wild type mice following infection with VV. .............................................................................................................803-3 Clinical progression of VV disease in mice ....................................................................... 813-4 Interferon beta levels in the lung tissue of VV infected wild type mice ............................833-5 Global expression profile changes in infected lung tissue of differentially expressed probe sets ...........................................................................................................................853-6 Expression profile of B cell specific genes ........................................................................ 863-7 Expression profile of T-cell specific genes ........................................................................ 873-8 Expression profile of probe sets iden tified as members of the immune response pathway. ...................................................................................................................... .......883-9 Immunohistochemitry staining of STAT-1 protein in wild type mouse lung tissue.......... 903-10 Immunohistochemistry and H&E staining of wild type mouse VV infected lung tissue ........................................................................................................................ ..........913-11 Immunohistochemistry and H&E staini ng of liver and spleen tissue from VV infected wild type mice ...................................................................................................... 92

PAGE 11

11 3-12 Tissue weights for VV infected wild type mice ................................................................. 934-1 Survival curves of knockout mouse constructs with wild type VV ................................. 1044-2 Comparison of survival of a ll mouse constructs with VV. ..............................................1054-3 Average body temperature and weight lo ss of RNaseL mice infected with VV ............. 1064-4 Average body temperature and weight loss of PKR mice infected with VV ..................1074-5 Average body temperature and weight loss of DKO mice infected with VV ................. 1084-6 Interferon beta levels in th e lung tissue of infected mice ................................................ 1104-7 Immunohistochemistry staining of STAT-1 protein in lung tissue of infected animals at day 3 and 5 post infection from DKO mice ................................................................. 1114-8 Immunohistochemistry and H&E staining of infected lung tissue from DKO mice .......1124-9 Immunohistochemistry and H&E staini ng of liver tissue from VV infected DKO mice ..................................................................................................................................1134-10 Tissue weights of mi ce infected with VV ........................................................................1145-1 Diagram of E3L protein and area deleted in VV E3L::gfp ............................................ 1295-2 Survival Curves for mice infected with VV E3L::gfp .................................................... 1315-3 Average body temperature and weight loss of wild type mice infected with VV E3L::gfp. .................................................................................................................. 1325-4 Average body temperature and wei ght loss of DKO mice infected with VV E3L::gfp. .................................................................................................................. 1335-5 Average body temperature and weight loss of RNaseL mice infected with VV E3L::gfp. .................................................................................................................. 1345-6 Average body temperature and wei ght loss of PKR mice infected with VV E3L::gfp. .................................................................................................................. 1355-7 Protein levels of STAT-1 in VV E3L::gfp infected lung tissue .....................................1385-8 VV E3L::gfp histology of wild type mouse lung tissue ................................................. 1395-9 Histology of DKO mouse lung tissue from animals infected with 104 pfu VV E3L::gfp ................................................................................................................... 1405-10 Histology of DKO mouse lung tissue from animals infected with 108 pfu VV E3L::gfp ................................................................................................................... 141

PAGE 12

12 5-11 Histopathology of spleen and liver tissue from VV E3L::gfp infected animals ............ 1425-12 Tissue weights for mice infected with VV E3L::gfp ...................................................... 1436-1 K3L protein domain diagram ...........................................................................................1566-2 Survival data for mice infected with VV K3L::gfp ........................................................ 1586-3 Average body temperature and weight loss of wild type mice infected with VV K3L::gfp. ................................................................................................................. 1596-4 Average body temperature and weight loss of RNaseL mice infected with VV K3L::gfp. ................................................................................................................. 1606-5 Average body temperature and wei ght loss of PKR mice infected with VV K3L::gfp. ................................................................................................................. 1616-6 Average body temperature and wei ght loss of DKO mice infected with VV K3L::gfp. ................................................................................................................. 1626-7 Staining of STAT-1 from VV K3L::gfp infected wild type mice .................................. 1656-8 Staining of STAT-1 from VV K3L::gfp infected DKO mice. .......................................1666-9 Histopathology of VV K3L::gfp infected wild type mouse lung tissue .........................1676-10 Histopathology spleen tissue from mice infected with VV K3L::gfp. ........................... 1686-11 Histopathology of VV K3L::gfp infected DKO mouse lung tissue ............................... 1696-12 Tissue weights for mice infected with VV K3L::gfp. .................................................... 170

PAGE 13

13 LIST OF ABBREVIATIONS M Micromolar 2-5A 2-5 linked oligoadenylates BHK-21 Baby hamster kidney cell line bp Base pairs cDNA Complementary dexoyribonucleic acid CEF Chicken embryo fibroblasts CEV Cell associated enveloped virion CPE Cytopathic effect CPV Cowpox virus cRNA Complementary ribonucleic acid CTL Cytotoxic Tlymphocytes CV-1 Green monkey kidney cell line DNA Deoxyribonucleic acid dNTP Deoxynucleotide triphosphate dsRNA Double stranded RNA EEV Extracellular enveloped virion ER Endoplasmic reticulum FBS Fetal bovine serum IC Intracranial ID Intradermal IEV Intracellular enveloped virion IFN Interferon IKK Inhibitor of NF kinase IMV Intracellular mature virion

PAGE 14

14 IN Intranasal IP Intraperitoneal IPS1 IFN promoter stimulator 1 IT Intratracheal ITR Inverted terminal repeats kbp Kilobase pair MDA5 Melanoma differentiatio n associated protein 5 MHC Major histocompatibility complex mL Milliliter mM Milimolar mm Millimeter MOI Multiplicity of infection NF Nuclear factor NK Natural killer cell OAS 2-5 Oligoadenylate Synthase PBS Phosphate buffered saline PCR Polymerase chain reaction pfu Plaque forming units PK-15 Pig kidney cell line PK-15 Pig kidney cell line PKR RNA dependent protein kinase RIG-1 Retinoic acid-inducible gene 1 RNA Ribonucleic acid RNase Endoribonuclease RPV Rabbitpox virus

PAGE 15

15 STAT Signal transducers and activators of transcription TBK1 TANK binding kinase TLR Toll like receptor TRIF TLR adaptor molecule 1 VCP Vaccinia complement control protein VV Vaccinia virus wt Wild type

PAGE 16

16 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PATHOGENESIS OF VACCINIA VIRUS ATTR IBUTED TO VIRAL GENES E3L AND K3L IN THE CONTROL OF THE HOST INTERFE RON RESPONSE GENES PKR AND RNASEL By Amanda Dawn Rice May 2008 Chair: Richard W. Moyer Major: Medical Sciences-Genetics The interactions between the vaccinia virus (VV) genes E3 L and K3L and their predicted host immune modulator targets, PKR (RNA dependent protein kinase) and RNaseL, have been investigated using both knockout mice and de letion viruses. The importance of the OAS/RNaseL and PKR pathways in response to double stranded RNA (dsRNA) within a host has been well documented. The virally encode d dsRNA binding protei n, E3L, and PKR pseudo substrate, K3L, have been reported to target these host pathways specifically and reported to prevent the induction of the dsRNA induced interferon response. To determine the importance of these host pathways in c ontrolling VV infection, single mouse knockouts of RNaseL and PKR and a double knockout of PKR/RNaseL (DKO) were studied using an intr atracheal inoculation of VV. Animals were examined for clin ical symptomology, virus dissemination, and histological findings of the lung, liver and spleen. VV caused lethal disease in all the mouse constructs. The single knockout animals were 10 times more susceptible wh ile the DKO mice were 100 time s more susceptible. VV E3L was determined to be nonlethal in wild type mice, suggesting that E3L play s a critical role in controlling the host immune response. Lethal disease was however observed in DKO mice inoculated with 108 pfu, exhibiting a distinct pathology fr om that seen with a wild type VV infection. Wild type and the RNaseL single kn ockout mice did not exhibit severe disease while

PAGE 17

17 20% of the PKR single knockout mice exhib ited lethal diseas e at a dose of 108 pfu. VV K3L exhibited no differences in viru lence among any of the mouse cons tructs, suggesting that PKR is not the exclusive target of K3L. It was conclu ded that K3L was involved in virus spread from the lung based the on the lack of virus dissemination and hist ological findings.

PAGE 18

18 CHAPTER 1 INTRODUCTION AND BACKGROUND Introduction The struggle between a pathoge n and its host is a delicate ba lance in which the pathogen attempts to subvert the host imm une system while the host attempts to block the infection from progressing. If successful the pathogen c ontrols the host immune system long enough to replicate and spread to anothe r host. The ability of a pa thogen to subvert the host is accomplished in a number of various ways from altering their membrane protein profiles as observed for some bacterial pathogens to encodi ng genes that target specific host response pathways. There are two defined defense responses that a host mounts against a pathogen, the innate and adaptive immune responses. The innate immune response is char acterized by a general response that is a non-specific cell mediated response to the re cognition of a specific pathogen triggered by the presence of cer tain sugar residues, double strande d RNA or other insult to the host organism. The adaptive immune response is focused on the specific pathogen and results in the production of antibodies against the offending pathogen. The innate immune response is the first response of a host that attempts to contro l the infection until the adaptive immune response has had time to get underway. The adaptive immune response takes seven to ten days to become fully activated. As part of the innate response, the induction of interfer on is a major innate response pathway that when activated up regulat es the production of not only interferon but many additional genes involved in host defense. One such way to activate the interferon response is through the presence of double stranded RNA (dsRNA) that is often present in virus infections.

PAGE 19

19 Poxviruses are viruses that infect a wide vari ety of organisms and are known to control the host immune system by encoding multiple proteins to control the host immune response and virus detection. The innate immune response has been identified as critical in controlling a poxvirus infection, where the cell mediated response is the major determinant of virus clearance in nave hosts. Interferon Interferons were identified in 1957 and are now known to play a role in cell death and tumor development, enhance imm une responses, a nd regulate resistance to viral infection.(Isaacs & Lindenmann, 1957) Currently there are multiple FDA approved therapies that use interferon to treat human diseases including multiple sclerosis and leukemia.(Maher, Romero-Weaver et al., 2007) Of particular interest to this study is the regulation of cellular resistance to viral infections. While controlling virus infections involve both the inna te and adaptive immune responses, the innate, or first level of defense of an organism against viral infections, is the most critical. (Haga & Bowie, 2005) Types of Interferon There are three classes of interferon, Type I, II, and III. Typ e I interferons are comprised of IFN IFN and IFN .(Samuel, 2001) INF is encoded by 13 genes in the mouse and secreted by leukocytes. IFN is encoded by a single gene in the mouse and secreted by all cells.(Takoaka & Yanai, 2006) Mi ce do not have a functional IFN gene. Type II interferon is comprised of a single member, IFN which is a part of the adaptive immune response and secreted by activated T-cells and natural killer ce lls. The Type III interferon family is comprised of three members, IFN 1 to 3. Type III interferons have not been well studied to date, but respond to infection in a manner si milar to that of T ype I interferons. (Takoaka & Yanai, 2006)

PAGE 20

20 For this study we will focus on the innate immune response, and therefore the pathways involved with Type I interferons. Type I Interferon Receptor Signaling Type I interferons share a heterodimeri c receptor encoded by the genes IFNAR1 and IFNAR2 present on all cells.(Randall & Goodbourn, 2008) T he extracellular component of the receptor is responsible for the binding of IFN / and signaling via the cytoplasmic tails. Signaling is separate and prot ein specific through the 2 diffe rent cytoplasmic tails. The cytoplasmic tail of IFNAR1 binds Tyk2 while th e cytoplasmic tail of IFNAR2 binds JAK-1 and STAT-1 and STAT-2. STAT -1 and STAT-2 are members of the signal transducers and activators of transcription family of transcri ption factors known to be important in IFN responses. Upon activation of the recep tor, both kinases Tyk2 and JAK-1 undergo phosphorylation. Tky2 then phosphorylates STAT-2 while STAT-1 is phosphorylated by JAK1. The phosphorylated STAT-1 and STAT-2 und ergo conformational changes that initiate a strong heterodimer formation. This STAT-1/2 heterodimer possesses a nuclear localization signal for trafficking to and sequestering within the nucleus. The heterodimer is then able to bind IRF9 (interferon stimulated transcription factor) and theref ore become a transcriptional enhancer for interferon stimulated genes (ISG) by binding to IFN stimulated response elements (ISRE) within the promoters of interferon response genes.(Kat ze, He et al., 2002;Takoaka & Yanai, 2006) (Figure 1-1) Activation of Type I Interferons Interferon is produced in response to ma ny stresses and recognition of pathogenic pattern associated molecular patterns (PAMPs)(Haller, Kochs et al., 2007). One of the ways to activate IFN production is via toll like receptor (TLR) signaling. There are twelve types of TLRs that respond to a wide variety of bacterial, fungal, para sitic, and viral PAMPs. Those that detect the

PAGE 21

21 presence of viral infection ar e all found intracellular and incl ude TLR9 that responds to CpG (DNA sequences that have a methylated cyto sine followed by guanine nucleotide) DNA; TLR7 and TLR8 that are activated in respond to ssR NA (single stranded RNA); and TLR3 that is activated in response to dsRNA.(Randa ll & Goodbourn, 2008) Poxviruses have dsDNA genomes and therefore do not activate TLR7 or TLR8, do not have CpG DNA that activates TLR9, but do produce dsRNA (Colby & Duesber g, 1969;Duesberg & Colby, 1969) that has the capacity to activate TLR3(Harte, Haga et al., 2003). Signaling through TLR3 is complex with multiple transcription factors activated. TLR3 is typically found in an endosome where dsRNA binds to the receptor causing phosphorylation of the receptor, dimerizatio n, and binding to CD14. CD14 is a membrane protein that has been identified as a pattern recognition receptor responding strong ly to LPS. This allows the signaling pathway to become activated, and is initiated by TLR adaptor molecule 1 (TRIF). TRIF activates multiple pathways including the activ ation of IRF3 via TRAF3 and TBK1, and the activation of NF via the activation of TR AF6/TAK1 complex and I phosphorylation by IKK. The TRAF6/TAK1 complex also activates JUN by JNK (c-Jun N-terminal kinases activated in times of stress) activation and ATF2 (activating transcription factor 2 is a leucine zipper family of DNA binding proteins) by p38 (mit ogen activated protein kinase activated under cellular stress) activation. The activation of th e TLR3 pathway ends in the transport of IRF3, NF JUN, and ATF2 transport to the nucleus wher e these proteins serve as promoter specific transcription factors that transcribe inte rferon response genes (I SGs).(Randall & Goodbourn, 2008) Figure 1-2 shows the pathway diagram. Response to free intracellular dsRNA causes the activation of multiple pathways. One pathway involves the cytoplasmic RNA helicas es RIG-1 and MDA5 that can bind dsRNA and

PAGE 22

22 signal through the same protein cascades as TLR3, using IPS1 as the adaptor protein rather than TRIF. IRF-3 (interferon regulat ory transcription factor) can also bind dsRNA independently inducing IRF-3 to become activated and tran scriptionally active. The binding of 2,5oligoadenylate synthetase (OAS) to dsRNA leads to activation of the OAS/RNaseL pathway that ends with the degradation of all the RNA within the cell. These degraded RNA fragments can then act as a positive feedback loop to fu rther amplify activation of RIG-1 or MDA5. RNA dependent protein kinase (PKR) is activated by the binding of dsRNA. Activated PKR then phosphorylates eIF2 (eukaryotic translation in itiation factor) leading to the inactivation of the eIF2 translational machinery and global cell transl ational shutdown. The relationships between these pathways and their signaling cascades are outlined in Figure 1-2.(Takoaka & Yanai, 2006;Borden, Sen et al., 2008;Randall & Goodbourn, 2008) All these responses to viral inf ection act to increase transcriptional regulation of a class of genes identified as ISGs. These ISGs are only actively transcribed in response to stress or infection and are the basis for the antiviral state. A positive feedback is also observed with the up regulation of IFNs. Cells in which the IF N response pathway is activated by intracellular detection of dsRNA rather than direct binding of IFN to the receptor are able to not only produce IFN to induce an antiviral state for neighboring cells but also enhance its own response by the produced IFN acting in an autocrine manner.(Bor den, Sen et al., 2008;Katze, He et al., 2002) RNA Dependent Protein Kinase (PKR) RNA dependent protein kinase (PKR) is a se rine threonine protein kinase encoded by a single gene in the mouse that is ubiquitously expressed in all ti ssues at low levels. PKR has 2 m ajor protein domains, the C-terminal serine /threonine kinase catalytic domain and the Nterminal dsRNA binding domain.(Ga le, Jr. & Katze, 1998;Katze, He et al., 2002) PKR is both a IFN production inducer and an IFN response up regulated protein, in that it both induces the IFN

PAGE 23

23 response in the presence of dsRN A and is transcriptionally up regulated in the presence of IFN.(Garcia, Meurs et al., 2007) PKR activation occurs after binding to dsRNA or RNA with complex secondary structure. (Figure 1-3) Upon binding to dsRNA PKR homodimerizes and undergoes autophosphorylation thereby b ecoming active. Active PKR acts to phosphorylate eIF2 thereby causing translation in the cell to halt. This is accomplished by increasing eIF2 s affinity for GDP and decreasing the levels of eIF2 -GTP-Met-tRNA required for translation initiation in the cell. PKR also phosphorylates IKK that in turn phosphorylates IkB (inhibitor of kappa B) which releases active NF The activated NF in turn up regulates transcription of apoptosis genes such as Fas, FasL, and p53. PKR also phosphor ylates p53 to enhance the transcriptional activation of p53. It also phosphorylates the apop totic proteins IRF-1, STAT-1, and NF-90 that in turn up regulate the transc ription of IFN inducible gene s.(Gale, Jr. & Katze, 1998) Cells derived from PKR knockout (KO) mice ha ve demonstrated a pa rtial inhibition of encephalomyocarditis virus (ECMV) replication, and no impact on vesicular stomatitis virus (VSV) or Vaccinia virus (VV) replication in vitro .(Yang, Reis et al., 1995;Gale, Jr. & Katze, 1998) These mice have shown an increased sensitivity to reovirus (Stewart, Blum et al., 2003), dugbe virus (Boyd, Fazakerley et al., 2006), encepha lomyocarditis virus (Zhou, Paranjape et al., 1999), vesicular stomatitis virus (Stojdl, Abra ham et al., 2000), hepatitis B virus (Guidotti, Morris et al., 2002), and bunyamwera virus (Streite nfeld, Boyd et al., 2003). PKR has also been implicated in functioning as a tumor suppressor in a mouse tumor transplant model in which the lack of functional PKR causes tumors to grow qu ickly. Infection with VV via the IN route of these knockout mice was previously reported to exhi bit responses identical to that of wild type mice.(Xiang, Condit et al., 2002a)

PAGE 24

24 2-Oligoadenylate Synthetase (OAS) and RNaseL The 2,5-oligoadenylate synthetase (OAS) a nd RNaseL proteins are a part of the interferon response innate response pathway that responds to dsRNA, also known as the 2-5A pathway. There are two main proteins in this pathway that respond to the presence of dsRNA, OAS and RNaseL proteins. There are 11 OAS encoding genes in the m ouse, 8 encoding OAS1, 1 encoding OAS2, 1 encoding OAS3, and 2 encodi ng OASL.(Kakuta S., Shibata et al., 2002) These different forms of the OAS protein are loca lized into different cellular locations, exhibit differential preference for dsRNA species, and gene rate 2-5-oligadenylates (2-5As) of varying lengths. The active forms of these proteins ar e also found with different numbers of 2-5As bound. OAS1 has a single catalytic domain and four OAS1 proteins bind to form homotetramers, OAS2 has 2 catalytic domains and forms homodimers, while OAS3 has 3 catalytic domains and is found as mono mers.(Silverman & SenGupta, 1990) RNaseL is found expressed ubiquitously in a ll cell types as an inactive monomer until activation. RNaseL has both endoribonuclease and phosphodieste rase activities. The Nterminus of the protein contains ankyrin repeats allowi ng for protein-protein interactions and the binding of 2-5As. The kinase a nd RNase domains of the protein are located in the C-terminal portion of the protein.(Samuel, 2001;Silverman, 2007) The 2-5A pathway is activated by the binding of dsRNA or RNA with secondary structure by OAS. OAS then uses ATP to generate 2-5 li nked oligoadenylates (2-5A) of varying lengths depending on the OAS species involved. The 25As have the general formula ppA(2p5A)n (1 n 30) and are short lived in the cells as they undergo rapid degradation by cellular phosphodiesterases and 2-5-exoribonucleases; therefore the activati on of RNaseL by 2-5As is rapid and transient. These 2-5As are then bound by RNaseL whereby inducing homodimerization and subsequent activation of the RNaseL protein complex. The RNaseL

PAGE 25

25 homodimer is then able to hydrolyze the 2-5 oligonucleotides and us e the endoribonuclease activity to destroy the RNA present in the cell, thereby halting protein synthesis and inducing apoptosis.(Samuel, 2001) (Figure 1-4) RNaseL knockout mice exhibit an increased susc eptibility to viral infections including encephalomyocarditis virus (Zhou, Paranjape et al., 1999) and deficient in cellular apoptosis (Zhou, Paranjape et al., 1997;Samuel, 2001). VV in fection of these knockout mice via the IN route was reported to exhibit normal responses as compared to wild type mice with no increase in sensitivity.(Xiang, Condit et al., 2002a) Poxviruses Poxviruses have been of global importance historically as both hum an and animal pathogens. The concern of variola, the causative agent of small pox, being intenti onally released and the well publicized accidental infections of humans with monkeypox virus has increased the desire to understand the relations hip of poxvirus genes to their function during an infection that leads to pathogenic infection. This desire to understand the host-virus in teraction has been the focus of this study, focusing on the host detectio n of dsRNA in response to viral infection and the genes in the virus that control these responses. Poxvirus virions are large, brick shaped particles approxima tely 300nm by 270nm in size, making them one of the largest virus particles. The virion s contain an envelope, outer membrane, nucleosome containing viral DNA, and 2 lateral bodies. See figure 1-5 for diagram of virion. Research has yet to determine what composes the lateral bodies and the entire composition of the membranes surrounding the virion. The DNA of Poxviruses is double stranded and these complex viruses replicate exclusively within the cytoplasm of cells.(Moss, 1996;Moss, 1996)

PAGE 26

26 In addition to Variola, the poxvirus family incl udes a wide variety of viruses that infect mammals, birds, and insects. (Table 1-1) The members of the Chordopoxvirinae infect vertebrates, while the Entomopoxvirinae members infect invertebrates. The poxviruses that infect mammals include the Orthopoxvirus, Leporipoxvirus, Parapoxvi rus, Capripoxvirus, Molluscipoxvirus, Suipoxvirus, and Yatapoxvirus generas. Th e Orthopoxvirus genus is very large and includes vaccinia, variola (smallpox), rabbitpox, and cow pox virus. These viruses are able to infect most mammalian cells and are ca pable of causing system ic disease in humans. Vaccinia virus was utilized historically as a vaccine to eradicate smallpox in humans. Currently the major use for VV is in the labo ratory as the prototypic orthopoxvirus for both in vitro and in vivo studies. It was the first animal viru s to be successfully grown, characterized, tittered, and observed by microscopy. The exact origin of VV is unclear, but it has been hypothesized that it arose from the seri al passage of smallpox vaccine. Within the classification of VV there are many different virus strains that are identified by their origin of isolation. These isolates exhibit different phenot ypes in animal models, with little sequence variation. It has also been shown that ev en within a specific strain cultured in different laboratories there can be large difference in anim al pathogenesis.(Adams, Ri ce et al., 2007) Two different laboratory strains of Vaccinia virus We stern Reserve (VV-WR) exhibit different levels of virulence in rabbits. For this reason a single laboratory isolate was ut ilized in these studies and all the preliminary studies in animals had to be performed to obtain a baseline from which all comparisons are made. Genomic Structure Poxviruses have large genomes ranging from 130 to 300kbp that encode 136 to 260 open reading frames transcribed in both directions. Vaccinia v irus, the prototypic orthopoxvirus, is 190kbp in size encoding 185 genes. Within the la st 10 years many poxviruses have been fully

PAGE 27

27 sequenced and annotated. This advance in genomics for the virus has allowed for in silico comparisons between many species of poxviruses a nd the elucidation of gene functions based upon sequence similarity to known gene functions. All poxvirus genomes have the same general layout of genes within the genome consisting of a middle conserved section and two outer more virus specific flanking regions that include the inverted terminal repeat s (ITRs).(Moss, 1996) All the open reading frames found in poxvirus genomes l ack introns, have shor t promoter sequences, and are relatively small in size.(Moss, 1990) The central most portion of the genome is highly conserved within each family, and encodes the genes required for re plication, transcription, and viru s assembly.(Upton, Slack et al., 2003) This core collection of ge nes is absolutely required for re plication in tissue culture. The flanking regions on each side of the conserved core exhibit the area of highest divergence between viruses from the same genera. These flanking regions encode the genes necessary for replication and disease in the animal model, but not tissue culture, enc oding genes that control host range, virus disseminati on, and immune modulation. Poxvirus Life Cycle The life cycle of poxviruses is characterized by entry, temporal expression of 3 classes of gene transcripts and virion assembly and rel ease.(Moss, 2001) The entry m echanism has not been fully elucidated; however necessary first steps are for the virus to attach to the cell, and fuse with the plasma membrane. This is thought to be accomplished in multiple steps that include the attachment to the cell and bindi ng of the virion entry complex of proteins with the cellular receptor. The virion then fuses with the outer membrane to release the virus core into the cytoplasm.(Moss, 2006) Once the virus core is rel eased into the cytoplasm it is then transported on microtubules into the cellu lar cytoplasm where it undergoes the primary uncoating and transcription and translation of early genes using virus encoded proteins. The cores undergo a

PAGE 28

28 secondary uncoating in which the viral DNA is becomes more accessible to allow for both intermediate gene expression and DNA replica tion.(Broyles, 2003) After DNA replication late gene products are made and virions begin to assemble in the cytoplasm within virus factories.(Condit, Moussatche et al., 2006;Schramm & Locker, 2005) Assembly is characterized by the appearance of crescent shaped particles as seen by electron microscopy and packaging of DNA into round shaped particles. The particles then mature into the typical brick shaped virions and are termed intracellular mature virions, IMV, a fully infectious form of the virus that has one outer membrane. The IMV particles then proceed through the golgi apparatus fo r the acquisition of two additional membranes becoming intracellular enveloped virions (IEV). The virions are then transported to the plasma membrane for fusion of the outermost virus membrane and th e plasma membrane to give cellular enveloped virus (CEV), that are bound to the outer membrane of the cell. It is the CEV that is responsible for spread of virus to adjacent ce lls, or locally. Actin tails are used to propel CEV virions off the cell surface and generate ex tracellular envelope virions (EEV). EEV is the form of the virus that is responsible for dissemination within an animal and distant spread of the virus from the local site of infection and contains two outer memb ranes.(Condit, Moussatche et al., 2006;Schramm & Locker, 2005) Virus mutants which are blocked in formation of EEV form normal levels of infectious virions in cell cultu re, but cause no disease in an imals.(Stern, Thompson et al., 1997;Wolffe, Isaacs et al., 1993;Payne, 1980) This is diagramed in Figure 1-5. Many of the virally encoded proteins are located in the virion and are therefore present at the time of infection including a fully functi onal RNA polymerase. Early viral mRNA is immediately transcribed from the cores derived from input virus and translated immediately. These early viral genes transcribed include those necessary for DNA replication, the second

PAGE 29

29 stage of uncoating, inhibition of host cell RNA and protein synthesis, host immune modulatory factors, and proteins needed for intermediate viral gene transcription and translation. Intermediate transcriptio n and translation is coupled to vira l DNA replication and is initiated from within virus factories located in the cytoplas m, providing late transc ription and translational proteins.(Condit, 2007) Poxvirus genomes are double stranded DNA (dsDNA) molecules that are continuous due to their hairpin loop termini (ITR). DNA replicati on occurs after sites of single stranded nicks occur at the ITRs. The free 3OH is then us ed for primer extension by the VV encoded DNA polymerase forming concatamer molecules of DNA. The concatamers are cleaved into single genome copies by virally mediated Holliday junc tion endonuclease. Late transcription and translation occur post DNA replication and produces all the structural vi rion proteins and other proteins ultimately packaged in to assembled virions.(Beaud, 1995) Vaccinia Virus Infection of Mice Animal models of poxviruses have been studied for decades g iving invaluable insight into the host-virus interaction. The responses to virus infection from a host animal are far more complex than can be observed in tissue culture and often give very different results as far as virulence and viral replication. Many orthopoxviruses have been studied in animals including vaccinia, cowpox, and rabbitpox virus.(Turner, 1967) The majority of the literature focuses on vaccinia virus as the representative virus and ha s been well characterized as far as the pathology and disease progression. Infection of poxviruses within a host is characterized by replica tion at the primary site of infection and then spread throughout the host via cell association. The virus migrates to the spleen, bone marrow, and lymph nodes by virtue of infected macrophages. An innate immune response to infection by cytotoxic T cells (CTL) is necessary to fight and clear the poxvirus

PAGE 30

30 infection, where as a humoral or antibody response is important for a full recovery and protection from reinfection. Increases in levels of interferon are also necessary to mount a sufficient immune response to poxvirus infections. Upon hist ological investigation, all poxviruses have the appearance of B-type, or Guarni eris bodies, inclusion bodies char acterized by oval bodies in the cytoplasm close to the nucleus. Late infections are characterized by necrosis and edema at the sites of virus replication.(Zaucha, Jahrling et al., 2001;Martinez, Bray et al., 2000;Fenner, 1990) Mice have been an invaluable model syst em for studying the pa thology and host-virus interactions. Several routes of infection for VV in mice have been studied and include intranasal (IN), intratracheal (IT), intrad ermal (ID) inoculation of the ea r pinnae (Tscharke & Smith, 1999) or footpad, scarification of the tail, intraperitoneal (IP), a nd intracranial (IC). (Figure 1-6A) Lung and upper respiratory infections result from both IN (Wilcock, Duncan et al., 1999) and IT infections, skin lesions result from ID (Read ing & Smith, 2003) and scarification methods of infection, peritoneal infections result from IP infections (Tur ner, 1967), and brain and central nervous system infections result from IC me thod of infection (Brandt Heck et al., 2005). The spread of VV from the primar y site of infection to distal sites in the mice, or viremia, has been shown to be variable depending upon th e route of infection used. The IC route of infection kills mice rapidly, with most virus repli cation in the brain with little spread to other distal organs.(Brandt, Heck et al., 2005) ID and scarification infections do not always lead to systemic involvement, but rather exhibit a locali zed necrotic lesion that either heals or becomes necrotic and gangrenous.(Reading & Smith, 2003) Th e IP route of infectio n allows the virus to non-specifically spread to multiple organs due to the centralized site of the virus introduction and therefore generates an almost in stant viremia.(Turner, 1967) The IN route of infection gives an upper respiratory system infection in which the virus is able to spread systematically.(Reading &

PAGE 31

31 Smith, 2003) The IT route of infection gives an upper and lower respirator y infection with rapid systemic spread of the virus. This study is focused on the lung infections of mice and therefore the choices for inoculation were the IT or IN rout es of infection. The IN route of infection is technically simple, but not highly reproducible due to the tendency of the animal to expel the virus inoculum. The IN route is also problematic in that the major ity of the inoculating virus remains in the nasal passages, giving a poor lung infectio n, but efficient spread to most organs within the animal. Despite the shortcomings of th e IN route of infection, the ma jority of the information on VV infectious if animals including viral spread has been derived from this route due to the technical ease of infection. The IT route of infection give s a true lung infection that rapidly causes viral spread to distal organs, and because of this the IT route of infection has been chosen for these studies. (Figure 1-6B) IT inoc ulations also are more reproduc ible than in IN inoculations because the animals do not sneeze following infecti on leading to inherent variability. However, due to the lack of literature su rrounding the IT infection of VV in mice, the data for the IN route of inoculation of VV will be utilized as a comparison. Several mouse strains have also been utilized in the investigation of the host-virus response to VV. C57BL/6 and Balb/C mice have been the most studied. Balb/C mice exhibit a humoral, or antibody based response to VV infections that has been demonstrated to be important for recovery from infection, but not sufficient to clear an acute infection.(Smith & Kotwal, 2002) C57BL/6 mice have a cell mediated response to VV infections which has been demonstrated to be the most effective in clearing VV infecti ons.(Xu, Johnson et al., 2004) The knockout animals selected for use in this study are also on the C57BL/6 background; therefore the C57BL/6 mouse will be used as wild ty pe mice for these studies.

PAGE 32

32 Control of Host Immune Responses by Poxviruses Poxviruses have developed multiple mechanis ms of evading the host immune response agains t viral infections both intracellular and secr eted. These proteins target multiple host innate responses that include, but are not limited to, complement, chemokine and immune cell recruitment, intracellular signaling, TNF signalin g, and interferon activation.(Seet, Johnston et al., 2003;Haga & Bowie, 2005) A se lect set of viral proteins a nd their cellular targets are described in Table 1-2. The i nnate immune response and IFN responses to VV will be the focus of this research. Therefore, viral proteins c ontrolling these early responses will be outlined with the focus on the IFN response. Vaccinia virus control of innate immune response The complement is activated by either of two pathways, the classical and alternative pathways, which ultim ately result s in cell killing. The products from these pathways not only serve to aid in the killing of in fected cells and inactivation of fr ee virus particles but to enhance the inflammatory response and attract immune cells, to the site of infection. The VV C3L gene encodes the vaccinia complement control protei n, or VCP, that has been shown to block complement activation by binding C3b and C4b to accelerate the decay of the C3 convertase and thereby prevent inactivation of both virus particles and killing of infected cells. (Seet, Johnston et al., 2003) Chemokine and cytokines are prot eins secreted by the host cells and serve as an attractant for immune cells to the site of infection. There are two classes of chemokines up regulated during a poxvirus infection, the CC chemokines th at attract macrophages and T-cells and CXC that attract neutrophils. These chemokines are se questered by virally encoded pseudo receptors and binding proteins such as th e one C23L and A41 encodes.(Bahar, Kenyon et al., 2008) This

PAGE 33

33 prevents the recruitment of imm une cells to the site of infec tion and killing of the infected cells.(Seet, Singh et al., 2001;Smith, 1999) A38L is a protein that is made in orde r to prevent immune cell recruitment and proliferation. Activated macrophages and T-cells produce TNF that signal for a proinflammatory response and death of the inf ected cells. The viral TNF receptor encoded by A53R and B28R prevents TNF from binding to it s host cell encoded receptors and activation of the pathway leading to cell deat h.(Seet, Johnston et al., 2003) Vaccinia virus control of interferon Vaccinia virus also encodes several proteins that control the interferon response and activ ation pathway. The proteins target multiple sections of the interferon response pathways including IFNs themselves, act ivation of the IFN pathway via recognition of PAMPs, and inactivation of both transcripti on and translational machinery. B8R is a virally encoded IFN pseudo receptor that is secreted and therefore sequesters IFN and prevents action of the cytokine not only at the site of infec tion but systemically. However, B8R has been shown not to have a high binding affinity for mouse IFN and most likely does not play a crucial part controlli ng in the pathogenesis of VV in mice.(Mossman, Upton et al., 1995;Alcami & Smith, 1995) B18R is a secreted type I IFN binding protein that works to prevent the activation of the type I IFN response by sequestering extracellular IFN and IFN .(Colamonici, Domanski et al., 1995) B18R has been shown to bind mouse, human, rat, rabbit, and bovine type I interferons.(Symons, Alcami et al., 1995) Not only does the virus block the binding of IF Ns to their receptors, but the responses within a cell to viral infection that cause IFN production are also blocked. One of the major inducers of IFN production is the presence of dsRNA, which poxviruses produce late in infection due to the bidirectional nature of transcription. A virus encode d dsRNA binding protein, E3L, is

PAGE 34

34 encoded by the virus to block the host cell fr om recognizing the presen ce of dsRNA.(Kibler, Shors et al., 1997) K3L is an eIF2 pseudo substrate that has been hypothesized to bind and sequesters PKR from being activated in th e presence of dsRNA and shutting down host translation.(Carroll, Elroy Stein et al., 1993) A viral encoded phosphatase encoded by H1L blocks the activation of STAT-1 by dephosphorylating and preventing the transcriptional enhancement function of STAT-1. This study will focus on the inhibition of interferon activation by the control of cellular responses to dsRNA by the virus, spec ifically the roles of E3L and K3L and their corres ponding host protein targets. Vaccinia Virus Gene E3L Normally VV is refrac tile to the action of IFN. E3L was first identified as a protein that when deleted from VV the resulting virus, VV E3L, exhibited IFN sensitivity at 2 hours post infection.(Beattie, Paoletti et al., 1995) This se nsitivity to IFN could be reversed if the protein were supplied in trans by a plasmid during in fection, linking the E3L gene directly to the resistance of VV to IFN.(Chang, Uribe et al., 19 95) The sensitivity to IFN suggested a close relationship between E3L and control of the host immune response, specifically IFN. The E3L protein is expressed ear ly in infection and remains up to 4 hours post infection to allow for virus replication.(Beattie, Paoletti et al., 1995) E3L has 2 domains necessary for function. The C-terminal end of the protein contains a dsRNA binding motif (Chang & Jacobs, 1993;Ho & Shuman, 1996) and the Nterminal domain that has b een characterized to bind both PKR and Z-DNA (Kim, Lowenhaupt et al., 2004). The binding and sequestering of PKR is an important function and aids in preventing activat ion of the antiviral st ate and IFN production. The sequestering of dsRNA from cellular dsRNA response proteins protects the infected cell from apoptosis.(Kibler, Shors et al., 1997)

PAGE 35

35 VV E3L has been studied since the 1990s and is currently under investigation as an alternative vaccine for smallpox.(Vijaysri, Jentarra et al., 2008) In tissue culture VV E3L exhibits a host range phenotype in which the virus has been previously re ported to only replicate only in chicken embryo fibroblas ts (CEF), baby hamster kidney (BHK-21), and rabbit kidney (RK-13) cells.(Beattie, Kauffman et al., 1996);(Ch ang, Uribe et al., 1995) The restricted host range phenotype exhibited by VV E3L was able to be restored in tissue culture by transfection of a plasmid encoding E3L. VV E3L has also been investigated in multiple mouse models and has been shown that the entire protein is needed for virulence in animals, unlike in tissue culture cells where only the C-terminal dsRNA domain is needed for replication.(Brandt & Jacobs, 2001) Infection of mice via intranasal or intracranial routes with VV E3L failed to produce disease symptoms in the mice(Brandt & Jacobs 2001), yet provided prot ection from challenge with the pathogenic VV-WR.(Brandt, Heck et al., 2005;Vijaya, Jentarra et al., 2008) Several dsRNA binding proteins from other viruses and bacteria have been shown to replace the function of E3L in vitro but have failed to comp letely restore virulence in vivo .(Vijaysri, Talasela et al., 2003;Shors & Jacobs, 1997;Beattie, Denzle r et al., 1995) E3L has been shown in vitro to rescue ECMV when provided transiently during an ECMV infection from the antiviral effects of IFN .(Shors, Beattie et al., 1998) Transgenic mice that express the E3L protein are more susceptible to VV infection and exhibit an inabilit y to resolve dermal lesions due to in inability to mount an immune response to the infecti on.(Domingo-Gil, PerezJimenez et al., 2008) Utilizing both knockout mice for the targets of E3L and VV E3L in vivo the interactions of E3L with the PKR and RNaseL pathways will be evaluated. Vaccinia Virus Gene K3L Vaccinia virus gene K3L, similar to E3L, was initially id entified as a protein responsible for allowing the virus to repli cate in the presence of IFN.(Beattie Tartaglia et al., 1991;Beattie,

PAGE 36

36 Paoletti et al., 1995) VV deleted for K3L, VV K3L, exhibited sensitivit y to INF that could be rescued by providing the protein in trans.(Beattie, Paoletti et al., 1995) K3L is a 10.5kDa protein that has been crysta llized and the structure determined. The structure consists of 5 beta stra nds and 2 alpha helices to form a beta barrel and shares sequence homology with eIF2 .(Dar & Sicheri, 2002) This N-term inal portion of K3L that shares homology with eIF2 is thought to bind PKR through the same C-terminal domain where it has been determined that eIF2 binds.(Dar & Sicheri, 2002) K3L is expressed early during poxvirus infection, from 0 to 2 hours post infection (Beattie, Paoletti et al., 1995) and serves to control of the host response to dsRNA by binding to PKR and preventing the phosphorylation of eIF2 .(Davies, Elroy Stein et al., 1992;Davies, Chang et al., 1993) This response is important to prevent the shutdown of host tr anslational machinery required for virus growth and cell de ath as the virus requires the cell to stay alive for replication. Vaccinia virus lacking the K3L gene, VV K3L, has been previously shown to be sensitive to IFN 30 minutes after infection of cells.(Beattie, Paoletti et al., 1995) VV K3L has been previously reported to exhib it a restricted host range in vitro replicating in Vero and HeLa cells, but failing to replicate in BHK cells.(Langland & Jacobs, 2002) K3L has been demonstrated in vitro to partially rescue VSV fr om the antiviral effects of IFN when provided transiently.(Shors, Beattie et al., 1998) Animal studies with VV K3L have shown that the protein is necessary for full virulence in mice in an IN model. Utilizing both knockout mice for the target of K3L and VV K3L in vivo the interactions of K3L with PKR will be evaluated. Study Objectives This study is focused on investigating the inte ractions between the V accini a virus proteins E3L and K3L with host cellular proteins PKR and RNaseL. E3L has been identified as a dsRNA

PAGE 37

37 binding protein necessary for pat hogenesis in a mouse model. K3L has been shown to bind PKR and prevent phosphorylation of eIF2 in vitro and also necessary for full virulence in a mouse model. VV proteins E3L and K3L have been previously identified to control the IFN response to infection in tissue cultu re and their target cellular protei ns are known to upregulate the IFN response. PKR and RNaseL are known to be involved in the response to dsRNA and upregulate IFN in the host as a method to control viral infections. The goal was to establish the relative importance of the host and virus gene inte ractions during a lung infection with VV. Using PKR, RNaseL, and PKR/RNaseL knockout mice the importance of these host genes in controlling wild type VV infection was determined by su rvival, virus dissemination, and histopathology. It was hypothe sized that the knockout animal s would exhibit an increased sensitivity to VV infections as compared to wild type mice, with the double PKR/RNaseL knockout animals being more sensitive than either of the tw o single knockouts. VV deleted for E3L or K3L was also invest igated in wild type, the single PKR and RNaseL knockout, and the double PKR/RNaseL knockout mice. The infection of mice with VV deleted for E3L allowed for determination of the importance of E3L during a VV infection. It is expected that the virus would be highly attenua ted and not cause lethal systemic disease. VV deleted for E3L in the knockout animals allows for the determination of the interaction of E3L with both the RNaseL and PKR pathways. Th e single knockout animals were expected to exhibit mild, if any disease due to the fact that only one of the dsRNA pathways that E3L is hypothesized to target is missing. The double PKR/RNaseL knockout mouse was hypothesized to exhibit an increased sensitivity to VV delete d for E3L due to both dsRNA response pathways targeted by E3L are missing. The level of dis ease was determined by clinical symptomology, survival, virus dissemination, and histopathology.

PAGE 38

38 The infection of wild type mice with VV deleted for K3L was hypothesized to be attenuated because the PKR pathway was present in the wild type mice. The PKR single knockout mice infected with VV deleted for K3 L was hypothesized to exhibit no attenuation because the target for K3L was absent in th e mice. The RNaseL single knockout mice were hypothesized to exhibit responses si milar to those seen with wild type mice as the PKR pathway was still intact in these animals. The PKR/ RNaseL double knockout mice were hypothesized to exhibit wild type VV responses to VV deleted for K3L because the PKR pathway missing. The level of disease was determined by clinical symptomology, survival, virus dissemination, and histopathology.

PAGE 39

39 Figure 1-1. Interferon signaling through the Type I IFN receptor. See text for details. (Randall & Goodbourn, 2008;Katze, He et al ., 2002;Takoaka & Yanai, 2006)

PAGE 40

40 Figure 1-2. Response to dsRNA within a ce ll. The responses and downstream signaling molecules that up re gulate the production of interferon and interferon response genes are denoted in multiple pathways. (Ra ndall & Goodbourn, 2008;Harte, Haga et al., 2003;Takoaka & Yanai, 2006;Borden, Sen et al., 2008)

PAGE 41

41 Figure 1-3. PKR activation. PKR is activated in response to dsRNA to inhibit translation within the cell and prevent a productive viral infection. (Gale, Jr. & Katze, 1998)

PAGE 42

42 Figure 1-4. OAS/ RNaseL activat ion. The OAS/RNaseL pathway is activated in response to dsRNA. This activation lead s to cell apoptosis by degrad ation of all RNA within a cell and prevents a productive viral infection. (Samuel, 2001)

PAGE 43

43 Table 1-1. Poxvirus classifications Subfamily Genera Example Chordopoxvirinae Orthopoxvirus Vaccinia, Smallpox Leporipoxvirus Myxoma Parapoxvirus Sealpox Avipoxvirus Fowlpox Capripoxvirus Goatpox Molluscipoxvirus Molluscum contagiosum Yatapoxvirus Yaba monkey tumor Suipoxvirus Swinepoxvirus Entomopoxvirinae Alphaentomopoxvirus Anomala cuprea Betaentomopoxvirus Amsacta moorei Gammaentomopoxvirus Chironimus luridus Chordopoxviruses infect vertebrate s while the Entomopoxviruses inf ect invertebrates. Within each subfamily the genera and an example from th at genera are listed.(Fenner, 2000;International Committee on Taxonomy of Viruses, 2002)

PAGE 44

44 Figure 1-5. Poxvirus life cycle. See text for specifics. (Moss, 1996;Moss, 1996;Moss, 2001;Moss, 2006;Broyles, 2003;Condit, Moussatche et al., 2006;Schramm & Locker, 2005)

PAGE 45

45 A B Figure 1-6. Infection of Mice. A) Various routes of infection used in mice and the relative location on the mouse. Mouse diagram from Science Slides. (Turner, 1967) B) Photograph of the surgical infection of a mouse by the intratracheal (IT) method.

PAGE 46

46 Table 1-2. Vaccinia virus encoded genes that control the host immune system Immunomodulatory protein Viral Gene Pathway involved TNF receptor A53R, B2 8R Sequesters TNF Phosphatase H1L Dephosphorylation of STAT-1 eIF2 homolog K3L PKR inhibitors IFN receptor B8R Competitive antagonists of IFN IFN / binding proteins B18R Sequester s and inhibits IFN extracellularly dsRNA binding protein E3L Sequest er dsRNA to avoid apoptosis IL-1 receptor B16R Sequesters IL-1 extracellularly Toll like receptor A46R, A52R Di srupt IL-1 receptor signaling Chemokine binding protein C23L A41L Binds CC chemokines 3 -hydroxysteroid dehydrogenase A44L Regulate steroid hormones, DHT, neural steroids Semaphorins A39R Mediates receptor binding specificity Viral growth factor C11R Viral growth Complement inhibition C3L Inhibits C3b/ C4b (alternative and classical pathway) CD47-like protein A38L Immune cell proliferation/recruitment SPI 1 C12L Host Range SPI 2 B13R Fas TNF mediat ed apoptosis inhibitors SPI 3 K2L Protease inhibition IL-18 BP D7L IFN induction inhibition The poxvirus protein classification, VV gene that encodes the prot ein and the proposed role of the protein during infection. (Seet, Johnston et al., 2003;Haga & Bowi e, 2005;Bahar, Kenyon et al., 2008;Seet, Singh et al., 2001; Smith, 1999;Mossman, Upton et al., 1995;Alcami & Smith, 1995;Colamonici, Domanski et al., 1995;Symons, Alcami et al ., 1995;Kibler, Shors et al., 1997;Carroll, Elroy Stein et al., 1993)

PAGE 47

47 CHAPTER 2 MATERIALS AND METHODS Tissue Culture and Virological Techniques Tissue Culture CV-1, BHK-21, and PK-15 cells were mainta ined in Minimu m Essential Media (MEM) with Earles Salts (Gib co, Grand Island, NY) supplemented with 2 mM glutamine (Media Tech, Herndon, VA), 50 U/mL penicillin G & 50 g/mL st reptomycin (Media Tech), 1 mM sodium pyruvate (Media Tech), and 0.1 mM nonessential amino acids (Media Tech), and 10% v/v FBS (Gibco). CV-1 cells were maintained at a split ratio of 1:10, BHK-21 cells were maintained at a split ratio of 1:15, and PK-15 cells were maintained at a split ratio of 1:6. CV-1 and PK-15 cells were maintained in 5% CO2 incubators. BHK-21 cells were maintained in 10% CO2 incubators. Growth of Virus Stocks for Animal Injections Vaccinia virus Western Reserve (V V) was obtai ned from R. Condit and maintained as the seed stock. This stock was obtained by R. Condit from the ATTCC as a mouse brain homogenate in which he subsequently plaque purifie d. It is this first round of plaque purification that is maintained as the seed stock for all VV grown for these experiments. Vaccinia virus expressing the gfp protein was generated by Peter Turner. This virus was generated by placing the gfp gene under the control of the poxviru s early late promoter in the ATI locus of VV. All viruses were grown on confluent cell monolayers in 150mm tissue culture dishes (Corning). Viruses were inocul ated at a multiplicity of inf ection (MOI) of 0.01 in 5mL of growth media without supplements per 150mm dish for one hour at 37C on a rocking platform for virus absorption. Post absorption the virus inoculum was removed and 25mL of supplemented growth medium was added to each dish. The dishes were then returned to normal

PAGE 48

48 growth conditions as stated in the tissue culture section. The vi rus was allowed to grow for 3-6 days, or until a complete CPE was observed. Once CPE was observed the dishes were scraped using cell scrapers to dislodge all the cells into the media. The media containing the cells was then collected into 500mL bottles (Corning) a nd centrifuged at 4C for 45 minutes at 12,000xg. The media was then separated from the ce ll pellet and discarded. The cell pellet was resuspended in 10mL of 10mM Tris-HCl and doun ced on ice for 30 strokes. The resulting cell homogenate containing the desired virus was th en centrifuged at 4C for 5 minutes at 500xg. The supernatant was removed and saved, the pell et resuspended in 2mL of 10mM Tris-HCl, vortexed to resuspend the cell pellet, and centrif uged again at 4C for 5 minutes at 500xg. The supernatant from the second centr ifugation was pooled with the fi rst supernatant and the cell pellet discarded. The supernatant was then so nicated using a probe sonicator (Sonics and Materials) for 2 minutes using the microtip at an output setting of 5 and 90% duty cycle, allowed to cool on ice for 2 minutes, and then sonicated again for 2 minut es. The sonicated supernatant containing the virus of interest is then placed onto a 17mL 36% sucrose pad and centrifuged at 4C for 80 minutes at 16,000xg in the ultra centr ifuge rotor SW28 (Beckm an Optima LE 80K). All aqueous layers are removed and the resul ting pellet is resuspende d in 1mL of phosphate buffered saline (PBS), stored in 0.1mL a liquots, and stored at -80C until use. Plaquing of Virus Cells are grown using standard growth conditi ons in 6 well p lates until confluent. All samples that are to be plaqued are thawed, sonicated for one minute in a water bath cup sonicator (Sonics and Materials), and then maintained on i ce. Dilutions of the samples are then made using growth media without any supplements and between 300-500l of the dilutions is placed onto each well. The plates are then placed at 37 C for one hour with constant rocking for virus absorption. After one hour absorp tion period the virus inoculum is removed and 3mL of a 1:1

PAGE 49

49 mixture of 2x growth media: 1% agarose (sterile ) at 55C is added to each well. The plates are then returned to normal growth conditions for 2-4 days. Once plaques are visible (2-4 days) the agar/m edia overlay is removed and the plates are stained with crystal violet a nd counted. For BHK-21 cells th e plaques are counted without staining using fluorescence to detect gfp and a dissecting microscope. Generation of Virus Mutants and R econstructed Wild type Revertants Vaccinia virus strain We stern Reserve (VV) deleted for the E3L gene (VV E3L::gfp) was generated by infection of BHK-21 cells with wild type VV at an MOI of 0.01 and then transfected using Lipofectamine 2000 (Invitrogen) with the E3L::gfp linear PCR fragment. The generation of these fragments is described in the cloning of VV E3L::gfp fragment section. Cells were harvested 48 hours later, underwent 3 freeze thaw cycles using dry ice and 37C water bath, and plaque purified on BHK-21 cells. BHK-21 cells we re the only cell line tested that would allow for the growth of VV E3L::gfp, and therefore were the cell line chosen for the screening. Plaques were screened for the presence of gfp by fluorescence microscopy and underwent a total of 3 rounds of plaque purification. VV E3L::gfp was rescued by infecting a 35 mm well BHK-21 monolayer at an MOI of 0.01 with purified VV E3L::gfp, then transfected with 1ug of wild type E3L gene and allowed to grow for 2 days. The cells were then harv ested, centrifuged at 16,000xg for 5 minutes at room temperature. The supernatant was removed and the cell pellet resuspended in 1mL of supplemented MEM plus 0.01M HEPES. Three cycles of freeze thaw were done and the virus mixture plaqued on PK15 cells. Due to the host range exhibited by VV E3L::gfp only viruses that have been successfully rescued will plaque on PK15 cells. Three days later plaques were isolated, and re-plaqued on PK15 cells for a total of 3 rounds of plaque picks to ensure no

PAGE 50

50 residual VV E3L::gfp was present. The resulting rec onstructed wild type virus is denoted as rVVE3L. Vaccinia virus strain Western Reserve deleted for the K3L gene (VV K3L::gfp) was generated by infection of RK13 cells at an MOI of 0.01 and then transfected using Lipofectamine 2000 (Invitrogen) with the K3L::gfp linear PCR frag ment. Cells were harvested 48 hours later, underwent 3 freeze thaw cycles using dry ice an d 37C water bath, and plaque purified on RK13 cells. Plaques were screened for the presence of gfp by fluorescence microscopy and underwent a total of 3 rounds of plaque purification. VV K3L::gfp was not rescued due to technical difficulty in screening for rescued viruses in that there was no cell line that allowed for selection. DNA Techniques Cloning of Wild Type E3L and K3L Genes The wild type E3L gene product was PCR amplifie d from wild type VV genomic DNA using forward primer IDT344: GAGAATT CCTTGGTTCATACATGAAATGATC and reverse primer IDT345: GAGAATTCCAC AAACATCAATGGCGGTAAC. The wild type K3L gene was PCR amplifie d from wild type VV-WR genomic DNA using forward primer IDT419: ATTAGACATACCGGATCTACG and IDT420: AAAGCGGTTAAATCATTGGTTC. Cloning of E3L::gfp Deletion Fragment The E3L::gfp frag ment (gfp fragment fla nked by E3L sequences) was generated in a PCR mediated ligation reaction devel oped by Peter C. Turner. In this reaction 3 fragments are generated, a left virus flanking sequence, a right virus flanking sequence and a middle reporter gene under its own promoter. Th e 3 fragments then undergo rest riction enzyme digestion and ligation as is the field standar d, however rather than transformi ng the fragments into a vector PCR of the ligation mixture is immediately perf ormed. The PCR using primers located in the

PAGE 51

51 left and right flanks allow for specific amplificat ion of the desired product for future use. The final construct is shown in Figure 2-1A. The E3L::gfp PCR fragment was generated by amplifying the left and right flanks. The left flank was amplified using ID T345 GAGAATTCCACAAACATCAATGGCGGTAAC and IDT594 AGCTCCTTCGATTCC. The right flank was amplified using IDT344 GAGAATTCCTTGGTTCAT ACATGAAATGATC and IDT595 CTCGTTTAGATTTTCC. The 874 bp gfp and promotor fragme nt was amplified using IDT 30 CCAGACATTGTTGAATTAGATCG; IDT130b GCTGGTACCGGTGGGTTTGGAATTA GTG from a plasmid containing gfp under the poxvirus early late promoter. All PCR products were cleaned up using Qiagen PCR clean up kit with all optional steps performed and DNA eluted in 50l. The E3L left fragment was digested with Acc65I (NEB), the E3L right fragment was di gested with NotI (NEB) and the gfp fragment was digested with NotI and A cc65I per NEB guidelines. The left and right fragments were treated with Antarctic Phosphatase (NEB) for 15 minutes. All restric tion digests were then cleaned up using the Qiagen PCR clean up kit with all optional steps performed and DNA eluted in 30l. The three fragments were then ligated overnight at 16C using T4 ligase (where from). The ligation mixture was then used for PCR template for the E3L::gfp PCR fragment. IDT599 GAGGTTCGTCAGCGGC and IDT 598 GTGATAATTTATGTGTGAGGC primers were used for PCR generation of the E3L::gfp PCR fragment. The E3L locus of 570bp was deleted from 81 to 488bp with gfp replacing this region. The deletion of E3L was confirmed by PCR using the following primers: IDT529 ATAGCCTTGTCCTCGTGCAG and IDT530 CG CAATCGATACATGAAAA CA. PCR with primers IDT 344 GAGAATTCCTTGGTTC ATACATGAAATGATC and IDT 272

PAGE 52

52 GAGTTATAGTTGTATTCCAGCT was also performed to ensure that gfp was located in the E3L gene locus. (Figure 2-1B) Cloning of K3L::gfp Deletion Fragment The cloning of the K3L::gfp fragment (gfp with K3L flanking seque nce) was done using the same protocol as that used for the E3L::gfp fragments. The K3L::gfp fragment PCR fragment was generated by amplifying the left a nd right flanks. The left flank was amplified using IDT639 CTTTTGTATAATCAACTCTAA and IDT531 ATGCAGGCAATAGCGACATA. The right flank was amplified using IDT638 CCTTCTCGTATACTCTGCCC and IDT532 AGATGCTCCACATGTAT. The 874bp gfp fragment was amplified using IDT 30 CCAGACATTGTTGAAT TAGATCG; IDT130b GCTGGTACCGGTGGGTTTGGAATTA GTG from a plasmid containing gfp under the poxvirus early late promoter. All PCR products were cleaned up using Qiagen PCR clean up kit with all optional steps performed and DNA eluted in 50l. The K3L right fragment was digested with Acc65I (NEB), the K3L left fragment was digested with NotI (NEB) and the gfp fragment was digested with NotI and A cc65I per NEB guidelines. The left and right fragments were treated with SAP for 15 minutes. All restriction digests were then cleaned up using the Qiagen PCR clean up kit with all optional steps performed and DNA eluted in 30l. The three fragments were then ligated overnight at 16C using T4 ligase (where from). The ligation mixture was then used for PCR template for the k3L::gfp PCR fragment. IDT531 and IDT532 primers were used for PCR generation of the K3L::gfp PCR fragment. The K3L locus of 267bp was deleted from 45 to 199bp with gfp re placing this region. (Figure 2-2A) The deletion of K3L was confirmed by P CR using IDT531 and IDT532 and the correct position of gfp in the K3L locus was confirme d with PCR using IDT419 and IDT272 as shown in Figure 2-2B.

PAGE 53

53 Purification of Viral DNA Cells (virus specifi c) in a single 35mm dish were infected with the desired virus at an MOI of 1 and grown until CPE was complete. Cells were then scraped from the dish into the growth media, centrifuged at 14,000xg for 5 minutes and media removed from the pellets. The pellets estimated at 2x106 cells were then placed in the Qi agen DNeasy kit and processed per the manufacturer instructions for tissue culture cells. DNA was eluted in 50ul of water. Sequencing of Viruses The E3L locus of both the wild type and deletion virus was sequenced using primers IDT529 ATAGCCTTGTCCTCGTGCAG and IDT530 CGCAATCGATACATGAAAACA. The K3L locus of both the wild type and deletion virus was sequenced using IDT531 ATGCAGGCAATAGCGACATA a nd IDT532 AGAT GCTCCACATGTAT. All sequencing was performed by the ICBR Sequencing Core (Unive rsity of Florida). Th e resulting sequences were aligned using AlignX (Invitrogen) to ensure correct viral sequences. PCR All primers used for PCR were made by IDT (where). Reactions were perform ed in a PTC-100 Thermal Controller (MJ Research Inc, Watertown, MA). All PCR reactions were carried out in a final volume of 100l contai ning 1.5mM MgCl2, 2U of Vent DNA polymerase (New England Biolabs) with the manufactur er supplied buffer, and 200uM dNTPs. The conditions were denature at 94C for 4 minutes; 30 cycles of: denature at 94C for 1 minute, primer binding at 35C for 1 minute, and extension at 72C for 1 minute per 1000bp expected band size; and a final extensi on step of 72C for 5 minutes.

PAGE 54

54 RNA Techniques RNA Isolation from Lung Tissue Entire lungs weighing approxima tely 0.2mg from animals flas h frozen in liquid nitrogen and maintained at -80C were used for RNA isolation. RNA isolati on was performed using 1mL of Trizol (Invitrogen) and 1mL of 24 grit s ilicon carbide (Electro Abrasives, Buffalo, NY) placed into 2mL screw top tubes (Sardstedt). Th e tubes are then placed in the Mini-Beadbeater 8 (Biospec, Bartlesville, OK) for 1 minute on the homogenization setting. The resulting lung/ Trizol homogenate is then separated from the s ilicon carbide and placed into a separate tube; the remaining silicon carbide was then washed with an additional 1mL of Trizol, removed and pooled with the initial homogenate. 0.2mL of chloroform per 1mL of Trizol was added and shaken vigorously for 15 seconds. The tubes we re then incubated at room temperature for 3 minutes and centrifuged at 1200xg for 15 minutes at 4C. The upper layer is then removed, placed into a new tube, and 1U of glycogen added (Ambion). Isopropyl alcohol is then added at 0.5mL to 1mL of Trizol initially used and incu bated at room temperature for 10 minutes. The samples are then centrifuged at 12000xg for 10 mi nutes at 4C. The supernatant was removed from the pellet, the pellet washed with 1mL of 75% ethanol, vortexed, and centrifuged at 7500xg for 5 minutes at 4C. The ethanol was then rem oved, the pellet air dried for 5 minutes at room temperature. The resulting RNA pellets were then placed through the Micro-to-Midi Total RNA Purification System (Invitrogen) for the removal of any residual cellular DNA and heme that was not removed in the Trizol purification. The cellular DNA was removed by DNaseI treatment. The RNA pellet was resuspended in 16ul of water, and protocol for DNaseI treatment to remove genomic DNA in the Micro-to-Midi Total RNA Purification System was followed. DNaseI was inactivated by direct proceeding to the purification of RNA from liquid samples. The RNA was brought up to 100ul with water, and 100ul lysis bu ffer, 1ul BME, and 100ul 100% ethanol added.

PAGE 55

55 The procedure for purification of RNA from liquid samples was then followed using the 300600g expected RNA yield elution conditions RNA was then stored at -80C. RNA Processing and Microarrays 2g of RNA was used for generation of cRNA for the Affy metrix microarrays. The first strand cDNA synthesis was performed using T7-(dT)24 primer and Superscript II RT (Invitrogen). Second strand synthesis was performed using Second Strand Reaction buffer (Invitrogen), of E. coli DNA ligase (NEB), E. coli DNA polymerase I (NEB), E. coli RNase H (NEB), and T4 DNA Polymerase (NEB). cDNA was then purified using GeneChip Sample Cleanup Module (Affymetrix). cRNA was then made using Affymetrix 3 Labeling IVT kit (Affymextrix) and cleaned up using GeneChip Sample Cleanup Module (Affymetrix). The samples were then fragmented, hybridized, and detected on the Affymetrix Mouse 430.2A arrays (Affymetrix). Microarray Data Analysis The analysis of the microarray data was pe rformed using D-Chip (who) for unsupervised analysis and BRB Array Tools (NIH) for the supe rvised analysis. The CEL files were brought into D-Chip and the perfect match/ mismatch va lues read. There was one array that had high background and was therefore excluded from furthe r analyses. The median probe intensity was used to normalize the arrays, with the median inte nsity of 84 and the percent call of ~50% across all arrays. The expression values were determined using the Model-based expression method and 5th percentile of region, perf ect match only, as background. Th e genes were filtered using a variation of standard deviati on/mean of 0.5 to 1000 with the pe rcent of genes present in the arrays of 1%. For the supervis ed analysis all arrays were im ported into BRB Array Tools and subjected to an F test with random variance with a significance level of 0.001 in which all arrays were assigned to classes based upon the day pos t infection the lungs were harvested. These

PAGE 56

56 resulting 1811 probe sets were then pictorially represented in a heat map generated by BRB Array Tools. BRB Array Tools was also used to generate heat maps of probe sets identified as members in specific cell types or pathways. These were performed on the entire data set, not the genes only identified in the F-test as significant. The desire d BioCarta annotated pathways supplied in BRB Array Tools were selected when generating heat maps of probes in specific pathways. Animal Techniques Mouse Breeding and Line Maintenance Mouse constructs used in these experiments were obtained from Robert Silverma n (Cleveland Clinic) and maintained as homozygous lines. The four constr ucts are on a C57B/L6 background. All breeders were genotyped from a tail snip using the following primers: PKR forward: GTTTGGCTATTTCTCTGTGTTCATT GGA; PKR wild type reverse: GTAATGGCTACTCCGTGGATCTGGG C; PKR knockout reverse: ATTCGCAGCGCATCGCCTTCTATCGCC; RNaseL forward: GCATTGAGGACCATGGAGAC; RNaseL reverse: GGAGGAGAAGCTTTACAAGGTG. PKR homozygous wild type mice only have a 1.5 kb band in the wild type PCR reaction; PKR homozygous knockout mice only have a 1.5kb band in the PKR knockout PCR reaction; RNaseL knockout mice have a 2.0 kb band, and RNaseL wild type band is 1.0 kb. Figure 2-3 shows the representative banding pattern s observed in the genotyping. Females were bred at 12 weeks or older, ma les at 8 weeks or older. All mice were maintained SPF on ventilated automatic watering housing racks. All breeding pairs were fed a 12% protein diet (Purin a 5062), all maintenance and non breedi ng mice were fed 6% protein diet (Harlan 7912). At the time of mating all mice were tagged for iden tification in the right ear and

PAGE 57

57 housed in amber cages with corncob bedding, co tton bedding, and a sterile paper mouse hut. Continuous breeding was performed when possible. At 15-17 days of age of the pups a water bottle and moist food was placed into the bottom of the cage. Water bottles and moist food in the bottom of the cage of the pups was maintained for 2-3 weeks post weaning. Mouse Infections All mice were infected via an intratracheal infection (IT). For this me thod, general anesthesia was used (isoflurane delivered via vaporizers with mixed oxygen). The plane of anesthesia of the animal is determined by a toe pinch, and failure to respond to the toe pinch is considered adequate anesthesia. A small incision through the skin is made along the throat of an anesthetized mouse and the musc le underlying the skin is separa ted via blunt dissection, and a microchip (Bio Medic Data Systems, Seaford, DE ) is placed subcutaneously. The esophagus is then displaced slightly to reveal the trachea, a nd 30l of virus diluted in PBS is then injection between the rings of the trachea to wards the lower airways. The incision is then closed with surgical glue (Nexaband, Abbott Animal Healt h, Chicago IL) and the mouse righted quickly to prevent the reflux of the virus in to the upper airways and nasal pa ssages. The mouse is allowed to waken before being returned to the cage. Monitoring of Infected Animals Each mouse is microchipped (Bio Medic Data System s, Seaford, DE) as described above at the time of infection and that microchip tr ansmits the temperature and mouse identification number to the DAS-5007 reader (B io Medic Data Systems, Seafor d, DE). Weight, temperature, and physical observations of the mice (grooming habits, facial swelling, secretions, removal of hair, etc) are recorded daily. Criteria for euthanizing the mice include open mouth breathing, severe difficulty in breathing, body temperature of 30C of lower as measured by the microchips, or greater than 30% body weight loss from beginning weight. A c linical scoring guide for mice

PAGE 58

58 infected with poxviruses has been created and us ed to determine the disease severity in the animals. (Table 2-1) Euthanasia is perf ormed by injecting 0.1 mL of Beuthanasia D intraperitoneally and then confirming th e cessation of a heart beat by palpitation. Processing of Tissues for Titering Lungs were removed while being careful to separate the major bronchi from the lung tissue. Th e tissue was flash frozen in liquid nitrogen a nd kept at -80C. Sections of the liver, spleen, one gonad, and the brain were placed into 2mL tubes and flash frozen in liquid nitrogen and kept at -80C until processing. The tissue was weighted prior to homogenization using an analytical balance (XS105 Dual Ra nge, Mettler Toledo) to determin e actual weight in mg of the tissue. The frozen tissues or whole blood are pl aced into 2mL screw top tubes (Sarstedt) with ~1.0 mL sterile 24 grit silicon carbide (Electro Abrasives, Buffalo, NY) and 1mL growth media without serum (Minimum Essential Media, Invitr ogen). The tubes are then placed in the MiniBeadbeater 8 (Biospec, Bartlesville, OK) fo r 1.5 minutes on the homogenization setting. The resulting supernatant is then s onicated for 1 minute in the wate r bath sonicator at an output setting of 9 and 70% duty cycle a nd virus concentration determined by plaque assay as described in the plaquing of virus and tissues section. Interferon Beta ELISA Assays IFN ELISAs assays were performed using the Mouse IFN-Beta ELISA kit (PBL Biomedical Laborato ries, Piscataway, NJ). All lung samples were weighed using an analytical balance (XS105 Dual Range, Mettler Toledo) to dete rmine actual weight in mg of the tissue. Tissue culture growth media without FBS was adde d to each lung tube for a final concentration of 0.2mg tissue/ 1mL of media, then approximately 1mL of sterile 24 grit s ilicon carbide (Electro Abrasives, Buffalo, NY) was added to each tube. The lung samples were then homogenized for

PAGE 59

59 30 sec using the Mini-Beadbeater 8 (Biospec, Bartlesville, OK), placed on ice and used immediately for the ELISA. Each ELISA kit had an indivi dual Certificate of Analysis in which the antibody standards and HRP detection antibody dilution s were prescribed to ensure consistency among kits used. For this study 4 kits were used with the standa rds and samples run on each plate in duplicate. The manufacturer supplied protocol was followed for these ELISAs. Histological Methods Tissue Processing Mice were euthanized as previously desc ribed and tissues rem oved immediately. The lung was sectioned horizontally multiple times per lobe, placed into a tissue cassette and immediately placed into 10% buffered formalin (F isher Scientific). Multiple sections cut horizontally from different liver lobes and the entire spleen cut transversely were placed into tissue cassettes and immediately placed into 10% buffered forma lin. Tissues were fixed for 1218 hours in formalin at room temperature and then transferred to PBS until embedded. All tissues were imbedded in paraffin by the Molecular Pathology Core at the Un iversity of Florida. Tissues were sectioned on continuous sect ions of 4um in thickness by the Molecular Pathology Core at the Un iversity of Florida. H&E Staining 4um sections of tissue we re stained with hematoxylin and eosin by the Molecular Pathology Core at the Un iversity of Florida. Anti-gfp Immunostaining Sections of lung, liver and spleen were stained with anti-gfp antibody. This antibody wa s used to detect virally infected cells as all the viruses used for histopathology express the gfp

PAGE 60

60 protein. The paraffin was removed using the following solvent series: xylene for 5 minutes 2 times, 100% ethanol for 2 minutes 2 times, 3% H2O2 in methanol for 10 minutes, 95% methanol for 3 minutes, 70% methanol for 1 minute, water for 1 minute, TBS-Tween buffer for 5 minutes. The tissues were blocked using Sniper serum (Biocare Medical, Concord, CA) for 15 minutes followed by 2 washes in TBS for 5 minutes. The primary antibody, anti -gfp antibody (Abcam) diluted at 1:40,000 was added for 60 minutes follow ed by 2 washes in TBS for 5 minutes. The secondary antibody used was the Mach 2 goat anti rabbit HRP polymer (Biocare Medical, Concord, CA) for 30 minutes. Cardassian DAB ch romagen (Biocare Medical) was then used to develop the slide color for 5 minutes and then sl ides were washed with water to stop the color development. The nuclei were stained with Hematoxylin (Vector) for 1 minute, washed with water, placed into TBS for 1 minute to develop the blue color and rinsed w ith water. The tissues were then dehydrated by soaking in 80% ethano l for 1 minute, a quick dip into 95% ethanol, followed by a 30 sec soak in 95% ethanol, 100% ethanol 2 times for 1 minute and, a soak in xylene 2 times for 1 minute. The slides were th en coverslipped using cytoseal (Richard-Allen Scientific) Anti-STAT-1 Immunostaining The paraffin was removed from the tissue se c tions using the following solvent series: xylene for 5 minutes 2 times, 100% ethanol for 2 minutes 2 times, 3% H2O2 in methanol for 10 minutes, 95% methanol for 3 minutes, 70% meth anol for 1 minute, water for 1 minute, TBSTween buffer for 5 minutes. Antigen retrieva l was performed in Antigen Retrieval Citra Solution (Biogenex, San Ramon, CA) for 30 minutes in a steamer then rinsed in TBS buffer 2 times for 5 minutes. The tissues were blocke d using Sniper serum (Biocare Medical, Concord, CA) for 15 minutes followed by 2 washes in TB S for 5 minutes. Rabbit anti-Stat I antibody (Abcam, Inc., Cambridge, MA) was diluted in Antibody Diluent (Zymed, Invitrogen, Carlsbad,

PAGE 61

61 CA, USA) at 1:500 and incubated for 1 hour at room temperature followed by 2 washes in TBS for 5 minutes. Normal rabbit i mmunoglobulin (Vector Laboratori es, Burlingame, CA) was used as a negative control, diluted in Antibody Dilu ent (Zymed, Invitrogen, Carlsbad, CA, USA) and incubated on tissues for one hour at room te mperature followed by 2 washes in TBS for 5 minutes. Secondary antibody, Mach 2 Rabbit-HRP Polymer (Biocare Medical), was incubated on tissues at room temperature for 30 minut es and then rinsed in TBS quickly. Color development was performed with Cardassian DAB chromagen (Biocare Medical) for 5 minutes then slides were washed with water to stop the color development. The tissues were counterstained with Hematoxylin (Vector Laboratori es) for 1 minute, washed with water, placed into TBS for 1 minute to develop the blue color and rins ed with water. The tissues were then dehydrated by soaking in 80% ethanol for 1 minute, a quick dip into 95% ethanol, followed by a 30 sec soak in 95% ethanol, 100% ethanol 2 times for 1 minute and, a soak in xylene 2 times for 1 minute. The slides were then coverslippe d using cytoseal (Richard-Allen Scientific). Analysis of Slides and Photography Results are representative of reads from at leas t 2 slides from different animals. All slides were analyzed by 2 separate indi viduals and results compiled. Photographs of the tissues were taken using an Axioskop2 mot plus upright micr oscope (Zeiss) with th e color camera AxioCam HRc (Zeiss). All photos were processed usi ng AxioVision 4.2 program (Carl Zeiss Vision, Zeiss).

PAGE 62

62 A B Figure 2-1. Generation of VV E3L::gfp. A) Diagram of the VV E3L::gfp locus. B) Agarose gel of the VV E3L::gfp virus to ensure complete deletion of the E3L locus and correct placement of the gfp gene. Lane 1 and 7 contain the 1kb ladder by NEB. Lanes 2 to 6 are are wild t ype virus; lanes 8-12 are VV E3L::gfp; lanes 13 to 17 are rVV E3L::gfp. lanes 2-4, 8-11 and 13-15 are th e left, right and internal regions of E3L that only amplify with the presence of wild type E3L locus; lanes 5-6, 11-12 and 16-17 are the left and right flanks amp lified using both the genome primers and internal gfp primers that only amplify wh en the deletion fragment is present.

PAGE 63

63 A B Figure 2-2. Generation of VV K3L::gfp A) Diagram of the VV K3L::gfp locus. B) Agarose gel of the VV K3L::gfp virus to ensure complete deletion of the K3L locus and correct placement of the gfp gene. Lane 1 contains the NEB 1kb ladder. Lanes 2 to 6 are amplified using wild type VV; lanes 7 to 11 are amplified using VV K3L::gfp. Lanes 2-4 and 7-9 are the wild type locus amplified the left, right, and internal sequences only present in the wild type gene. Lanes 5-6 and 10-11 are the the left and right flanks amplified using both the genome primers and internal gfp primers that only amplify when the de letion fragment is present.

PAGE 64

64 Figure 2-3. Mouse genotyping. Agarose gel of mouse genotyping for all mouse constructs. Lanes 1 and 6 contains the NEB 1kb ladder, la ne 2 is a wild type copy of the RNaseL gene, lane 3 is the knockout mouse copy of the RNaseL gene. Lanes 4 and 5 show the wild type PKR locus primers when amplif ied from a wild type mouse (lane 4) and the PKR knockout mouse (lane 5). Lanes 7 and 8 are amplified using the PKR knockout primers with a wild type mouse in lane 7 and the PKR knockout mouse in lane 8. A PKR gene deletion is determined by the absence of a band using the wild type PKR locus primers (lane 5) and the presence of a band using the PKR knockout primers (lane 8).

PAGE 65

65 CHAPTER 3 PATHOLOGY OF VACCINIA VIRU S IN WILD TYPE MICE Introduction Vaccinia virus (VV) has been reported to cause d isease in a variety of mice using a number of infection routes including intr anasal (IN), intratracheal (IT), intradermal of bot h the ear pinn, two of which, the intranasal (IN), intratrachea l (IT), intradermal (ID) inoculation of the ear pinnae (Tscharke & Smith, 1999) or footpad, scarific ation of the tail, intr aperitoneal (IP), and intracranial (IC). (Figur e 1-6A) The IN and IT routes ar e intended to result in primary lung infections.(Wilcock, Duncan et al., 1999) VV inf ection by the IN route has been studied in mice and shown to exhibit an upper resp iratory infection with systemic lethal disease at a dose of 106 pfu. (Brandt, Heck et al., 2005;Brandt & Jacobs 2001;Turner, 1967) The IT route of infection has been reported to cause similar disease with 10 to 100 fold less virus required. (Amanda Rice, Amy MacNeill, Unpublished data) Reproducibility has been a problem with the IN infections because mice have a tendency to expel the inoculum resulting in a variable infectious dose and the sequestration of the inoculum in the nasal pa ssages rather than the lung.(Brandt & Jacobs, 2001) IT infections have been documented to be more reproducible delivering a defined amount of virus for each mouse infection directly to the lung. (Amy MacNeill, Unpublished data) C57BL/6 mice infected by the IT route will be examined in this study for a baseline determination of disease caused by VV. C57BL/6 mice have been selected for the wild type control because the knockout mice us ed in later sections of this study are on this background. The C57BL/6 mice are also known to have a cell mediated response to virus infections, rather than a humoral or antibody response found in Balb/C mice.(Smith & Kotwal, 2002) A cell mediated response has been shown to be important for controlling and cl earing virus infections.

PAGE 66

66 It is necessary to have a thorough disease baseline for comparison of the disease in knockout mice with both VV and deletion viruses examined later in this study. The disease outline includes survival and determination of the LD80, virus dissemination through the animal as assayed by tissue titers, host responses at th e primary site of infection determined by both histopathology and microarray analysis, a nd clinical symptomology documentation. Results Survival Studies of Wild Type Mice Infected with Vaccinia Virus To f irst establish baseline survival parameters of VV infection of wild type mice via the IT route, mice were inoculated with VV diluted to 104 to 106 pfu in PBS to a total volume of 30l by the intratracheal route (IT) of infection. Animals were then observed for 10 days and daily measurements of temperature, body weight, and overall clinical appearance were noted. No animals were allowed to die of natural causes; therefore the time of death indicated on the survival curves is the time in which an animal had to be euthanized due to severe disease. Animals were euthanized when they reached 30% or more weight loss as compared to the starting weight, body temperatures dropped from 36 to 30C or le ss, or the animal exhibited severe respiratory distress as measured by open mouth breathing or gaspin g. A clinical scoring criteria chart was developed from the observations made in this study. (Table 3-1) This chart allows for a numerical value to be given to the clinical appearance of the animals and describes specific criteria for euthanasia of an infected animal. The survival curve for wild type mice is shown in Figure 3-1 and a summary of the doses and with survival percentages are shown in Ta ble 3-2. A dose dependent response was observed in this survival study in that as the virus dose increased a higher percentage of animals had to be euthanized following our criteria. Clinical sy mptoms were more severe and appeared sooner with higher doses of virus. The experimental dose that led to 80% eu thanization rate, or LD80,

PAGE 67

67 was determined to be 1x106 pfu. This experimentally determined dose is comparable to values previously published with VV inoculated into wi ld type C57BL/6 mice via the IN route.(Brandt, Heck et al., 2005;Brandt & Jacobs, 2001;Turner, 1967) The average day to euthanization was determined to be 4.95 days with all mice euthani zed by day 6, also consistent with previously published data.(Brandt, Heck et al., 2005) This LD80 dose of 106 pfu will be utilized in all future studies for wild type mice including all time courses of infection with VV and histological analysis. Clinical Symptoms in Wild Type Mice Infected with Vaccinia Virus It has been p reviously reported that weight loss serves to be an accurate indicator of disease in the animal, and that has been demonstrated in this study as well.(Thompson, Turner et al., 1993) Mice that received le thal doses of virus exhibite d a rapid, continual drop in body weight up to 30% as compared to the beginni ng body weight of the animal, serving as an indication of severity of disease. This weight loss severity and timing at which it occurred was dependent upon the dose of virus the animal recei ved. (Figure 3-2B) Although it is difficult to ascertain the exact cause of the animals dramatic weight loss over a period of 4 days (days 3 to 6) it is most likely due to anorexia and dehydrat ion. As the animals body weight decreased the body temperature also decreased as shown in Figure 3-2A, demonstr ating that temperature also serves as valid a measure of disease. Body temperature is measured by a microchip placed subcutaneously into the mouse at the time of infection. This method of temperature measure may not represent a true measure of core body te mperature late in inf ection; however it does serve to be an adequate predictor of di sease severity and likelihood of survival. Mice that succumbed to infection had a distinct disease presentation i llustrated in Figure 33. The mice demonstrated clinical symptoms at 4 days post infection until euthanasia that included hunched posture, failure to groom, nose, facial and head swelling, respiratory distress,

PAGE 68

68 weight loss, and decrease in body temperature. The hunched posture, failure to groom, and face and head swelling is shown in Figure 3-3BII and III. Respiratory distress in the animals is characterized by open mouth breathing or trembli ng when the animal is attempting to breathe. Occasionally eye secretions and lower abdominal swelling with intestines filled with gases at necropsy have been observed. Induction of Interferon Synthesis in Infected Lung Tissue Vaccinia virus is known to suppress the immune response to viral infection, and one of the first responses to infection is the up regulation of type I interferons. IFN is expressed by all cell types in response to tissue damage or virus infection quickly while IFN is expressed at high levels only after priming of the cells. Theref ore to measure the imme diate response to viral infection the total amount of IFN was quantified. ELISAs on w hole lung homogenate from animals infected with 106 pfu of VV was performed to detect the levels of IFN in the tissue as a function of total lung weight. Similar studies with other IFN activators, such as InfluenzaA, observe high levels of induction (Jewell, Vaghefi et al., 2007). Mice were infected with 106 pfu of VV via the IT route, which is the LD80 dose for wild type mice. Animals were then observed for clinical signs of illness until the previously prescribed day of euthanasia a nd tissue harvest. Lungs were re moved at the time of euthanasia, flash frozen to prevent any protein degradati on, and stored at -80C until homogenization and assaying. Levels of IFN in the lung tissue were markedly low, staying below 225 pg/mg tissue for every day assayed except for a small peak at day 3 of 500 pg/mg tissue as shown in Figure 34. This lack of activation of IFN as compared to values reported in other studies with average levels of 600 to 12000 pg/ml of IFN (Jewell, Vaghefi et al., 2007) in the infected lung tissue reinforces the hypothesis previously reported data that VV is able to control the host response to virus infection at multiple points in the IFN response pathways. (Seet, Johnston et al., 2003)

PAGE 69

69 Virus Dissemination from the Site of Inoculation Viremia is defined as the spread of virus from one site to another and characterized in the anim al as the successful infec tion at a local site, replicati on, and spread throughout the body to distal sites. This determination of the efficiency of the virus to spread within the host animal is one indicator of the virulence of the virus. It has been previous ly reported that mice infected with VV via the IN route exhibited high leve ls of viremia and spread throughout the animal.(Brandt, Heck et al., 2005; Brandt & Jacobs, 2001) In order to determine the baseline viremia of wild type mice infected with VV via the IT route, a time course measuring the virus spread was performed. These values for the spread of VV in wild type mice are utilized throughout the remainder of the study for comparison of knockout mouse constructs infected with VV as well as mice infected with de letion viruses used later in this study. Animals received 106 pfu of VV IT and were then sacrificed at days 2 to 6. Ten mice per day were harvested. The tissues harvested fo r evaluation from each mouse included lung, liver, spleen, gonad, and brain. After flash freezing of the tissues and processing as described in Chapter 2 the samples were then plaqued on CV-1 cells to determine the virus levels found in each tissue. Table 3-3 shows the levels of viru s found in each of these organs over time. From the virus titers it can be determined that VV replic ates in the lungs of wild type mice as indicated by the 100 fold increase in virus titer from th e total pfu inoculated in the lungs. The dissemination to other organs occurs at early time points and the virus continues to grow to high titers suggesting that VV can replicate efficiently within all the tissues sampled. There was no difference found between males and females for viru s spread or titers of virus in the organs harvested (data not shown) and levels of viru s in either the male or female gonads was comparable. This was in some disagreement with published data suggesting that VV replication

PAGE 70

70 was enhanced in ovaries. Our data showed that bo th testis and ovaries are equally susceptible to VV infection. Microarray Analysis of Vaccinia Virus Infected Lung Tissue Microarrays allow for the analysis of thousa nds of transcripts within a given RNA sample in a single experime nt, therefore making them a ttractive tools and generate large amounts of data quickly. The arrays can either be based on a competitive hybridization with two samples labeled with different dyes that give a ra tio of the level of a transcript as compared to a control or a single sample hybridization that assigns an absolute number to transcript levels which then must be compared to a control sample array. The dua l dye competitive arrays allow for the direct comparison between two samples to determine the genes differentially expressed, such as when used to determine the differences in expression between two closely related tumors or cancers; whereas the single sample hybridization arrays allow for a broad comparison for multiple treatments. Dual dye arrays often exhibit a hi gh level of variation betw een arrays due to the nature of which they are printed while the single sample arrays exhibit a high consistency due to the method of printing. While both methods have their uses, the single sa mple hybridization will be used in this study.(M cShane, Shih et al., 2003) This technology has also been utilized in the context of poxvirus infect ed cells to identify key genes that undergo an expression change afte r infection. Several stud ies have analyzed the changes of host cell transc riptional changes in res ponse to poxvirus infection in vitro These studies have utilized different cell lines and viruses, therefore identifying different gene profiles.(Guerra, Najera et al., 2007;Laassri, Meseda et al., 2007;Gr inde, Gayorfar et al., 2007;Ludwig, Mages et al., 2005;Guerra, Lopez-Fe rnandez et al., 2004;Brum, Lopez et al., 2003) One key feature of the microarrays perfor med with wild type vi ruses is the overall shutdown of the host transcript le vels observed within 4 hours of infection.(Brum, Lopez et al.,

PAGE 71

71 2003) This is most likely a function of using a high MOI to ensure synchronistic infections of the cells and necessitated a lternative methods of anal yzing data obtained from in vitro studies. (Brum, Lopez et al., 2003) While this informati on is useful it does not allow for extrapolation to the responses observed in animals. Infections in animals have a cell popu lation that constantly changes with the recruitment lymphocytes to the site of infection. It is also impossible to have all the cells in a given tissue of an animal inf ected simultaneously, so there are subpopulations of cells at varying stages of infec tion that give different profiles as well as neighboring uninfected cells that may or may not be responding to signals sent from the infected cells. The question of what changes in gene expression in the lung tissu e are observed and whether they occur early in infection was investigated in this study. Mice infected with 106 pfu were sacrificed at days 0 to 5 and whole lung with the major bronchial tissue were harvested for microarray analysis. The lung tissue harvested was flash frozen to preserve the RNA integrity and RNA isolated using Trizol. Three female mice per time point were used for this study. While both males and females exhibited identical susceptibility to VV infection only females were us ed to eliminate the possibility of sex related transcriptional differences. All mice exhibite d expected clinical sy mptoms before being sacrificed. The Affymetrix mouse 430_2 microarray was used for analysis of host gene transcriptional differences over the course of infection. The mouse 430_2 array has 45,101 probe sets that represent 39,000 tr anscripts and splice variants of 34,000 genes. The arrays were normalized according to the intensity across the arra y, with all the arrays exhibiting no decrease in transcript levels over time as had been previously observed with studies in tissue culture. This was due to not all the cells in the lung being infected as occurred with the in vitro studies. One

PAGE 72

72 replicate array for day 3 was excluded from the analysis due to an overly high background level of intensity. After normalization the arrays were assigned to groups based upon the day of harvest and subjected to an F-test for change in expression over time in BRB Array tools. This analysis gave 1811 probe sets that were differentia lly expressed over time at a significance of p= 0.001. Based upon the number of probe sets that would by chance be identified as differentially regulated, or the false discovery rate, the 1811 pr obe sets were more than the 450 genes that would be identified by chance. A heat map showing the relative up or down regul ation as compared to the average values for the probe sets identified in the F-test as be ing differentially expressed over time is shown in Figure 3-5. This heat map shows a clear diffe rence in expression between days 3 and 4 post infection. This change in expression profiles o ccurs at approximately the same time that the clinical symptoms of disease begin in the animals. Probe sets of particular interest to this study included IFN IFN OAS, RNaseL, PKR and the subseque nt signaling molecules in the IFN response pathways targeted by VV. The probe sets identified were from a wide variety of pathways with no consensus for up or down regulation within a pathway. Software from Inguinity and the BioCarta classifications in BRB Array Tools were used for the pathway cla ssifications were used to examine the relative regulation changes within pathways. Specific cla sses of probe sets were analyzed to determine whether the changes in expression patterns could be attributed to any pathway or cell type. Classifications of immune responses that were ex pected were used in th is analysis including T and B cell genes and those identified to be invo lved in an immune response. Probe sets classified to be members of T-cell and B-cell sp ecific pathways (BioCarta) were clustered in BRB Array Tools and no defined patterns were identified in either, Figures 3-6 and 3-7,

PAGE 73

73 respectively. The heat maps do not show a cha nge in expression that is consistent over time; rather the regulation is random and sporadic. There may be probes within these analyses that will prove to be crucial in future studies, how ever using these methods they are not readily identifiable. The class of probes identified to be part of the immune system response pathways in BRB Array Tools in Figure 3-8 do show a si milar pattern to the ove rall global change in which there are 2 opposite shifts observed. This da ta suggests that the response is highly focused within the immune response pathways, but it is s till not easily identifiable as to the most important genes or pathways. Probe sets specific for the genes of interest for the IFN activation pathways outlined in Chapter 1 that showed differential expression pattern s were identified and are listed in Table 3-4. The type I interferon re ceptor (IFNAR1) and IFN receptors were both down regulated over the course of the arrays. CD14, the membrane prot ein that is involved in TLR3 mediated IFN up regulation, was shown to be up re gulated over time. The ATF pr oteins that are transcription factors that up regulate the e xpression of IFN response gene s by response to dsRNA were up regulated over time, however the proteins upstrea m of ATF were not shown to be differentially expressed. PKR was not shown to be di fferentially expressed over time, but eIF2 was shown to be slightly up regulated. OAS2 and OAS3, bot h in the OAS/RNaseL pathway responding to dsRNA were shown to be highly up regulated in the lung tissue over time; however RNaseL did not show a change in expression. STAT-1 and STAT-2 both showed an increase in transcript level over time. STAT-1 and ST AT-2 are transcription enhancers that are activated by activation of the type I IFN receptor that act by binding to IFN stimulated response elements (ISRE) within the promoters of interferon res ponse genes. (Figure 1-1)

PAGE 74

74 STAT-1 Staining of Lung Tissue Validation of the microarray data change in gene expression between days 3 and 4 of infected lung tissue was perform ed using immunohistochemistry (IHC) to determine if gene expression correlated to an increas e in protein levels within a cell. Often RNA levels do not correspond to protein levels due to transcript longevity and level of translation, so to determine if the up regulation of transcript le vels correspond to protein levels in the tissue staining of lung tissue for STAT-1 total protein was performed. The lung environment also changes over the course of infection and it was unclear as to whether the changes in the microarray expression profiles were attributed to infiltrating cells. ST AT-1 is a transcriptional factor activated in the interferon pathway and important for the response of a cell to viral in fection. STAT-1 was identified in the microarray data as up regulated in 3 probes that represent STAT-1 between days 3 and 5 by an average of 3.94 times (Table 3-4) and was chosen for the IHC analysis. The antibody used for this IHC binds to both active and inactive STAT-1 in order to determine the overall up regulation of host cell translation. Lung tissue isolated from mice at days 3 a nd 5 post infection with VV were fixed in formalin and embedded in paraffin for sectioning. Both active and inactive STAT-1 protein was detected by IHC and photographs of representative sections are located in Figure 3-9. STAT-1 staining is shown by the presence of a brown colo r in the tissue with nuclei counterstained blue with hematoxylin. The 3 day post infection lung s ections clearly have STAT-1 present in most cells exhibiting a light mostly cytoplasmic stai ning pattern similar to that of the uninfected sections. However the staining of the 5 day post infection sections is more intense than that observed in both the uninfected and day 3 post in fection sections and de monstrates a different pattern. At day 5 all the epithelial cells are st aining in both the nucleus and cytoplasm darkly while the infiltrating cells (neutrophils, macropha ges, etc) show no staining as shown by the

PAGE 75

75 absence of brown color. This not only suggests th at the changes in expre ssion patterns correlate to protein levels in the tissue but are due to th e global tissue response not to the influx of immune cells to the site of infection. Histopathology of Infected Mice The histopathology of infected animal tissues allows for an in depth view of what responses are occurring at the tis sue and cellular level. Cells infected with VV were detected by the use of a VV expressing gfp under the poxvi rus synthetic early late promoter, VV ATI::gfp. This virus ha s been examined in animals (dat a not shown) and is not attenuated, with mice exhibiting all the same clinical symptoms and LD80 dose observed with wild type VV. While spread of VV was observed at high levels thr oughout the mice by tissue titer, histopathological analysis was performed on lung, liver and spleen tissue. Both H&E and immunohistochemistry (IHC) for detection of gfp were performed on all 3 tissues from at least 2 different mice at day 3 and 5 post infection. The H&E shows the overall appearance of the tissue and for identification of cell types while IHC staining detected of vi rally infected cells by the presence of the gfp protein expressed by the virus. Day 3 lung sections of wild type mice surprisi ngly showed a very localized and restricted infection by IHC with VV inf ection of lung epith elial cells surroundi ng the bronchi and bronchioles. There was no staining of the alveolar type I or type II pneumonocytes observed at day 3. (Figure 3-10A) The H&E of the lung tissue showed modest edema around areas of infection with infiltrating neutrophils. (Figure 3-10B) The liv er sections showed no pathology and no staining with IHC. The spleen secti ons were of normal pathology with no staining by IHC for VV infected cells. Day 5 histopathology showed local areas of VV infection that were centralized around bronchi with staining in the al veolar tissue surrounding these br onchi following IHC staining of

PAGE 76

76 lung sections. (Figure 3-10C, 3-10E) Macrophage in filtrates were also observed to be positive for VV infection by IHC. Both alveolar type I and II pneumonocytes were shown to be infected by IHC as were blood vessel epithelial and endothe lial cells. H&E of thes e lung sections showed severe edema around areas of infection with ma crophage and neutrophil infiltrate. (Figure 310D, 3-10F) Blood vessel integrity was not maintained and was mo st likely the source of the edema. Liver sections showed no pathology on H&E staining (Figur e 3-11B); however few epithelial cells were positive for VV infection by IHC (Figure 3-11A, 3-11C). Spleen sections showed evidence of an immune res ponse with neutrophillic and his tiocytic aggregates within the follicles. (Figure 3-11D, 3-11E) Few cells in the spleen stained positive for VV by IHC. Tissues harvested for histopathology were wei ghed prior to processi ng to determine gross weight. Figure 3-12A shows the ratios of body weight to lung weight from days 2 to 6 post infection. As the infection progresses the ratio s of body weight to lung weight decreases as compared to the uninfected PBS treated animals wh ile the ratio of lung weight to spleen weight increases over time. The weight of the lungs from infected animals as compared to PBS animals is shown to increase by more than 2 times in Figure 3-12C and 3-12D. This increase in lung weight over the course of the infection correlates with the appearance of both immune cell infiltrates and edema in the lung tissue. The spleen weights did not change over the course of infection while the body weight decreased, conf irming that the spleen composition does not change over the course of infec tion as seen with histopathology. Discussion VV infection of wild type C57BL/6 mice wa s shown to produce severe disease in which 80% of m ice had to be euthanized when given 106 pfu via IT route. The disease is characterized by a rapid weight loss that occurs simultaneously with a body temperature decrease. A decrease

PAGE 77

77 in grooming and face and head swelling late in the disease are also characteristic of VV infection of wild type mice. The primary site of infection in this infecti on is the lung, of which the virus infected cells are focused around the bronchi. The infection su rprisingly was localize d, not the generalized infection pattern that wa s expected from placing 106 pfu of VV direct ly into the lungs of mice. Although the lung infection was main tained as a localized infecti on, global tissue responses were observed in both transcription levels as measured by microarrays and protei n levels as measured by STAT-1 immunohistochemistry. This localiz ed infection caused edema around the infected bronchi and the influx of immune cells to the lung tissue. The influx of immune cells suggests a wide spread immune response to infection, howev er the response was not caused by an increase in IFN as measured by whole lung ELISA. The lack of IFN up regulation is consistent with VV controlling the host immune response to infection. The spread of virus from the lung is rapi d, showing dissemination to the brain, liver, spleen, and gonad within 2 days from infection. The levels of virus found in the lung are constant over the course of inf ection, suggesting that any virus made in the lungs is rapidly disseminated to the other organs. This rapid virus dissemination is consistent with the small localized areas of infection obser ved in the IHC lung sections. Since virus replication is not restricted to the lung nor does it cause the massive amount of tissue damage that would be expected from a direct inoculat ion into the lung, it can be hypo thesized that the disease and clinical symptoms observed are due to a systemic infection of the animal. This description of wild type C57BL/6 mice infected with VV will be utilized as a reference point for all the future experiments.

PAGE 78

78 Figure 3-1. Survival curves for wild type mice following infection with VV. Survival curve for wild type mice infected with VV-WR at th e doses indicated. The number of mice per treatment is notated in the Figure legend. Th e overall significance for the curves was p< 0.001 by the Kaplan Meier survival analysis. The Holm-Sidak method for multiple comparisons was used to determine significance between the curves. All comparisons were found to be significant at p 0.0008 except the 105 pfu curve compared to the 106 pfu curve.

PAGE 79

79 Table 3-2. Summary of Survival Data for W ild Type Mice Infected with Vaccinia Virus Dose (pfu) Live/Total % Survival PBS 25/25 100 104 19/30 80 105 10/36 27.3 106 7/43 16.3 The total virus dose, number of survivors ove r the total number of mice infected, and the calculated percent survival are indicated.

PAGE 80

80 A B Figure 3-2. Average body temperature and wei ght loss curves of w ild type mice following infection with VV. Number of mice for each of the measurements is shown in the legend for the survival curve gra ph in Figure 3-1. Curves for the 106 dose were terminated at day 6; there were 7 survivors that were not taken in to account after day 6. A) Average body temperature of the mice infected with VV-WR. Error bars represent SEM. B) Average weight lo ss over time of the mice infected with VVWR. Weight loss was calculated from beginni ng weight of the mice at day 0. Error bars represent SEM.

PAGE 81

81 A B Figure 3-3. Clinical progression of VV disease in mice. A) Time line of the progression of disease. The red boxes repres ent the symptoms for which an animal is euthanized, the blue other clinical symptoms obser ved. B) Photographs of the clinical progression. I) A healthy, uninfected mouse with a normal head profile. II) A mouse infected with VV at day 5 just before euthanization with the head swelling, face swelling, open mouth breathing, and failure to groom. III) A whole body photo of a mouse infected with VV at day 5 that is exhibiting face swelli ng, eye secretions, a hunched posture, and failure to groom.

PAGE 82

82 Table 3-1. Clinical scoring cr iteria for poxvirus infected mice. Observation Points Assessment Examples Posture 0 Normal Ears up, moving freely. 1 Abnormal Some reduction in spontaneous activity; will move with gentle encouragement but prefers to be inactive. Change in posture. Scruffy. 2 Very abnormal Little to no spontaneous activity. Frumpy. 3 Severe Hunched. Attitude 0 Normal Bright & alert, interested in surroundings. Usual temperament. 1 Abnormal A bit dull but sti ll active in investigating new situation. 2 Very abnormal Very dull, not interested in surroundings, preoccupied. Hydration 0 Normal No tent. 1 Abnormal Tent, but returns to shape after <30 seconds. 2 Very abnormal Tent and does not return to shape readily. Eyes 0 Normal Normal. 1 Abnormal Conjunctivitis either unilateral or b ilateral may be occasional nodules. Eyes are not closed. 2 Very abnormal Severe conjunctivitis with purulent discharge. Eyes may be partially closed. Lids swollen. Nose 0 Normal Normal. 1 Abnormal Slight swelling. 2 Very abnormal Severe swelling. Respiration 0 Normal Normal. 2 Abnormal Mild difficulty breathing. 4 Very abnormal Open mouth or gasping. Head 0 Normal Normal. 1 Abnormal Slight swelling 2 Very abnormal Severe swelling Temperature 0 Normal 35-36F 1 Abnormal 36-39F/34-32 F 2 Very abnormal +40F/below 31.9 F Weight 0 Normal Constant weight gain 1 Abnormal No weight gain 2 Very abnormal Weight loss Viremia 0 Normal No secondary lesions present 2 Abnormal Secondary lesions present, minimal 4 Very abnormal Secondary lesions present, severe Animals with 13-15 points require euthanasia. Any animals exhibiting weight loss >30% or severe respiratory distress require immediate euthanasia.

PAGE 83

83 Figure 3-4. Interferon beta levels in the lung tissue of VV infected wild type mice. Total IFN as measured by Elisa. 10 mice per time point were assayed. Error bars represent SEM. Horizontal dotted line is the limit of detection for the assay.

PAGE 84

84 Table 3-3. Virus spread of VV in wild type mice Tissue Day 2 3 4 5 6 Lung 6.12 6.49 6.53 6.45 5.92 Liver 3.55 5.05 6.39 7.75 7.57 Spleen 4.51 5.68 5.95 6.15 6.05 Gonad 3.72 4.39 5.15 5.31 7.29 Brain 5.82 6.58 6.45 6.64 7.26 Virus spread of wild t ype mice infected with 106 pfu as measured by virus titers. Numbers represent log10 values of the average titers for each organ at the specific time point. n 10 mice per time point.

PAGE 85

85 Figure 3-5. Global expressi on profile changes in infected lung tissue of differentially expressed probe sets. Microarray analys is heat map of probes diffe rentially expressed over time in lung tissue of inf ected mice as determined by an F-test performed on the data. 3 samples per time point, except for day 3 where one array was not included in the analysis due to processing failure. Each horizontal line represents a single probe expression profile. Vertical collections of lines represent the overall expression profile for a sample. Blue represents dow n regulation of the probe compared to the average expression value, white is the average expression value, and red is up regulation of the probe as compared to the average expression value.

PAGE 86

86 Figure 3-6. Expression profile of B cell specific genes. The e xpression profile was performed on the entire data set for probes identifi ed as B cell specific genes in BRB Array Tools. Each horizontal line represents a single probe e xpression profile. Vertical collections of lines represent the overall expression profile for a sample. Blue represents down regulation of the probe co mpared to the average expression value, white is the average expression value, a nd red is up regulatio n of the probe as compared to the average expression value. The bright red line of probes in the third sample in day 1 is due to a bright secti on of the array and no t considered in the analysis.

PAGE 87

87 Figure 3-7. Expression profile of T-cell specific genes. The e xpression profile was performed on the entire data set for probes identifi ed as T cell specific genes in BRB Array Tools. Each horizontal li ne represents a single probe expression profile. Each horizontal line represents a single probe expression profile. Vertical collections of lines represent the overall expression prof ile for a sample. Blue represents down regulation of the probe compared to the average expression value, white is the average expression value, and red is up regul ation of the probe as compared to the average expression value. The line of bright red in the third column under the day 1 set is not real up regulation, it is a bright area on the array, and therefore adds no information for those genes from that individual sample.

PAGE 88

88 Figure 3-8. Expression profile of probe sets identified as members of the immune response pathway. The expression profile was perf ormed on the entire data set for probes identified as immune response specific ge nes in BRB Array Tools. Each horizontal line represents a single probe expression pr ofile. Each horizontal line represents a single probe expression profile Vertical collections of lines represent the overall expression profile for a sample. Blue re presents down regulation of the probe compared to the average expression value, white is the average expression value, and red is up regulation of the probe as compared to the average expression value.

PAGE 89

89 Table 3-4. Expression values fo r differentially expressed genes in the IFN activation pathways. Gene Day 0 Day 1 Day 2 Day 3 Day 4 Day 5 p-value ATF3 210.3 257.7 291.5 164.2 778.6 773.8 0.00069 ATF4 2169 1737.6 2788.9 2127.4 4593 3977.7 0.000138 CD14 821.5 688.8 1079.9 961.7 1638.4 2214.8 0.0001174 eIF2b5 680.6 683.3 563.8 859.8 966.3 856.7 0.000975 IFN gamma receptor 417 541.9 266.5 286.7 195.2 128.3 0.000684 IFNAR1 (IFN / receptor) 342.1 351.9 290.7 257.7 50.7 144.6 0.000953 OAS2 191.1 195.1 211.1 159.4 677.2 1201.4 0.000393 OAS3 28.9 27.8 80.1 161.9 585.3 1012.5 9.50E-06 STAT-1 636.7 626.6 875.2 985.4 2930.7 3302.8 1.00E-07 STAT-1 622.6 683.5 1196.4 1447 4825.2 6195.1 2.00E-07 STAT-1 592.4 650.6 857.2 936 2892.8 3591.2 1.90E-06 STAT-1 453.8 576.2 1203 991.9 1674.5 2022.9 0.000391 STAT-2 117.4 108.5 117.4 168.7 457.6 533 0.000972 Genes differentially expressed that are found in the IFN activation pathwa ys of interest. The values for each probe set repres ents the standardized fluorescen ce from the array. The p-values were calculated in BRB Array Tools. STAT-1 was represented on the array 4 times and each probe set showed a change in expression pattern.

PAGE 90

90 Figure 3-9. Immunohistochemitry staining of STAT-1 protein in wild type mouse lung tissue. Immunohistochemistry staining of STAT-1 pr otein in lung tissue of infected animals at day 3 and 5 post infection. Nuclei are stained with hematoxylin and appear blue; STAT-1 staining detected with HRP and a ppears brown. Star shows location of a bronchi. A) Uninfected l ung tissue at 20X magnification. B) Uninfected lung tissue at 40X magnification. C) Day 3 lung ti ssue at 20X magnifica tion. D) Day 5 lung tissue at 20X magnification. Arrow shows area of infiltrating cells and debris. E) Day 3 lung tissue at 40X magnification. F) Day 5 lung tissue at 40X magnification. Star shows location of a bronchi.

PAGE 91

91 Figure 3-10. Immunohistochemist ry and H&E staining of wild type mouse VV infected lung tissue. A) Day 3 IHC stained lung sec tion at 5X magnification. Arrow shows bronchi infected with VV. B) Day 3 H&E lung section at 5X magnification. C) Day 5 IHC stained lung section at 5X magnification. Arrow show s the site of infection. D) Day 5 H&E lung section at 5X magnificat ion. E) Day 5 IHC lung section at 20X magnification. Arrow shows an area of edem a. F) Day 5 H&E lung section at 40X magnification. BV shows the location of a blood vessel; E shows an area of edema.

PAGE 92

92 Figure 3-11. Immunohistochemist ry and H&E staining of liver and spleen tissue from VV infected wild type mice. Arrows show ar eas of positive staining for virus infection. A) Day 5 IHC stained liver section at 10X magnifica tion. B) Day 5 H&E liver section at 10X magnification. C) Day 5 IHC stained liver section at 100X magnification. D) Day 5 H&E spleen sect ion at 10X magnification. Follicles are dark blue in color. E) Day 5 IHC spl een follicle section at 40X magnification.

PAGE 93

93 Figure 3-12. Tissue weights for VV infected wild type mice. Graphs of the ratios of lung, spleen, and body weights for mice harves ted for histopathology. Animals were infected with 106 pfu of VV and harvested at the times indicated. 3 animals per time point were used. PBS animals are used as the normal reference. A) Graph of the ratios of body weight to lung weight for mice. B) Graph of the ratios of lung weight to spleen weight for mice. C) Ratio of infected lungs to PBS lungs at the day indicated. D) Average rati o values for the infected l ung, spleen, and body weight to uninfected animal lung, spleen, and body weight over time.

PAGE 94

94 CHAPTER 4 PATHOLOGY OF VACCINIA VI RUS IN KNOCKOUT MICE Introduction PKR and RNaseL are important intracellular proteins that respond to the presence of dsRNA and up regulate interferon response pathways (Garcia, Meurs et al., 2007) VV has been shown to control these two pathways in vitro through the production of both a dsRNA binding protein (E3L) (Kibler, Shor s et al., 1997) and an eIF2 homologue (K3L) (Carroll, Elroy Stein et al., 1993). Since the virus spec ifically targets these pathways m ice deficient in either PKR or RNaseL has been hypothesized to be more sus ceptible to VV infection. To examine this hypothesis PKR or RNaseL single knockout mice, and a double knockout PKR/RNaseL (DKO) mouse were obtained from Bryan Williams and Robert Silverman. These mice were created on the C57BL/6 mouse background, the RNaseL knoc kout mouse from the Jax Labs C57BL/6 (Zhou, Paranjape et al., 1999), and the PKR knockout mouse on the Taconic C57BL/6 background (Yang, Reis et al., 1995 ). Although background strains of mice can give different responses to various pathogens, this was determined not to be the case with VV infections by the IT route by examination of the sensitivity of both sources of mice with VV. (Data not shown.) PKR and RNaseL single knockout mice and the PKR/RNaseL double knockout (DKO) mice have been previously examined for their responses to VV infection.(Xiang, Condit et al., 2002b) This study reported that all three mouse constructs res ponded the same as the C57BL/6 wild type animals, succumbing to virus disease at a dose of 106 pfu via the IN route.(Xiang, Condit et al., 2002b) This study did not perform a complete dose response study with analysis of disease such as the one done here with a complete analysis of disease in these knockout animals. Detailed examination of the survival and dis ease caused in the knockout animals will allow for us to elucidate the roles PKR and RNaseL play in controlling VV infections of the lung. The

PAGE 95

95 disease outline includes survival and determination of the LD80, virus spread through the animal as assayed by tissue titers, host responses at the primary site of infection determined by histopathological analysis, and cl inical symptomology documentation. Results Survival Studies of Knockout Mice Infected with Vaccinia Virus Survival studies were performed on all the knockout mouse constructs to determine the sensitivity of each knockout mouse to VV infec tion. It was hypothesized that the single knockout mice, RNaseL and PKR, would exhibit an increased sensitivity as compared to wild type mice to VV infection; while the DKO mice would exhibit a greater sensitivity to VV infection than that of the single knockout mice. Seven week old knockout mice were infected with VV at doses of 103 to 106 pfu in 30l of PBS by the IT route of inf ection. These mice were monitored twice a day and underwent complete physicals daily that included weight and temperature measurements and overall appearance. No animals were allowed to die of natural causes, therefore the survival curves are the time in which an animal had to be euthanized due to severe disease. Animals were euthanized when they reached 30% or more wei ght loss as compared to the starting weight, body temperatures dropped to 30C or less, or the an imal exhibited severe respiratory distress. The PKR and RNaseL knockout mice were determined to exhibit simila r sensitivities to VV, with 105 pfu causing approximately 80% of the animals to be euthanized due to severe disease. (Figure 4-1) This was 10 fold less viru s required to cause the same level of disease in wild type mice (Chapter 3). The DKO mice e xhibited an increased sensitivity to VV as compared to the single knockout mice (Figure 4-1), taking 104 pfu to cause disease that necessitated euthanasia of approximately 80% of the animals. All the animals exhibited clinical

PAGE 96

96 symptoms of disease like that of the wild type mice infected with VV shown in Chapter 3 (Figure 3-3A). A comparison of all mouse constructs at the construct specific LD80 dose was performed using the data from all the separate su rvival studies. This is shown in Figure 4-2A. All the survival curves reach th e 20% survival line at the same rate except for the DKO mouse. This was determined to be significant when analyzed in Sigma Stat by the Holm-Sidak method for pairwise multiple comparis ons. The DKO mouse survival cu rve was significantly (p=0.05) different from the other 3 curves. The differences in the rate of euthan asia are shown in Figure 4-2B in which the mean day to death for each mouse construct at their individual LD80 dose was calculated. Wild type, RNaseL and PKR single knockout mice all exhibited a mean day to death of 4.5 to 5 days while the DKO mouse had a calc ulated mean day to death of 6.85 days. The mean day to death was calculated using the sl ope of the survival curve over time for that particular virus dose. In an attempt to explai n the extended survival of the DKO mice, clinical profile data for the mice was examined. Clinical Symptoms in Knockout Mi ce Infected with Vaccinia Virus The RNaseL mice temperature and weight loss profiles for mice infected with 104 to 106 pfu are shown in Figure 4-3. RNaseL mice do not e xhibit a severe drop in temperatures as seen with the wild type mice. RN aseL mice infected with the LD80 dose of 105 pfu have no temperature deviations from that of PBS mice. The weight loss curves, however, demonstrate a dose response related to the amount of input virus. As the inoc ulum increased in total pfu the severity and rapidness of the weight loss increased. Approximately 20% weight loss was observed in RNaseL mice infected with the LD80 dose of 105 pfu. This weight loss as a measure of disease is consistent with both previously pu blished data and the data with wild type mice infected with VV.

PAGE 97

97 The profiles for temperature and weight loss for PKR knockout mice infected with 104 to 106 pfu are shown in Figure 4-4. PKR mice exhi bited a dose dependent response in both temperature decrease and weight loss curves. This was in contrast to RNaseL mice in that did not show a decrease in temperature as a function of virus input. Approximately 20% weight loss was observed with PKR mi ce infected with the LD80 dose of 105 pfu. The temperature and weight loss pr ofiles for DKO mice infected with 103 to 105 pfu of VV are shown in Figure 4-5. The infected DKO mice, like the RNaseL knockout mice, do not exhibit a severe temperature drop to 30C, and mice infected with the LD80 dose of 104 pfu have no temperature drop as compared to the PBS an imals. The weight loss exhibited by the DKO mice is similar when compared to the total weight loss exhibited by wild type animals infected with 106 pfu; animals take longer to exhibit the we ight loss. Approximate ly a 25% weight loss was observed in animals infected with the LD80 dose of 104 pfu. This increase in survival of the DKO mice may be in part due to their increased time to lose a sufficient amount of weight to necessitate euthanasia and the ability to maintain a normal body temperature during infection. Induction of Interferon Synthesis in Infected Lung Tissue The level of IFN activation was mea sured in whole l ung homogenate of mice infected at their experimentally determined LD80 doses is shown in Figure 4-6. The IFN levels for wild type mice infected with 106 pfu that was shown in Chapter 3 but are also included here to allow for comparisons with the knockout mice. As shown previously wild type mice have a peak in IFN production at day 3 where as both single k nockout mice constructs, PKR and RNaseL, had a peak a of IFN levels at day 4. The level of IFN detected in RNaseL mice was 2 times that of the PKR mice, but not different fr om wild type mice. The DKO mi ce also exhibited a peak of IFN production one day later on day 5 with leve ls similar to those of the PKR mice.

PAGE 98

98 While none of the mice exhibited high levels of IFN the data does allow for some suggestions as to the roles that RNaseL and PKR play. This data suggest s that by knocking out a gene responsible for detecting and responding to virus infections compensatory pathways are delayed as compared to a normal response. This is evident in the one day delay for the peak of IFN for both single knockout mice, while the do uble knockout mice took two days longer than the wild type mice for the IFN levels to peak. The le vel of activation of IFN was decreased in the PKR knockout mice as compared to the RNaseL mice suggesting that PKR is more important for the activation of IFN Virus Dissemination from the Site of Inoculation To determine the level of dissemi nation of the virus in the knockout mouse constructs virus titers of various organs was performed. Each mouse construct wa s infected with their previously determined LD80 virus dose and then 10 mice were sampled each day from day 2 to 6 post infection. The outcome of all the mouse constructs infected was the same, giving a clinically identical outco me of euthanasia. Samples of the lung, liver, spleen, gonad, and brain were removed from the animals, flash frozen, an d then processed for titering as described in Chapter 2. Titers of pfu/mg of tissue were determined, averages of the 10 replicates calculated, and the log10 values of the averages are represented in Table 4-1. The values for wild type mice from Chapter 3 are also included in Table 4-1 to allow for comparisons to the knockout mice. Virus replication in the lung was similar for all mouse constr ucts even though the single knockout mice received 10 fold less virus than the wild type mice a nd the DKO mice received 100 fold less virus than the wild type mice. This suggests that RNaseL and PKR are important in controlling virus infections for when they are deleted the virus is able to replicate more efficiently as observed in the knockout animals as compared to the wild type. The virus spread and replication in the liver was determined to be similar in the wild type, RNaseL, and PKR mice

PAGE 99

99 reaching 6-7 logs of virus within 6 days. Th e DKO mice, however, had low levels of virus detected in the liver, only reaching 3.65 logs by da y 6. This may be related to the increased time to euthanasia observed in the DKO mice, that the overall viral load may not be sufficient to cause severe disease at the same time as the other mice. A similar pattern was also observed in the spleens of the 4 mouse constructs. The wild t ype and RNaseL and PKR single knockout mice all had 6 logs of virus by day 6 while the DKO mice had 4 logs of virus. The virus levels detected in the gonads of all the animals was similar regardless of input, with 6 to 7 logs found in both ovary and testis. There were no differences observed between the virus titers found in the ovary or testis, suggestin g that there is no sex bi as for VV. The spread and replication in the brains of infected animals was also the same at day 6 with 6 to 7 logs of VV detected by day 6 post infection. The fact that both the brain and gonad are immune privileged sites in the body seemed to have he lped the virus replicate in the DKO mice for the decrease in amount of virus seen in the liver and spleen were not obser ved in these tissues. STAT-1 Staining of Lung Tissue The levels of STAT-1 were dete rmined for DKO mi ce infected with 104 pfu VV. This was performed to determine whether the DKO mice were able to generate a global response to the VV infection like the wild type mice have been shown to in Chapter 3. The DKO mouse lungs exhibited more overall st aining for STAT-1 at day 3 than the wild type mice shown in Figure 4-7. The entire lung stained positive for STAT-1 with the infiltrates excluded from staining. Day 5 IHC for STAT-1 showed no increase in staining in the DKO mice. This suggests that while the DKO mice ma y have more STAT-1 protein, this protein is insufficient to cause the up regulation of the interf eron response genes.

PAGE 100

100 Histopathology of Infected Mice The histopathology of lung, liver, and spleen was performed on DKO m i ce infected with 104 pfu on days 3 and 5 post infection. Histopat hology of RNaseL and PKR mice infected with 105 pfu on lung, liver, and spleen was also performed on day 5. Lungs from DKO mice on day 3 showed mild edema with few infiltr ating neutrophils at the site of infection on H&E. (Figure 4-8B) IHC for VV detection via gfp staining showed localized infections around the bronc hi as had been previously seen in the wild type mice at day 3 (Figure 4-8A), there are however, fewer sites of infection that are mo st likely due to the 100 fold decrease in virus that was us ed to inoculate the DKO mice. Day 5 H&E of the lungs showed edema around the bronchi with macrophage infiltrate. (Figure 4-8D, 4-8E) The IHC showed larger area s of infection in the lung, still focused around the bronchi and relatively few in number. (Figur e 4-8C, 4-8F) The lungs of PKR mice at day 5 showed severe edema with neutrophils on H&E a nd the same focal infection pattern the bronchi and areas of edema and infiltrate seen in both DKO and wild type mice (data not shown). Day 5 of the RNaseL lungs showed edema and infiltrate like that of the PKR and wild type mice with IHC staining undistinguishable from the other mice (data not shown). Liver histology from DKO mice at days 3 and 5 and PKR and RNaseL single knockout mice at day 5 had neutrophilic foci of necrosis. (Figure 4-9) There were a few randomly stained cells on IHC for detection of VV, however, they were isolated and not representative of the overall sample set. The spleen sections at day 3 from DKO mice showed an increase in histiocytes within the lymphoid follicles (Figur e 4-9) with no staining for virus infected cells with IHC. Day 5 spleen sections on H&E showed both lymphoid depletion and follicular hyperplasia. (Figure 4-9) Spl een sections from both the PKR and RNaseL single knockout mice at day 5 were normal on H&E and had no staining on IHC staining (data not shown).

PAGE 101

101 At the time of tissue harvest the lung and sp leen weights were recorded. The previously shown data for wild type mice from Chapter 3 will also be included in these Figures for allowance of comparison to knockout mice. Fi gure 4-10A shows the body to lung weight ratios for all 4 mouse constructs. The ratios of body we ight to lung weight for wild type and DKO mice show a general decrease, with the slope of the DKO mouse line being shallower than that of the wild type mice. The PKR and RNaseL mice were only harvested on day 5, however, the values of body weight to lung weight for these si ngle knockout mice are in the middle of the wild type and DKO mouse values. When the lung to spleen weight ratios are graphed for all 4 constructs (Figure 4-10B) there are large differences observed between the DKO and wild type mice. The wild type mice show an increase in th e ratio of lung to spleen weight over time, while the DKO mice exhibit no change at all in the ratios. This is due to two factors, one is that the lung weights do not increase as much as with the wild type animals and the second is that the spleens of the DKO mice weight more than the wild type mice before infection. The change in lung weights over time is shown in Figure 4-10C with all 4 mouse constructs. This comparison to PBS lung weight s is quite informative and generally correlates with histology results for the presence of edema and cell infiltrate. The wild type mice have a general increase in lung weight as compared to PBS lung weight over time, more than doubling the lung weights at day 6. The DKO mice show no difference in the weight of infected lungs to PBS lung weights. Both the RNaseL and PKR an imals have an increase in lung weight as compared to PBS animals, but it is not as severe as the wild type animals, but is in the middle of the DKO and wild type animals at day 5. The va lues for this graph are located in the lung column of Figure 4-10D, which is a table of l ung, spleen, and body weights as compared to PBS animals for all the mouse constructs.

PAGE 102

102 The differences of DKO from the wild type animal s seen in this data are most likely one of the sources of the longer survival of the DKO mice. These numbers located in Figure 4-10D demonstrate that the DKO mice are not experiencing the same level of disease as the other mice at day 6, supported by the presence of less edema in th e lungs of DKO animals on histopathological examination. Discussion The importance of both RNaseL and PKR in c ontrolling the VV infection in mice has been shown from the increased sensitivity to VV of the knockout mice. The question as to whether one gene is more important in controlling the infe ction than the other has also been answered in this study. It has been shown that both genes co ntribute to the sensitiv ity of the DKO mice and that the deletion of both RNaseL and PKR creates an additive effect This additive effect was observed in the survival study in which the DKO mice were two logs more susceptible to VV while both of the single knockout mice were 10 fold more susceptible to VV than wild type mice. The lack of temperature decrease in the DKO mice was most likely due to the absence of the RNaseL gene, since the RNaseL knockout mous e failed to have a temperature response to VV infection while the PKR knockout mice did exhibit these expect ed decreases. Virus titers in the knockout animals reach levels similar to that of wild type mice infected with VV even at 10 to 100 fold less virus inoculat ed. This suggests that the deletion of PKR or RNaseL allows for more efficient virus replic ation and spread within the animals, again supporting the hypothesis that viru s replication is controlled by PKR and RNaseL. The reason for the decrease in virus load in the liver a nd spleen in the DKO mice was not explained by the single knockout animal analysis. The DKO mice survived 2 to 3 days longer than the single knockouts and the wild type mice, therefore give n enough time the virus titers may have reached

PAGE 103

103 levels similar to those observed in the single knockout and wild type animals. However, an overall decrease in immune reac tion was observed in both the IFN assays and the staining for STAT-1 in the lung tissue in the DKO mice. Theref ore, one explanation for the lack of virus in the liver and spleen beyond that of survival is that with a poor immune response there are fewer white blood cells circulating through the animal able to infiltrate the lung ti ssue. The majority of the virus recovered from the live r and spleen is most likely bl ood cell associated and not from replication within the liver and sp leen. This is determined from the lack of numerous foci of infected cells in these organs that would be expected if the virus was actively replicating in them. The absence of infiltrating cells in the lung tissue that the virus would normally use for travel through the body are in lower numbers, thereby limiti ng the movement of the virus to the liver and spleen. The virus that does leave the lung then in fects immune privileged sites such as the brain and gonad, where it is able to grow to hi gh titers. The brain and gonad were specifically targeted by VV for the primary viremia. This wa s not completely unexpect ed as the isolate of VV used in these experiments is known to be ne urotropic and poxviruses have been identified to target gonads for infection. (Turner, 1967) The delayed responses in the PKR and RNas eL single knockout mice was observed in the IFN Elisa data in which the modest peaks of IFN production were one day later than wild type mice. The DKO mice also exhibited a modest peak of IFN production that was two days later than wild type mice. This delay in reaction is most likely due to the level of tissue involvement needed in the lung before compensatory viral response pathways were able to react.

PAGE 104

104 Figure 4-1. Survival curves of knockout mouse cons tructs with wild type VV. A) Survival for DKO mice. Overall signifi cance of the curve was p < 0.001 by the Kaplan Meier survival analysis. B) Survival curve for RNaseL mice. Overall significance of the curve was p < 0.001 by the Kaplan Meier survival anal ysis. C) Survival curve for PKR mice. Overall significance of the curve was p < 0.001 by the Kaplan Meier survival analysis. Each curve was analyzed usi ng the Holm-Sidak method for pair wise multiple comparison with an overall significan ce level p = 0.05. The survival curves for each mouse construct were analyzed for significance between the curves and all were found to be significant except for the DKO mice in the comparison between the PBS and 103 pfu dose mice.

PAGE 105

105 A B Mouse construct LD80 dose (pfu) Mean day to euthanization Wild type n=43 1x1064.95 PKR n=21 1x105 4.81 RNaseL n=25 1x105 4.51 DKO n=36 1x104 6.85 Figure 4-2. Comparison of surv ival of all mouse constructs with VV. Summary of VV LD80 virus dose survival data for all mouse constr ucts. A) Survival curves for all mouse constructs at the construct specific LD80 virus dose. Using the Kaplan Meier survival anaylsis it was determined that the si gnificance for the curves was p=0.012. There was no significance found using the Holm-S idak method for multiple comparisons between the wild type an d single knockout animals or the DKO and PKR mice. There was a difference between the DKO mice and wild type mice of p=0.0006 and between the DKO mice and the RNaseL mi ce of p=0.002. B) Table of all mouse constructs mean day to euthanization.

PAGE 106

106 A B Figure 4-3. Average body temperature and weight loss of RNaseL mice infected with VV. Number of mice per treatment is located in the Figure legends. The 105 and 106 lines were terminated at days 4 and 6 respectiv ely, and the surviving mice not included in the analysis. Error bars represent SEM. A) Average temperatures of RNaseL mice infected with VV. B) Weight loss for mice infected with VV.

PAGE 107

107 A B Figure 4-4. Average body temperat ure and weight loss of PKR mice infected with VV. Number of mice per treatment is located in the Figure legends. The 105 and 106 lines were terminated at days 5 and 6 respectively, a nd the surviving mice not included in the analysis. Error bars represent SEM. A) Average temperatures of PKR mice infected with VV. B) Weight loss for mice infected with VV.

PAGE 108

108 A B Figure 4-5. Average body temperature and weight loss of DKO mice infected with VV. Number of mice per treatment is located in the Figure legends. The 105 graphs were terminated at day 6, the 1 surviving mouse was not included in the analysis. Error bars represent SEM. A) Average temperatur es of DKO mice infected with VV. B) Weight loss for mice infected with VV.

PAGE 109

109 Table 4-1. Virus spread of VV in all mouse constructs. Mouse Construct Tissue Day 2 3 4 5 6 Wild type Lung 6.12 6.49 6.53 6.45 5.92 DKO Lung 5.39 5.18 5.74 6.04 6.10 RNaseL Lung 4.90 5.66 5.58 5.88 4.79 PKR Lung 4.78 5.49 5.69 5.55 5.58 Wild type Liver 3.55 5.05 6.39 7.75 7.57 DKO Liver 2.99 3.07 3.87 3.28 3.65 RNaseL Liver 2.94 4.53 5.37 5.44 6.34 PKR Liver 2.76 5.20 5.52 6.41 6.89 Wild type Spleen 4.51 5.68 5.95 6.15 6.05 DKO Spleen 2.56 2.84 4.79 4.15 4.16 RNaseL Spleen 4.67 5.26 5.83 5.49 5.93 PKR Spleen 4.14 5.49 6.20 6.55 6.64 Wild type Gonad 3.72 4.39 5.15 5.31 7.29 DKO Gonad 4.37 1.46 5.03 4.22 6.07 RNaseL Gonad 3.87 3.81 5.03 5.77 7.36 PKR Gonad 3.71 4.80 5.33 6.02 6.20 Wild type Brain 5.82 6.58 6.45 6.64 7.26 DKO Brain 6.53 3.46 5.89 6.60 6.06 RNaseL Brain 3.61 4.63 6.77 6.37 6.66 PKR Brain 4.24 5.25 6.58 6.89 7.22 Virus spread in all mouse constructs infected with VV. The data for wild type mice presented in Chapter 3 is also included as a comparison. Values represent log10 of the average values for each tissue on each time point taken.

PAGE 110

110 Figure 4-6. Interferon beta levels in the lung tissue of infected mice. Wild type and DKO mice had 10 mice per time point; RNaseL and PKR mice had 3 mice per time point. Bars represent average values with SEM. Data for wild type mice previously shown in Chapter 3 is included for comparison.

PAGE 111

111 Figure 4-7. Immunohistochemistry staining of STAT-1 protein in lung tissue of infected animals at day 3 and 5 post infection from DKO mice. Nuclei are stained with hematoxylin and appear blue; STAT-1 staining detected with HRP and appears brown. Stars show location of bronchi. A) Uninfected DKO lung tissue at 20X magnification. B) Uninfected DKO lung tissue at 40X magnifi cation. C) Day 3 lung tissue at 20X magnification. D) Day 5 lung tissue at 20X magnification. E) Day 3 lung tissue at 40X magnification. F) Day 5 l ung tissue at 40X magnification.

PAGE 112

112 Figure 4-8. Immunohistochemistry and H&E staining of infected lung tissue from DKO mice. B denotes bronchi, BV denotes blood vessels E shows areas of edema. A) Day 3 IHC stained lung section at 5X magnification. B) Day 3 H&E lung section at 5X magnification. C) Day 5 IHC stained lung se ction at 5X magnific ation. D) Day 5 H&E lung section at 5X magnification. E) Day 5 IHC lung section at 40X magnification. A shows the alveolar tissue staining positive for virus infection. F) Day 5 H&E lung section at 20X magnification.

PAGE 113

113 Figure 4-9. Immunohistochemist ry and H&E staining of liver tissue from VV infected DKO mice. F shows the sites of follicles within the spleen tissue. A) Day 3 IHC stained liver section at 5X magnification. B) Da y 3 H&E liver section at 40X magnification. The arrow shows the site of neutrophilic in filtrate. C) Day 5 IHC stained liver section at 5X magnification. D) Day 5 H& E spleen section at 5X magnification. E) Day 5 IHC spleen section at 5X magnificat ion. F) Day 5 H&E sp leen section at 20X magnification.

PAGE 114

114 Figure 4-10. Tissue weights of mi ce infected with VV. Graphs of the ratios of lung, spleen, and body weights for mice harvested for histopat hology. Animals were infected with the mouse construct specific LD80 dose of VV and harvested at the times indicated. 3 animals per time point were used. Wild type mouse data previously shown in Chapter 3 is included for reference. A) Graph of the ratios of body weight to lung weight for mice. B) Graph of the ratios of lung weight to spleen weight for mice. C) Ratio of infected lungs to PBS lungs at th e day indicated. D) Average ratio values for the infected lung, spleen, and body we ight compared PBS lung, spleen, and body weight over time for each mouse construct.

PAGE 115

115 CHAPTER 5 PATHOLOGY OF VV E3L::GFP IN WILD TYPE AND KNOCKOUT MICE Introduction The VV encoded E3L protein was initially identified as a ge ne that dictated IFN sensitivity, that when E3L was deleted VV did not replicate in cells in the presence of IFN.(Beattie, Kauffman et al., 1996;Brandt & Jacobs, 2001) The protein was later identified to be a dsRNA binding protein of 190aa in size. The dsRNA binding domain is located from amino acid 118 to 186 at the C-terminus of the prot ein.(Chang & Jacobs, 1993;Ho & Shuman, 1996) The dsRNA binding domain is required for viru lence in animals and infection of most mammalian cells in tissue culture and viruses la cking E3L exhibit a seve re host range.(Brandt, Heck et al., 2005) The N-terminus region of th e protein was later determined to encode a ZDNA binding domain. This Z-DNA binding domain that reaches from amino acids 5 to 70 and is required for full virulence in animal models, but is not required for virus replication in tissue culture.(Brandt & Jacobs, 2001) The protei n domains are diagramed in Figure 5-1. E3L has been implicated in controlli ng the host immune response to dsRNA by binding and sequestering of the dsRNA produced during VV infections. This protein has also been suggested to play a role separate from that of K3L in inactivating PK R, but this role has yet to be fully investigated.(Kibler, Shors et al., 1997) VV deleted for E3L has previously been shown to be highly attenuated in animals. To determ ine the importance of the E3L protein in VV infection, a virus with E3L deleted from VV generated (VV E3L::gfp), see methods for details. A deletion virus had to be generated due to th e variation in lab isol ates of VV for a true comparison to our wild type VV. The red sh aded box in Figure 5-1 shows the area of E3L deleted in VV E3L::gfp, removing the majority of bot h domains necessary for virulence in animals and tissue culture. This shaded area deleted was replaced with gfp under the poxvirus

PAGE 116

116 synthetic early late promoter. This VV E3L::gfp virus is was then used to also determine the interaction of E3L with PKR and RNaseL and th eir respective pathways by infection of wild type, PKR and RNaseL single knockout, and DKO mice. Results In Vitro Data of VV E3L VV E3L had been previously reported to exhibi t a severe host range only replicating in BHK-21, CEF, and RK13 cells. The VV E3L::gfp virus generated for these studies was also analyzed for host range. For the host range st udy virus grown in the permissive BHK-21 cell line was purified and tittered on BHK-21 cells to determine the baseline titer. The titer f or VV E3L::gfp on BHK-21 cells was 2.81x108. The virus was then tittered on various cell lines from pigs, rabbits, humans, and monkeys in which the parental virus grows and replicates with no decreases in titer. VV E3L::gfp did plaque on most cell lines as shown in Table 5-1, however the titer was decreased by at least 5 logs. This decrease of 5 logs or more suggests that the VV E3L::gfp does have a host range phenotype confirming the previously published data.(Beattie, Kauffman et al., 1996;Brandt & Jacobs, 2001) Rescue of VV E3L::gfp VV E3L::gfp was rescued by recombining the wild type E3L gene back into the natural E3L locus, r eturning the deletion to a wild type virus to ensure that no other unintended mutations were present in the de letion virus that cont ributed to any phenotypes observed in tissue culture or animal studies. The rescue was accomplished by infection of BHK-21 cells with VV E3L::gfp and transfection with a linear PCR pr oduct of the entire E3L gene amplified from the parental virus. The resulting virus mixtur e was harvested and then plaqued on PK15 cells to screen for viruses that had integrated the wild type gene product into the E3L locus and the loss of gfp. This rescue was relatively straight fo rward and easy due to th e host range phenotype

PAGE 117

117 exhibited by VV E3L::gfp allowing for selection of virus containing the wild type E3L gene on the non permissive cell line PK-15, as VV E3L::gfp only grows on the highly permissive BHK21 cell line. The re sulting virus, rVV E3L::gfp, was then amplified and purified for evaluation in animals. rVV E3L::gfp was used to infect 10 PKR and 6 wild type mice at 105 and 106 pfu, respectively, via the IT route. These mice infected with RVV E3L::gfp exhibited severe systemic disease and required the euthanization of approximately 80% of infected animals. rVV E3L::gfp caused disease identical to that caused by VV in these animals, with body temperature decrease and weight loss, abse nce of grooming, hunched posture, head and face swelling, and respiratory distress ne cessitating euthanasia (data not shown). From these results it was determined that VV E3L::gfp had successfully been rescued to full virulence and that any effects observed in tissue culture or animals was i ndeed the result of the de letion of the E3L gene and not a random unintended mutation elsewhere in the virus genome. Survival Studies of Mice Infected with VV E3L::gfp Wild type mice inf ected with VV E3L::gfp were hypothesized to be resistant to the virus and able to control the infection. These animals were expe cted to exhibit few symptoms of disease because of the intact immune system th at includes wild type copies of both PKR and RNaseL. The PKR knockout mice were expected to have responses similar to that of the wild type mice since these mice still have a wild type copy of the RNaseL gene to mount responses to the presence of free dsRNA. The RNaseL knoc kout mice were hypothesized to exhibit a slight resistance to the VV E3L::gfp virus, to require one to two logs more of virus to cause disease because these mice are lacking the main protein that E3L was thought to target. With the lack of both the host genes RNaseL and PKR in the DKO mouse was hypothesized to exhibit susceptibility similar to that of wild type VV. This would occur if the soul targets for E3L were

PAGE 118

118 PKR and RNaseL, however it was recognized that there are redundant pathways. The DKO mouse was hypothesized to exhibit a response similar to that of the RNaseL knockout mice, with a response to VV E3L::gfp that was similar to that of a VV infection. These responses would be due to the mice lacking both RNaseL and PKR, th e main and secondary targets, respectively, for E3L. Survival curves for all four mous e constructs is shown in Figure 5-2. Wild type mice were infected with 106 to 109 pfu of VV E3L::gfp with all mice at all virus doses surviving. (Figure 5-2A) RNas eL knockout mice were infected with 105 to 109 pfu of VV E3L::gfp and also had 100% survival of all mi ce at all doses. (Figur e 5-2C) These results were not as expected if RNaseL was the majo r responder to dsRNA within a cell, a slight attenuation was expected, but at high doses of virus it was expect ed that the mice would become ill and succumb to disease. The fact that the mice did not succumb to disease simply reinforces the idea that the host responses to infectio ns are redundant, and capable of controlling an infection. The PKR knockout mice were infected with 105 to 108 pfu of VV E3L::gfp and exhibited a 100% survival of all mice infected with 105 to 107 pfu. 20% of PKR mice infected with 108 pfu, however, were euthanized due to severe disease. (Figure 5-2D) DKO mice were infected with 104 to 108 pfu, of VV E3L::gfp with 100% of all mice infected with 104 to 106 pfu surviving. At a dose of 107 pfu, 70% of the DKO mice were euthanized due to seve re disease, and at 108 pfu 90% of mice exhibited severe disease requiring euthanization. (Figure 5-2B ) The increased susceptib ility of the DKO mice to VV E3L::gfp as compared to wild type and both single knockout s was expected. This result was expected because the deletion of both PKR and RNaseL, th e major responders to dsRNA in a cell were absent in these mice. Current literature has fail ed to report the generatio n of such a disease in animals. The studies examined mi ce infected with doses of up to 106 pfu, in which the results

PAGE 119

119 from the DKO mice do agree. Had the prev ious studies examined higher doses of VV E3L in mice similar results may have been observed. Clinical Symptoms in Mice Infected with VV E3L::gfp Wild type mice infected with 106 to 109 pfu of VV E3L::gfp exhibited no changes in body temperature at any dose, but a weight loss of up to 15% was seen at the 108 and 109 pfu doses. (Figure 5-3) The weight loss s uggests an illness at high doses th at is most likely due to the excessively high titer of virus used for infection of the animals. The survival of these mice was as expected since both PKR and RNaseL are pr esent and can respond to the unmasked presence of dsRNA in infected cells. RNaseL single knockout mice exhibited no chan ges in body temperature over time and a weight loss of up to 10% was seen at the highe r doses. (Figure 5-4) These measurements suggest a mild disease in which the animal is able to adequately respond and control virus infection. PKR single knockout mice did not exhib it a change in body temperature even at high doses of virus, but did exhibit weight loss. Up to 8% weight loss was observed in mice infected with 107 pfu while mice infected with 108 pfu exhibited weight loss of up to 20%. (Figure 5-5) The weight loss exhibited by the mice infected with 108 pfu supports the se verity of disease caused in the animals that were euthanized. Th is result was surprising because these mice still have the OAS/RNaseL pathway intact and it was theorized that E3L was produced by VV to inhibit the OAS/RNaseL de gradation of RNA. DKO mice that were euthanized due to severe disease exhibited similar clinical symptoms to mice infected with VV, failure to groom hunched posture, head and face swelling, and respiratory distress. Mice infected with 104 to 106 pfu exhibited no signs of disease with no temperature changes or severe weight loss. At a dose of 107 pfu, however, a slight decrease in body temperature was observed with up to 20% wei ght loss, indicating seve re disease. Severe

PAGE 120

120 weight loss of 28% and temperature decreases typical of moribund animals was observed at a dose of 108 pfu. (Figure 5-6) Virus Dissemination from the Site of Inoculation Virus spread as measured by tissue tite r was perform ed on animals infected with VV E3L::gfp. This experiment was to establish whether the virus was ca pable of controlling the host immune system enough to allow for spread and replication in the tissues similar to that observed with wild type VV. Mice were infected with VV E3L::gfp via the IT route of infection at the mouse construct specific VV LD80 dose previously presented in Chapter 4. Wild type mice infected with 106 pfu of VV E3L::gfp exhibited a generalized virus clearance in the lungs between days 2 and 6 post in fection. (Table 5-2) This virus clearance was not complete by day 6 most and this was most likely due to both a low leve l of virus replication and the maintenance of virus particles that did not infect lung cells at th e time of inoculation, but just remained in the lung tissue. Spread from the lung tissue was not observed in wild type mice (Table 3-3), supporting the survival and clinical data that these mice were not severely ill with a systemic infection. RNaseL and PKR single knockout mice were infected with 105 pfu of VV E3L::gfp and tissues harvested at days 2 to 6 post infection. While both single knockout mice cleared the virus from the lung tissue over time, th e RNaseL mice were able to clea r the virus faster than the PKR mice. This more rapid clearance of virus by more than a log a day correlates with the survival data in which RNaseL mice were more resistant to VV E3L::gfp than the PKR mice. VV E3L::gfp was unable to spread from the l ung tissue of either mouse construct, again indicating that there are redundant compensatory pathways that are able to control the virus infection.

PAGE 121

121 DKO mice were infected with 104 pfu of VV E3L::gfp and tissues harvested at days 2 to 6 post infection. The lung tissue of DKO mice support the maintenance of VV E3L::gfp over time at the level of input virus until day 5, most likely due to low level replication within the lung tissue. This is not an unexpected result, when considering the survival data in which DKO mice were susceptible to VV E3L::gfp. These mice would most likely clear the virus infection over time based on the lack of disease seen in the survival data for DKO mice infected with 104 pfu of VV E3L::gfp in which all mice survived with no clinical symptoms. The virus was unable to exit the lung tissue in th ese mice at this virus dose, agai n consistent with the animals controlling the virus infection. DKO mice were shown to be susceptible to VV E3L::gfp at high doses and the animals that were euthanized due to severe disease exhi bited clinical symptoms similar to those observed in VV infected animals. The question as to whether the disease caused by VV E3L::gfp was the same as that of VV in viremia and tissue titers. 3 DKO mice, 2 female and 1 male, were infected with 109 pfu of VV E3L::gfp and the disease allowed to progress until euthanasia. These animals were then necropsied and tissues removed for virus titer at day 5. The results from these mice are shown in Table 5-3. The titer of VV E3L::gfp in the lungs of these animals was higher than that observed with wt VV, with 8 logs of VV E3L::gfp in the lungs of the animals sampled. The animals did exhibit spread, but at lo wer levels than was observed with wt VV, 4 to 5 logs to the liver, spleen, gonad, and brain. Although disease was clinically similar to that caused by wt VV, the amount of VV E3L::gfp in the lung and lower levels of spread to distal sites distinguished this disease from that of wt VV (Table 4-1). STAT-1 Staining of Lung Tissue STAT-1 is known to be important in the IFN response signaling pathway by serving as a me mber of the transcriptional complex that up re gulates the IFN response genes. This gene was

PAGE 122

122 identified in microarray analysis as being up regulated late in infection and has been used to determine the type of response, lo cal versus global, in the lung tissue of infected mice. Based upon clinical symptoms and the su rvival curves it was hypothesized that the infected animals were mounting a successful immune response against VV E3L::gfp and therefore levels of STAT-1 staining should be hi gh at early time points. Lungs from wild type mice infected with 108 pfu, DKO mice infected with 104, and DKO mice infected with 108 pfu of VV E3L::gfp via the IT route of in fection were isolated, fixed in formalin for 12-18 hours, and imbedded in para ffin for processing and tissue sectioning. Sections were stained with the anti-STAT-1 antibody and detected with the chromogen DAB that gives a brown color when reacting with the anti body. Nuclei were counterstained blue with hematoxylin to allow for determination as to the cell type staining positive for STAT-1 presence. Lung sections from days 3 and 5 post infection were used for this study since it has been established in the wild type m ouse study with VV there is a larg e difference in both transcript and protein levels of STAT-1 between these days (Chapter3). Wild type mice infected with 108 pfu of VV E3L::gfp showed high levels of staining for STAT-1 at day 3 with both dark nuclear and cytoplasmic staining. Day 5 also showed a high level of nuclear staining, although no t as dark as day 3, with little cytoplasmic staining. This suggests that the virus was unable to control the host immune response that occurred quickly and the response was global as all the lung tissue stained for STAT -1. (Figure 57A&B) DKO mice infected with 104 pfu of VV E3L::gfp also showed high levels of STAT-1 staining within the nuclei at day 3. The levels of STAT-1 staining increased at day 5 with both cytoplasmic and nuclear staining remaining high. (Figure 5-7C&D) This up regulation of STAT-1 appears to be slower than wild type mice, but does suggest that the DKO mice are able to control the virus

PAGE 123

123 infection through a host response pa thway. DKO mice infected with 108 pfu of VV E3L::gfp showed very high levels of nucl ear staining with light cytoplasmi c staining at day 3 and 5 post infection. The majority of seve ral lung sections consisted of i mmune cell infiltrate and lymphoid cell proliferation that remained unstained. Howe ver the staining of the bronchial epithelial and alveolar cells for STAT-1 was strong. (Figure 5-7E&F) The levels of STAT-1 detected in the lungs of these animals suggests that the VV E3L::gfp virus is unable to control the hos t response and up regulati on of STAT-1 protein production in response to virus infection. The inf iltrating cells did not stain positive for STAT-1, but the response was rather a gl obal lung tissue response, as had been previously observed (Chapter 3 and 4). Histopathology of Infected Mice Histopathology of the lung, liver, and sp leen tissues from mi ce infected with VV E3L::gfp to determine the pathology of the virus in animals. Wild type mice infected with 108 pfu, DKO mice infected with 104 pfu, and DKO mice infected with 108 pfu of VV E3L::gfp were used in this study. Tissues were harvested from the animals, fixed in formalin overnight, and then embedded in paraffin. Each tissue was s ectioned for IHC staining for gfp to determine virus replication in the tissue a nd H&E for overall ti ssue composition. Wild type mice infected with 108 pfu of VV E3L::gfp showed no staining for virus replication in the lung tissue sections at either day 3 or day 5. (F igure 5-8A) This suggests that the virus that infected cells and replicated were killed earlier than day 3. The lack of virus replication also suggests that the response to this virus was within the first 2 days. Lung H&E of day 3 showed mild edema with few infiltrating ce lls. This edema and infiltrate was located near bronchi in a pattern similar to that seen in the VV infected sections in Chapter 3. There was also a proliferation of lymphoid tis sue observed that suggests a high host response to the virus

PAGE 124

124 infection. (Figure 5-8B, 5-8C) This again suggests that the viru s infected bronchial epithelial cells and because E3L was absent from this in fection the host responded quickly and efficiently to clear the virus. The edema and infiltrate that is seen at day 3 is most likely what is remaining after the host clearance of inf ected cells. Day 5 H&E showed some edema present, but the lung structure appeared to be return ing to normal with little infiltra te left and the lymphoid tissue proliferation. (Figure 5-8 D to 5-8F) Spleen sections from days 3 and 5 showed no staining for virus infection by IHC. H&E showed follicular hyperplasia in both days 3 and 5 post infection. (Figure5-11A &B) Liver sections for day 3 showed no staining in IHC and normal morphology in H&E, which was expected since the virus does not spread from the l ung in these animals. Liver sections from day 5 showed no virus staini ng by IHC but neutrophilic foci of necrosis and apoptosis were present on H&E. (Figure 5-11C, 5-11D) DKO mice infected with 104 pfu of VV E3L::gfp at day 3 and 5 in lung tissue exhibited no staining for virus replication by IHC. (Figure 5-9A, 5-9D) Day 3 H&E of lung tissue showed little edema with cell in filtrates and large areas of lymphoi d tissue prolifera tion located around bronchi that were probably infected initially. (Figure 5-9B, 5-9C) These areas of infiltrate and lymphoid tissue proliferation pockets are what remain after this response of the host to virus infection. This activation of an immune response demonstr ates that the DKO mice are not deficient in their responses to vi ral infections at relatively low virus doses and in fact have a more robust response that that of wild type mice. Day 5 H&E of the lung tissue sections showed relatively normal lung structures with moderate alveolar neutroph ilic infiltrate (Figure 5-9E) and hyaline membranes present (Figure 5-9F). Hyalin e membranes suggest a stressed alveolar tissue and subsequent scaring from virus infections. Liver sections from both days 3 and 5 post infection showed random small ar eas of neutrophilic necrosis. Spleen sections from day 3

PAGE 125

125 showed spleenic activation with follicular hyperplasia, while da y 5 sections showed hypoplasia. (Figure 5-11E) DKO mice infected with 108 pfu of VV E3L::gfp that are clin ically ill and exhibit symptoms that require euthanasia by day 6 we re also sampled for histopathology. Lung samples from day 3 showed staining for virus replicat ion around epithelial cells of bronchi, areas of edema, and infiltrates. (Figure 5-10A) This staining was localized as was seen in the IHC of VV infected lungs. The H&E of day 3 lungs show ed edema around the bronch i and alveolar tissue with neutrophilic infiltrate. (Figure 5-10B) IH C of day 5 lung tissue show ed increased areas of virus infection focused at the site of bronchi (Figure 5-10C) Day 5 lung sections showed extensive edema and fibrin located around infected bronchi and at the ou ter edges of lung with no virus present. These lung sections also exhibited lymphoid tissue proliferation mostly contained in small areas causing obstruction of entir e sections of the lung tissue. (Figure 5-10 D to 5-10F) Liver sections for both day 3 and 5 showed normal tissue structures by H&E, with a few epithelial cells staining positive for virus rep lication at day 3 by IHC. The spleen sections exhibited normal tissue structure by H&E at da y 3 with no staining for virus by IHC. Day 5 spleen sections showed severe lymphoid depleti on, with the lymphocytes from the spleen being having been recruited to the lung following infection. (Figure 5-11F) Tissue weights were also measured in these studies and are located in Figure 5-10. The lung weights of infected mice as compared to PBS animals showed a general increase in weight, up to twice that of the PBS animals, for both wild type and DKO mice infected with 108 pfu of VV E3L::gfp. (Figure 5-10A) This was not the case for the DKO mice infected with 104 pfu of VV E3L::gfp where the lung weight did not change as compared to the PBS animals. The spleen weights of infected animals as compared to the PBS controls were very different. (Figure

PAGE 126

126 5-10B) The wild type mice infected with 108 pfu of VV E3L::gfp showed a general increase in weight suggesting activation of the spleen where as DKO mice infected at either 104 pfu or 108 pfu of VV E3L::gfp showed a decrease in weight, sugge sting that the infilt rating cells seen in the lungs of the infected animal s were from the spleens. The body weight to lung weight ratios of mice infected with 108 pfu showed a clear difference between the wild type mice that survive and the DKO mice that are euthanized due to severe disease. (Figure 5-10C) The wild type mice show a stable ratio that does not change over time while the DKO mice show a general decrease in the ratio of body to lung weight at the 108 pfu dose. The ratio of lung to spleen weights for the in fected animals was also very informative. Figure 5-10D shows that the wild type mice infected with 108 pfu exhibited as stable ratio of 1-2 for the lung to spleen weights while the DKO mice exhibited high ratios of 3 or more. This suggests that the animals that su ccumb to infection have a massive emptying of their spleens into their lung tissue, as suggested by the H&E as well. When the DKO mice at both doses are compared in body to lung weight ratios it is seen that the mice that do no t exhibit severe disease have a stable ratio that does not change over time (Figure 5-10E) while the 108 pfu dose animals exhibit a general decrease in ratio s of body to lung weight. This has been observed in previous experiments to be an indicator of disease as well, and that a deviation from a stable ratio signifies severe disease. Discussion The VV E3L::gfp virus was found to be highly attenuated in both tissue culture and animals as previously reported. (B eattie, Kauffman et al., 1996);(Ch ang, Uribe et al., 1995) It is clear from t he animal survival data that the vi ral gene E3L is important in controlling the host response to dsRNA and interacti on with PKR. Animals respond ve ry efficiently and robustly to infection with VV E3L::gfp as observed by the up regulation of STAT-1 pr otein staining and

PAGE 127

127 by histopathological analysis. This strong immune response is most likely due to the lack of the E3L protein that normally sequesters ds RNA and prevents cellular responses. Survival data with the double PKR/RNaseL knockout mice also suggest that E3L targets and controls host responses othe r than that of PKR and RNaseL and are most likely the other cellular response pathways activ ated by dsRNA, i.e. TLR3, RIG, MDA5. These pathways are most likely secondary to that of PKR and RNaseL in a poxvirus infection as the DKO mice were able to be lethally infected at high doses of VV E3L::gfp. The comparison of the mice infected at the wt VV LD80 doses with VV E3L::gfp (Table 5-2) showed a high level of attenuation and inabi lity of the virus to replicate well in the lung tissue. This poor level of replication and lack of spread to dist al organs is consistent with previously reported virus phenotypes in mice. This data also supports th e hypothesis that E3L is critical in controlling the host immune response and that when deleted the host responds to give little disease in animals. The lethal infection any animal with VV deleted for E3L (VV E3L::gfp) has not been previously reported. We have shown that DKO mice at inoculated with 108 pfu causes lethal systemic disease that is undistinguishable clinically from a lethal wt VV infection. The differences observed were at the level of virus spread and histopathology. VV E3L::gfp replicated to high levels within the lungs of infected animals and induced a major response in which massive cell infiltrates and lymphoid tissu e proliferation were observed. The lymphoid depletion observed in the spleens at day 5 is mo st likely due to the recr uitment of cells to the lung. The level of edema observed in the lung tissue was higher than observed with wt VV lung infections in DKO mice. This increase in edema is most likely due to the major influx of immune cells to the lung thr ough the blood vessels causing damage to the vessels. In this case

PAGE 128

128 the vigorous immune response appears to highly contribute to the disease observed in the DKO animals.

PAGE 129

129 Figure 5-1. Diagram of E3L pr otein and area deleted in VV E3L::gfp. The E3L protein is 190aa long and has two domains. The N-terminal domain binds Z-DNA and is from amino acids 5 to 70. The C-terminal domain is the dsRNA binding domain that is from amino acid 118 to 186. The overlaid red box that reaches from amino acid 27 to 163 shows the area of the protein delete d and replaced with gfp under the poxvirus synthetic early late promoter in VV E3L::gfp. (Chang & Jacobs, 1993;Ho & Shuman, 1996;Kim, Lo wenhaupt et al., 2004)

PAGE 130

130 Table 5-1. In vitro titer data VV E3L::gfp host range Cell Line Virus Titer BHK-21 2.81 x 108 CV-1 1.4 x 103 RK13 3.6 x 103 PK15 4 x 102 Vero < 100 BSC40 6 x 103 A549 < 100 MCF7 2 x 102 In vitro data for host range in VV E3L::gfp. A stock of VV E3L::gfp grown and tittered on BHK-21 was tittered on various cell lines to determine the host ra nge phenotype for this virus.

PAGE 131

131 Figure 5-2. Survival Curves for mice infected with VV E3L::gfp. The inoculums doses are noted in each curve. Number of animals per dose is noted in the figure legends. A) Survival curves for wild type mice infected with VV E3L::gfp. B) Survival curves for DKO mice infected with VV E3L::gfp. The Kaplan Me ier survival analysis determined an overall significance for the survival curves of p<0.001. The HolmSidak method for multiple comparisons showed significance of p 0.00003 for the 108 pfu survival curve as compared to the other curves; and p 0.001 for the 107 pfu survival curve as compared to the other curves except the 108 pfu curve. C) Survival curve for PKR mice infected with VV E3L::gfp. The Kaplan Me ier survival analysis determined an overall significance for the su rvival curves of p=0.035. There were no significant differences between the curves in a multiple comparison by the HolmSidak method. D) Survival curve for RNaseL mice infected with VV E3L::gfp.

PAGE 132

132 Figure 5-3. Average body temperature and wei ght loss of wild type mice infected with VV E3L::gfp.

PAGE 133

133 Figure 5-4. Average body temperature and weight loss of DKO mice infected with VV E3L::gfp.

PAGE 134

134 Figure 5-5. Average body temperature and we ight loss of RNaseL mice infected with VV E3L::gfp.

PAGE 135

135 Figure 5-6. Average body temperature and weight loss of PKR mice infected with VV E3L::gfp.

PAGE 136

136 Table 5-2. Tissue titers of mice infected with VV E3L::gfp Mouse Construct Tissue Input Titer Day 2 Day 3 Day 4 Day 5 Day 6 Wild type Lung 106 3.68 3.37 2.73 2.50 2.64 DKO Lung 104 3.51 3.28 3.67 3.20 2.62 RNaseL Lung 105 2.42 2.60 2.26 1.87 <0.3 PKR Lung 105 3.86 3.71 3.61 2.26 1.78 Wild type Liver 106 <0.3 <0.3 <0.3 <0.3 <0.3 DKO Liver 104 <0.3 <0.3 <0.3 <0.3 <0.3 RNaseL Liver 105 <0.3 <0.3 <0.3 <0.3 <0.3 PKR Liver 105 <0.3 <0.3 <0.3 <0.3 <0.3 Wild type Spleen 106 <0.3 <0.3 <0.3 <0.3 <0.3 DKO Spleen 104 <0.3 <0.3 <0.3 <0.3 <0.3 RNaseL Spleen 105 <0.3 <0.3 <0.3 <0.3 <0.3 PKR Spleen 105 <0.3 <0.3 <0.3 <0.3 <0.3 Wild type Gonad 106 <0.3 <0.3 <0.3 <0.3 <0.3 DKO Gonad 104 <0.3 <0.3 <0.3 <0.3 <0.3 RNaseL Gonad 105 <0.3 <0.3 <0.3 <0.3 <0.3 PKR Gonad 105 <0.3 <0.3 <0.3 <0.3 <0.3 Wild type Brain 106 <0.3 <0.3 <0.3 <0.3 <0.3 DKO Brain 104 <0.3 <0.3 <0.3 <0.3 <0.3 RNaseL Brain 105 <0.3 <0.3 <0.3 <0.3 <0.3 PKR Brain 105 <0.3 <0.3 <0.3 <0.3 <0.3 Average log10 values for tissue titer data for mice infected with VV E3L::gfp. Mice were infected with their construct specific VV LD80 dose and tissues harvested at day 2 to 6. n 8 mice per mouse construct per time point. Lim it of detection for th e titer assay was 0.3.

PAGE 137

137 Table 5-3. Tissue titers fo r DKO mice infected with 109 pfu VV E3L::gfp Lung Liver Spleen Gonad Brain 8.04 4.91 5.25 4.29 5.78 Average log10 values for tissue titer data for DKO mice infected with 109 pfu VV E3L::gfp. 3 mice were infected with 109 pfu VV E3L::gfp and tissues harvested at the time of euthanization, day 5.

PAGE 138

138 Figure 5-7. Protein levels of STAT-1 in VV E3L::gfp infected lung tissue. Immunohistochemistry staining for STAT-1 levels of infected lung tissue. B represents bronchi, BV represents blood ve ssel, A shows the location of alveoli, L shows the location of lymphoi d tissue proliferation. A) Wild type mouse lung tissue taken at day 3 post infection; 5x. B) Wild type mouse lung tissue taken at day 5; 20x. C) DKO mouse infected at 104 pfu day 3 lung tissue; 20X D) DKO mouse infected at 104 pfu day 5 lung tissue; 20X. E) DKO mouse infected at 108 pfu day 5 lung tissue; 5x. F) DKO mouse infected at 108 pfu day 5 lung tissue; 40x.

PAGE 139

139 Figure 5-8. VV E3L::gfp histology of wild type mouse lung tissue. B shows the location of bronchi; E denotes areas of edema; BV s hows the location of blood vessels. A) Day 3 IHC stained lung section at 5X magnification. B) Da y 3 H&E lung section at 5X magnification. C) Day 3 H&E stained lung section at 20X magnification. D) Day 5 H&E lung section at 5X magnification. E) Day 5 IHC lung section at 20X magnification. Arrow shows an area of lym phoid tissue that underw ent proliferation. F) Day 5 H&E lung section at 40X magnification.

PAGE 140

140 Figure 5-9. Histology of DKO mouse lung tissue from animals infected with 104 pfu VV E3L::gfp. B shows the location of bronch i; A shows the location of alveoli. A) Day 3 IHC stained lung section at 5X ma gnification. Arrow shows area of lymphoid proliferation. B) Day 3 H&E lung section at 5X magnification. Arrow shows area of lymphoid proliferation. C) Day 3 H&E stai ned lung section at 40X magnification. Arrow shows area of lymphoid proliferation. D) Day 5 IHC lung section at 5X magnification. E) Day 5 IHC lung section at 10X magnification. F) Day 5 H&E lung section at 20X magnification. Arrows s how the location of hyaline membranes.

PAGE 141

141 Figure 5-10. Histology of DK O mouse lung tissue from animals infected with 108 pfu VV E3L::gfp. B shows the location of br onchi; BV shows th e location of blood vessels; E shows areas of edema. A) Day 3 IHC stained lung section at 5X magnification. Arrow shows infected bronc hi. B) Day 3 H&E lung section at 5X magnification. C) Day 5 IHC stained lung se ction at 20X magnification. D) Day 5 H&E lung section at 5X magnification. Arrows show areas of fibrin and edema. E) Day 5 H&E lung section at 20X magnification. AW shows the thickening of the alveolar wall; arrow shows an area of ed ema. F) Day 5 H&E lung section at 20X magnification. Arrows show the lo cation of lymphoid proliferation.

PAGE 142

142 Figure 5-11. Histopathology of sp leen and liver tissue from VV E3L::gfp infected animals. A) Spleen section at 10X magnification from wild type mice at day 3. B) Day 3 H&E spleen section at 20X magnification from w ild type mice. C) & D) Day 5 liver section at 40X magnification. Arrows show the areas of necrosis and neutrophil infiltrate. E) DKO mouse infected with 104 pfu day 5 IHC spleen section at 5X magnification. F shows the area of a follicle F) Day 5 H&E spleen section at 5X magnification from DKO mice infected with 108 pfu. F shows the site of a follicle.

PAGE 143

143 Figure 5-12. Tissue weights for mice infected with VV E3L::gfp. Graphs of the ratios of lung, spleen, and body weights for mice harvested for histopathology. 3 animals per time point were used. PBS animals are used as the normal reference. A) Graph of the ratios of body weight to lung weight for mice. B) Graph of the ratios of lung weight to spleen weight for mice. C) Body weight to lung weight ratios over time for wild type and DKO mice infected at 108 pfu. D) Lung weight to spleen weight ratios over time for wild type and DKO mice infected at 108 pfu. E) Body weight to lung weight ratios over time for D KO mice infected at 108 or 104 pfu. F) Lung weight to spleen weight ratios over time for DKO mice infected at 108 or 104 pfu.

PAGE 144

144 CHAPTER 6 PATHOLOGY OF VV K3L::GFP IN WILD TY PE AND KNOCKOUT MICE Introduction Vaccinia virus K3L was identified as a gene that when deleted from the virus caused sensitivity to interferon in vitro The role of K3L in the viru s infection has not been fully elucidated as of yet. K3L has been shown in vitro to bind PKR and to prevent the phophorylation of eIF2 and host cell translation shutdow n in response to the presence of dsRNA. K3L is expressed early in the infecti on, suggesting that it is needed for immune modulation in some fashion.(Beatti e, Paoletti et al., 1995) The prot ein has been crystallized and is known consist of 5 beta strands and 2 alpha helices to form a be ta barrel.(Dar & Sicheri, 2002) The protein is diagramed in Figure 6-1. The area shaded in red is the area deleted in the virus generated for these studies, VV K3L::gfp. The N-terminal porti on of K3L that shares homology with eIF2 and is hypothesized to bind PKR through th e same C-terminal domain where it has been determined that eIF2 binds.(Dar & Sicheri, 2002) This interaction has not been shown in vivo as of currently. In tissue culture VV deleted for K3L was reported to exhibit host range in vitro failing to replicate in BHK-21 cells. (Langland & Jacobs, 2002) VV deleted for K3L showed a 1 to 2 log attenuation in mice by the IN route of infection. Results In Vitro Data Vaccinia virus deleted for the K3L gene was evaluated in cell culture for com parison to previously reported data of VV K3L exhibiting a host range phenotype. The reported host range phenotype was that it failed to re plicate in BHK-21 cells with normal replication in other cell types tested. The deletion virus, VV K3L::gfp, generated for this study was evaluated in cell

PAGE 145

145 culture for a host range phenotype. VV K3L::gfp was grown and tittered on CV-1 cells and then evaluated for host range by titer on the cell lines from pigs, hamster, rabbits, humans, and monkeys shown in Table 6-1. The titer for VV K3L::gfp on CV-1 cells was 6.8 x 109 and exhibited no decrease in tit er on any of the cell lines ev aluated. Therefore, the VV K3L::gfp virus generated for these studies does not have a host range phe notype in vitro, which contradicts what has been previously reported. Survival Studies of Mice Infected with VV K3L::gfp All the mouse constructs were evaluated fo r survival and clini cal symptomology by IT inf ection with VV K3L::gfp. Based on previous literature it was expected that wild type and RNaseL single knockout mice would exhi bit a decreased sensitivity to VV K3L::gfp. Both the wild type and RNaseL single knockout animals ha ve wild type copies of the PKR gene, which K3L has been suggested to target. The PKR single knockout and DKO mic e were expected to exhibit sensitivity very similar to that of wt VV. With the de letion of the host target PKR and viral K3L proteins both missing fr om the infections there should be a canceli ng of effects of the proteins giving a wild type VV like infection. Wild type mice were infected with 105 to 107 pfu of VV K3L::gfp and exhibited survival similar to that observed with wt VV. (Figure 6-2A) 106 pfu was determined to be the LD80 for wild type mice, which was also the dose of wt VV required for 80% lethality in wild type mice. RNaseL single knockout mice evaluated fo r survival after infection with 104 to 106 pfu VV K3L::gfp. (Figure 6-2C) The LD80 was determined to be 106 pfu with mice demonstrating severe disease being euthanized by day 6. This was a decrease in virulence as compared to VV, where the LD80 of wt VV in RNaseL knockout mice was 105 pfu. (Chapter 4)

PAGE 146

146 PKR knockout mice were infected with 104 to 106 pfu VV K3L::gfp for evaluation of survival and clinical sympto mology. (Figure 6-2D) The LD80 was determined to be 106 pfu with mice demonstrating severe disease being euthanized by day 6. This also was a decrease in virulence as compared to VV, where the LD80 of wt VV in PKR knockout mice was 105 pfu. (Chapter 4). The DKO mice were infected with 103 to 106 pfu VV K3L::gfp for survival and clinical symptomology. DKO mice exhibited an increas ed survival when infected with VV K3L::gfp as compared to VV. (Figure 6-2B) The LD80 was determined to be 106 pfu with mice demonstrating severe disease eu thanized by day 6, where the LD80 of wt VV was 104 pfu. (Chapter 4) Clinical Symptoms in Mice Infected with VV K3L::g fp The wild type mice temperature and weight loss profiles for mice infected with 105 to 107 pfu of VV K3L::gfp. (Figure 6-3) Mice infected with 105 pfu exhibited no decrease in body temperature but did exhibit a weight loss maximu m of 17%. These animals were clearly ill, although not all of them required euthanization. Mice infected with 106 and 107 pfu of VV K3L::gfp all mice succumbed to disease by da y 6 exhibiting temperature decrease to 30C or lower and a 25% body weight decrease from beginning weight. The animals that succumbed to infection all exhibited a failure to groom, head and face swelling, hunched posture and respiratory distress. RNaseL single knockout mice temperature and we ight loss profiles for mice infected with 104 to 106 pfu VV K3L::gfp. (Figure 6-4) Mice infected with 104 pfu of VV K3L::gfp did not exhibit a temperature or weight lo ss indicating a lack of disease in these animals. Mice infected with 105 pfu did show a slight body temp erature decrease at day 7 and a normal weight loss of up

PAGE 147

147 to 20% as expected from animals that exhibit symptoms of disease. Mice infected with 105 pfu all required euthanizatio n by day 6 and exhibited severe disease with a body temperature drop to 25C or lower with a 28% average body weight lo ss by day 6. All the mice also exhibited the same clinical symptoms of failure to gr oom, head and face swelling, hunched posture and respiratory distress observed in the wild type and DKO mice. PKR single knockout mice were infected with 104 to 106 pfu VV K3L::gfp. The temperature and weight loss profiles for these mice. (Figure 6-5) Mice infected with 104 pfu of VV K3L::gfp did not exhibit a temperature decreas e, but did exhibit a we ight loss of up to 17%, suggesting some systemic disease at this virus dose. The animals infected with 105 pfu did exhibit a temperature and weight loss decrease suggesting severe systemic disease, however only 50% of the mice at this dose re quired euthanization for disease symptoms. Mice euthanized for severe disease infected with 106 pfu exhibited a body temperature decrease to 30C or lower and a 22% body weight loss as compared to starting weight by day 6. The PKR mice exhibited all the clinical symptoms that the ot her mouse constructs did at the 106 pfu dose, but also had severe eye secretions with secondary lesions present on the eyelids. This was the first time in the course of this study that secondary dermal lesions had been observed. DKO mice temperature and weight loss profiles for mice infected with 103 to 106 pfu. (Figure 6-6) Animals infected with 103 and 104 pfu showed modest temperature decreases and weight loss profiles up to 10% suggesting mild, transient disease. Mice infected with 105 pfu also showed little in the way of a temperature decrease, but did exhib it a maximum weight loss of 22%, indicating systemic dis ease. Mice infected with 106 pfu exhibited severe body temperature decreases as well as a weight loss up to 20% and clinical symptoms observed in the other mouse constructs that required euthanization.

PAGE 148

148 Virus Dissemination from the Site of Inoculation Virus spread as measured by tissue titer was perform ed on animals infected with VV K3L::gfp. This experiment was to establish the ability of the viru s to control the host immune system enough to allow spread and to exam ine the tissues the virus infects. Mice were infected with VV K3L::gfp via the IT route of infection. There were two sets of tissues taken, one done at the mouse construct specific VV LD80 dose previously presented in Chapter 4 for direct dose comparison. The second set was perf ormed at the experimentally established LD80 dose for VV K3L::gfp in all the mouse constructs for an outcome comparison. Titers for the lung, liver, spleen, gonad and br ain of mice infected w ith their respective VV LD80 dose. (Table 6-2) All tissu es were tittered on CV-1 cells. Wild type mice were infected with 106 pfu of VV K3L::gfp and sampled at days 2 to 6 post infection. The wild type mice showed high levels of virus present in the lungs, bu t little spread to the other distal organs. The most spread exhibited was to the brain, which had 3.68 logs at day 6, nearly 4 logs less than what was seen with wt VV (Table 3-2). RNaseL and PKR single knockout mice were infected with 105 pfu of VV K3L::gfp. Both mouse constructs showed high levels of virus in the lungs, with 7-8 logs at day 6. The PKR and RNaseL single knockout mice, from th e survival and weight loss data were clearly ill, exhibiting a 50% and 60% survival, respectively, and a 20% we ight loss, yet virus titers in the lungs were higher than that observed in VV. (Table 4-1) As with the wild type mice infected with VV K3L::gfp, there was little spread to the distal organs. The spread to the brain was the highest level found at 3 logs at day 6, 3 logs less than that of RNaseL and PKR single knockout mice infected with the same titer of wt VV.

PAGE 149

149 DKO mice were infected with 104 pfu VV K3L::gfp, a dose in which an 80% survival was observed with no signs of disease. Titers higher than those observed in wt VV infections were also observed in the lungs, at 8 logs for VV K3L::gfp. Spread from the lung to the liver, spleen, gonad and brains were very low as compar ed to wt VV at the same dose in DKO mice, typically 3 logs or more lower th an that of wt VV. (Table 4-1) All the mouse constructs also underwent tissu e sampling at days 3 and 5 post infection when infected with 106 pfu of VV K3L::gfp, the LD80 dose of VV K3L::gfp for all the mouse constructs. (Table 6-3) This was to compare the tissue titers when similar outcomes were observed, that is at a dose that varies, but al l the mice exhibit severe disease and require euthanization in the survival studies. PKR and RNas eL single knockout and DKO mice were infected with 106 pfu of VV K3L::gfp and harvested at days 3 and 5 post infection. Previous data for wild type mice infected with 106 pfu of VV K3L::gfp was used in Table 6-3. All mice had 8.0 to 8.8 logs of VV K3L::gfp in the lungs at both da y 3 and 5, again much higher than seen with VV in these mice. (Table 4-1) There wa s limited spread to distal sites in the mice with the most virus being found in the gonads and brains at day 5. Secondary lesions observed on PKR mice eyelids were harvested and tittered for confirmation of virus present as compared to eyelids harvested from a wild type VV infected PKR mouse. The eye lesions from the VV K3L::gfp infected mice had confirmed virus present, while the wild type VV infected eyelids did not. This data suggests that the K3L gene product co ntributes to the virus ability to spread from the lung. VV K3L::gfp is able to infect cells and re plicate efficiently in other tissues as evident by the increase in titers ove r time, it takes longer for the viru s replicate to high titers due to its delay in arrival at the distal sites.

PAGE 150

150 STAT-1 Staining of Lung Tissue STAT-1 immunostaining of lung tis sue from mi ce infected with VV K3L::gfp was performed. Lung tissue from wild t ype and DKO mice infected with 106 pfu of VV K3L::gfp at day 3 and 5 post infection were analyzed for the levels of STAT-1 protein. Lung tissue from wild type mice at day 3 post infection showed low levels of cytoplasmic staining with no infiltrating cells staining pos itive for STAT-1. The areas immediately surrounding both the edema and infected cells was slightly darker staining than the remainder of the lung tissue. (Figure 6-7C&D) Day 5 post infection staining was very similar to that observed at day 3 with low cytoplasmic staini ng throughout the lung with higher levels of nuclear staining around areas of edema and infecti on. The staining patterns observed in the wild type mouse lungs was different from th at seen with both wild type VV and VV E3L::gfp in that there was not an increase in STAT-1 protein present to suggest a strong immune response to viral infection. DKO mice at day 3 and 5 post infection (Figure 6-8E&F) showed moderate levels of both nuclear and cytoplasmic staini ng. The bronchi epithelial cells showed increase pattern of cytoplasmic and nuclear staining at both time points. Compared to the staining observed in uninfected animals, the staining obser ved with DKO mice infected with VV K3L::gfp was not considerably different. As with wild type mice infected with VV K3L::gfp, the DKO mice do not appear to respond strongly to VV K3L::gfp infection in the l ungs as had been previously observed. Histopathology Histopathology of the lung, liver, and spl een tissues from mi ce infected with VV K3L::gfp to determine the pathology of the viru s in animals. All mouse constructs were

PAGE 151

151 infected with 106 pfu VV K3L::gfp, to allow for comparison of disease with VV infected animals with identical outcomes of severe diseas e. Tissues were harvested from the animals, fixed in formalin overnight, and then embedded in paraffin. Each tissue was sectioned for IHC staining for gfp to determine virus replica tion in the tissue and H&E for overall tissue composition. Wild type mouse lungs at day 3 post infecti on demonstrated severe edema with necrosis of the bronchioles and loss of st ructure integrity. The infiltrate was mild, consisting of mainly neutrophils. The blood vessels loca ted in close proximity to the n ecrotic bronchioles exhibited a loss of integrity leading to the edema. (Fi gure 6-9B) Upon examination of the IHC focal infections located mainly at the bronchioles wa s observed. (Figure 6-9A) Positive staining for gfp was also detected in the alveolar epithelial cells The lung sections at day 5 were similar in both H&E findings and IHC, with more severe ed ema (Figure 6-9D & E) and larger areas of infected tissue around the bronchioles (Figure 6-9C & E). The liver sections showed no virus at day 3 or 5, but did have random sites of neutr ophillic necrosis on H&E examination. The spleen sections at both day 3 and 5 had a few random IHC positive cells, but normal H&E on day 3. Day 5 spleen H&E showed depletion in cells in the follicles, suggesting an active immune response in the host. (Figure 6-10) RNaseL mouse lungs at day 3 post infecti on showed severe edema with necrosis and neutrophilic cell infiltrate on H&E. The bronchioles showed a loss of structural integrity and a minor loss of blood vessel integrity leading to the edema. IHC staining showed the localized infection at the bronchiolar epithelial cells. Day 5 lung H&E showed severe edema and hemorrhage at both the area around the bronchiolar blood vessels and alveolar structures. Both vascular and bronchiolar structural integrity are lost and the pres ence of neutrophilic infiltrate.

PAGE 152

152 IHC showed bronchiolar epithelium and vascul ar endothelium staini ng, suggesting the edema was again due to virus infection damage to the va scular cell walls. Liver sections for both day 3 and 5 were normal for both H&E and IHC. Spleen sections at both day 3 and 5 showed no IHC staining, but severe lymphoid de pletion on H&E. The lymp hoid depletion most likely contributed to the infilt rates observed in the lung tissue. (Data not shown.) PKR mouse lungs had edema, hemorrhage, n ecrosis and neutrophilic infiltrate located around the IHC positive bronchioles at day 3. Day 5 lung sections had more severe edema around the bronchioles and in the alveolar spaces. An increase in necrosis and an expansion of lymphoid tissue with macrophages a nd neutrophilic infiltrate was seen on H&E. IHC for day 5 lungs showed positive bronchial epit helial cells and cellular debris around infected bronchioles. Liver sections were normal on both H&E and IHC for day 3 and 5. Spleen sections at day 3 were also completely normal. (Data not shown.) Day 5 spleen sections showed no IHC staining, but did show severe lymphoid tissue depletion. (Figure 6-10B) DKO mouse lungs at day 3 post infection showed mild edema with few infiltrating cells on H&E and a bronchiolar epithelial cell localized infection on IH C. (Figure 6-11B & C) On IHC the lung at day 3 showed bronchiolar epithe lial cells positive for gfp and virus infection in a localized, isolated pattern. (F igure 6-11A) Day 5 lungs showed severe edema both around the bronchioles and alveolar compartm ents with severe necrosis and neutrophilic infiltrate on H&E. (Figure 6-11E & F) There were also lymphocytic cell aggregates observed that are most likely from lymphoid cell proliferation and cell inf iltrate. Day 5 also had positive bronchiolar epithelial cells, but also a high level of endothelial cell stai ning. (Figure 6-11D) This suggests that the edema seen is due to the loss of blood ve ssel integrity from virus infection and necrosis. The liver sections were negative on IHC for gf p and virus replication, but there were random

PAGE 153

153 sites of neutrophilic necrosis at day 3 on H&E. The day 3 spleen sections were normal on both H&E and IHC, and day 5 showed no IHC staining but a moderate lymphoid depletion, most likely a contributor to the lymphocytic cell aggregat es seen in the lungs at day 5. (Figure 6-10A) Weights of the tissues taken for histop athology are shown in Figure 6-12. The lung weights as compared to PBS animals over time is shown in Figure 6-12A. The weight of the lung tissue increases over time as would be expect ed from the amount of edema present late in infection with no differences between mouse constr ucts seen. Spleen weights as compared to PBS spleen weights over time are shown in Figur e 6-12B. The weights are normal at day 3, but decreased at day 5 coinciding with the lymphoid depletion seen at day 5 by histopathology. The body weight to lung weight ratios is shown in Fi gure 6-12C. The ratios decrease over time for all the mouse constructs indicating severe disease. The lung wei ght to spleen weight ratios, shown in Figure 6-12D, exhibited a general increase in ratio with do differences between mouse constructs. Discussion The major role of K3L in the pathogenesis of the IT model for VV infection of m ice is not PKR. This is deduced from the mouse survival data in which there are no differences between the wild type and knockout mice in survival or approximate LD80 dose. This is possibly explained by the presence of E3L in VV K3L::gfp and E3L having been previously described as both a dsRNA binding protein and responsible for sequestering PKR from activation by the presence of dsRNA. (Kim, Lowenhaupt et al., 20 04) In the presence of E3L the K3L protein may be a backup or redundant protein that is responsible for aidi ng E3L in blocking the activation of PKR during periods of time in whic h E3L is unable to control the immune response completely. This is also supported by the lack of increase in STAT-1 immunostaining of lung

PAGE 154

154 tissue late in infection, suggesting that K3L is no t needed to control the immune responses in the lung tissue. Using a different infection model a more dramatic survival phenotype may be observed based upon this hypothesis. The deletion of K3L from VV doe s not affect the virus abil ity to grow as shown by the increase in titers observed in all organs, but does affect its ability to spread. The areas of virus infection and growth are much larger in the animals infected with VV K3L::gfp (Table 6-2 and Table 6-3) than those observed with wild type VV (Table 4-1), leading to more tissue damage within the lung. The infection of the endothelial cells surround th e blood vessels causing leakage from the blood vessels is most likely the source of the massive edema observed in the tissues and contributes to the virus inability to spread. This inability to spread is most likely due to the loss of integrity of the blood vessels. Poxviruses have been theorized to spread by a cell associated mechanism through the blood in animals and with decreased blood vessel integrity this would not occur as rapidly. K3L may contribute to virus growth in the animal by controlling or sequestering some undetermined protein or prot eins to keep the virus from overgrowing the primary site of infection in or der to allow for dissemination th rough the animal via the blood. A time course for the levels of virus in blood would have to be performed in order to determine if this was the cause for the lack of virus spread. Another possibility is that in the absence of K3L this pattern of spread is not altered, but the host immune is mounting a systemic immune response that is controlling the virus growth throughout the body except in sites that are immune privileged. In this case K3L would function as an immune suppressor, but again this target has yet to be discovered as illustrated by the survival curves with VV K3L::gfp. There is an immune response occurring systemically as illustrated by the severe lymphoid depletion from the spleen. This suggests a high level of

PAGE 155

155 circulating immune cells. The higher titers of VV K3L::gfp found in the brain and gonads as well as the appearance of secondary eye lesions and profuse eye discharge suggest a preference for immune privileged sites. Again, virus le vels in whole blood as well as blood cell counts would provide valuable insight into this as a explanation for the diffe rential disease patterns observed with VV K3L::gfp.

PAGE 156

156 Figure 6-1. K3L protein do main diagram. K3L is 88 amino aci ds in length and comprised of 5 beta sheets and 2 alpha helices. The sh aded red area from amino acids 15 to 66 shows the area deleted and replaced with gfp in VV K3L::gfp. (Dar & Sicheri, 2002)

PAGE 157

157 Table 6-1. In vitro titer data VV K3L::gfp host range Cell Line Virus Titer BHK-21 3.2 x 109 CV-1 6.8 x 109 RK13 4.2 x 109 PK15 5.2 x 109 Vero 1.22 x 109 BSC40 8 x 109 A549 6.6 x 109 MCF7 5.4 x 109

PAGE 158

158 Figure 6-2. Survival data for mice infected with VV K3L::gfp. The numbers of mice used for each virus dose are located in the respectiv e Figure legends. The overall significance for all the curves was p < 0.001 by the Kaplan Meier survival analysis. A) Survival curve for wild type mice infected with VV K3L::gfp. The Holm-Sidak method for multiple comparisons was used to determine significance between the curves. All comparisons were found to be significant at p 0.004 except the 106 pfu curve compared to the 107 pfu curve. B) Survival curve for DKO mice infected with VV K3L::gfp. The Holm-Sidak method for multiple comparisons determined that the 106 pfu curve was significantly different (p 0.0001) compared to the other doses. C) Survival curve for PKR mice infected with VV K3L::gfp. The Holm-Sidak method for multiple comparisons determined that the 106 pfu curve was significantly different (p 0.001) compared to the other doses. D) Survival curve for RNaseL mice infected with VV K3L::gfp. The Holm-Sidak method for multiple comparisons determined that the 106 pfu curve was significantly different (p 0.0001) compared to the other doses.

PAGE 159

159 Figure 6-3. Average body temperature and wei ght loss of wild type mice infected with VV K3L::gfp.

PAGE 160

160 Figure 6-4. Average body temperature and we ight loss of RNaseL mice infected with VV K3L::gfp.

PAGE 161

161 Figure 6-5. Average body temperature and weight loss of PKR mice infected with VV K3L::gfp.

PAGE 162

162 Figure 6-6. Average body temperature and weight loss of DKO mice infected with VV K3L::gfp.

PAGE 163

163 Table 6-2. Titer of tissues from all mouse constructs infected with VV LD80 doses of VV K3L::gfp. Mouse Construct Tissue Day 2 3 4 5 6 Wild type Lung 7.83 8.27 8.66 8.55 8.34 DKO Lung 4.98 6.99 6.96 7.46 7.09 RNaseL Lung 7.04 7.60 8.07 8.22 7.75 PKR Lung 6.87 7.49 8.12 8.28 8.55 Wild type Liver 0.92 0.64 0.90 1.82 0.89 DKO Liver 1.26 1.00 1.08 1.05 2.29 RNaseL Liver 1.26 0.91 0.37 0.3 0.72 PKR Liver 1.20 0.95 1.05 1.67 0.88 Wild type Spleen 0.79 0.95 1.69 1.46 0.80 DKO Spleen 1.30 0.97 1.17 0.83 2.35 RNaseL Spleen 1.15 0.79 0.89 0.3 0.67 PKR Spleen 1.15 1.28 1.58 1.30 1.78 Wild type Gonad 1.01 0.94 0.83 3.76 3.06 DKO Gonad 1.30 1.20 1.09 1.25 1.08 RNaseL Gonad 1.26 1.21 1.01 1.01 0.67 PKR Gonad 1.26 1.22 1.90 3.01 2.59 Wild type Brain 1.32 2.56 3.14 3.52 3.68 DKO Brain 1.26 1.10 1.37 2.19 2.71 RNaseL Brain 1.08 1.62 2.38 2.30 2.07 PKR Brain 1.05 1.69 2.41 3.37 3.59 Average log10 values for tissue titer data for mice infected with VV K3L::gfp. Mice were infected with their construct specific VV LD80 dose and tissues harvested at day 2 to 6. n 8 mice per mouse construct per time point. Lim it of detection for th e titer assay was 0.3.

PAGE 164

164 Table 6-3. Titer of tissues for all mouse constructs infected with VV K3L::gfp 106 pfu Mouse Construct Tissue Day 3 5 Wild type Lung 8.27 8.55 DKO Lung 8.01 8.72 RNaseL Lung 8.60 8.47 PKR Lung 8.45 8.78 Wild type Liver 0.64 1.82 DKO Liver 0.11 2.61 RNaseL Liver 1.62 1.76 PKR Liver 1.18 2.09 Wild type Spleen 0.95 1.46 DKO Spleen 0.96 2.16 RNaseL Spleen 1.58 1.44 PKR Spleen 1.79 2.58 Wild type Gonad 0.94 3.76 DKO Gonad 0.33 3.03 RNaseL Gonad 1.19 2.03 PKR Gonad 0.63 2.73 Wild type Brain 2.56 3.52 DKO Brain 2.42 3.84 RNaseL Brain 3.17 3.16 PKR Brain 2.38 3.44

PAGE 165

165 Figure 6-7. Staining of STAT-1 from VV K3L::gfp infected wild type mice. Stars show the location of bronchi. A) Uninfected mice at 20X magnification. B) Uninfected mice at 40X magnification. C) Day 3 post inf ection lung at 20X magnification. D) Day 3 post infection lung at 40X magnification. E) Day 5 post infection at 20X magnification. F) Day 5 post in fection at 40X magnification.

PAGE 166

166 Figure 6-8. Staining of STAT-1 from VV K3L::gfp infected DKO mice. Stars show the location of bronchi. A) Uninfected mice at 20X magnification. B) Uninfected mice at 40X magnification. C) Day 3 post inf ection lung at 20X magnification. D) Day 3 post infection lung at 40X magnification. E) Day 5 post infection at 20X magnification. F) Day 5 post in fection at 40X magnification.

PAGE 167

167 Figure 6-9. Histopathology of VV K3L::gfp infected wild type mouse lung tissue. B shows the location of bronchi; BV the location of blood vessels; E areas of edema. A) Day 3 IHC stained lung section at 5X magnification. The arrow points to areas of infected lung tissue. B) Day 3 H&E lung section at 5X magnification. C) Day 5 IHC stained lung section at 5X magnification. D) Day 5 H&E lung sec tion at 5X magnification. E) Day 5 IHC lung section at 20X magnificat ion. F) Day 5 H&E lung section at 40X magnification. N shows the location of neutrophilic infiltrating cells.

PAGE 168

168 Figure 6-10. Histopathology spleen tissue from mice infected with VV K3L::gfp. A) Spleen H&E from a DKO mouse 5 days post infect ion. B) Spleen H&E from a PKR knockout mouse 5 days post infection.

PAGE 169

169 Figure 6-11. Histopathology of VV K3L::gfp infected DKO mouse lung tissue. B shows the location of bronchi; BV the location of blood vessels; E areas of edema; F areas of fibrin collection A) IHC of infected lung tissue from day 3 at 5X magnification. Arrow shows the location of infected lung tissue. B) H&E at day 3 using 5x magnification. C) H&E at day 3 at 20x magnification. D) IHC at day 5 at 5x magnification. E) H&E at day 5 at 10x magnification. F) H&E at day 5 at 20X magnification.

PAGE 170

170 Figure 6-12. Tissue weights for mice infected with VV K3L::gfp. A) Average lung weights over time of all mouse constructs infected with 106 pfu of VV K3L::gfp. B) Average lung weights over time of all mous e constructs infected with 106 pfu of VV K3L::gfp. C) Body weight to lung wei ght ratios for all m ouse constructs over time. D) Lung weight to spleen weight ratios for all mouse c onstructs over time.

PAGE 171

171 CHAPTER 7 CONCLUSIONS AND DISCUSSION Overall Conclusions VV infection of wild type mice was shown to produce severe disease ch aracterized by a rapid weight loss that occurs simultaneously with a body temperature decrease. The primary site of infection in this infection is the lung, of which the virus infected cells are focused around the bronchi. This localized infection caused edem a around the infected bronc hi and the influx of immune cells to the lung tissue. The influx of immune cells suggests a wide spread immune response to infection, however the respons e was not caused by an increase in IFN as measured by whole lung ELISA. Vaccinia virus is known to encode multiple proteins to control both the production and downstream signaling of IFN, and this most likely contributes to the lack of IFN up regulation. Although the lung in fection was maintained as a lo calized infection, global tissue responses were observed in both transcription le vels as measured by microarrays and protein levels as measured by STAT-1 immunohistochemistry. The importance of both RNaseL and PKR in controlling the VV infection in mice has been shown from the increased sensit ivity of the knockout mice to VV. The question as to whether one gene is more important in controlling the infe ction than the other has also been answered in this study. It has been shown that both genes co ntribute to the sensitiv ity of the DKO mice and that the deletion of both RNaseL and PKR shows an additive eff ect, suggesting that both genes are important in controlling wild type VV infecti ons. This was also supported by the ability of the virus to replicate to simila r levels in all organs sampled (with exception of the liver and spleen in the DKO mice) regardle ss of the input inoculums titer.

PAGE 172

172 The virus lacking E3L was found to be attenua ted in both tissue culture and animals as previously reported, validating the findings that E3L is necessary for pathogenesis in wild type animals. It is clear from the DKO mouse survival and hi stological data that the viral gene E3L is important in controlling the hos t response to viral infection and not only targets PKR and RNaseL but other pathways that have yet to be el ucidated. E3L may also influence the spread of the virus to other sites as observed with the decreased titers of virus found in the distal organs with DKO mice infected with high doses of VV E3L::gfp. The data for K3L suggests that the major func tion of K3L in an animal is for virus dissemination, although it may play a limited role in controlling eIF2 phosphorylation. The survival data in which there are no differences between the wild type kn ockout mice in survival or doses needed to reach a LD80 do suggest that K3L is not require d to cause clinical disease. The deletion of K3L from VV does affect the virus ability to sp read, most likely due to the loss of integrity of the blood vessels, as the virus sp reads by a cell associated mechanism through the blood and with severe blood vessel leakage this becomes very diffi cult. Finally, K3L does seem to control host cell tropism and without K3L the vi rus tropism is altered; the virus can now infect the blood vessel endothe lial cells and moves towards a derm al infection. Although all the animals that succumbed to disease clinically l ooked the same the disease was very different upon viremia and histology. Final Thoughts The major lesson to be learned from this seri es of studies is that tissue culture does not predict what effects will be obser ved in an anim al. Also it has been shown that although animals may have the same clinical symptoms (weight loss, temperature decrease, and respiratory

PAGE 173

173 distress) the influences of the virus in the animal s can be very different. These lessons are shown in both infections of animals with VV E3L::gfp and VV K3L::gfp. E3L was demonstrated to be highly attenuat ed, but when inoculation of an immune compromised animal (DKO) at high titers was performed the animals presented with wild type VV symptomology and disease, contrary to other published works. Although the virus dissemination was not as high as with wild type VV and the primary responses in the lung were far different from that seen in wild type VV the animals still succumbed to a systemic disease. K3L in tissue culture may very well associate with PKR and prevent eIF2 phosphorylation, but its major role in the pathogenes is of an animal is virus spread. VV virus lacking K3L is unable to effectively spread thr ough an animal yet lethally infected animals succumb to systemic disease undistinguishable from that of a wild type VV infection.

PAGE 174

174 LIST OF REFERENCES Adams, M. M., Rice, A. D., Moyer, R. W., 2007. Rabbitpox virus and Vacci nia virus infection of rabbits as a model for human sma llpox. The Journal of Virology 81, 11084-11095. Alcami, A., Smith, G. L., 1995. Vaccinia, cowpo x, and camelpox viruses encode soluble gamma interferon receptors with novel broad species specificity. Journal of Virology 69, 4633-4639. Bahar, M. W., Kenyon, J. C., Putz, M. M., Abrescia, N. G., Pease, J. E., Wise, E. L., Stuart, D. I., Smith, G. L., Grimes, J. M., 2008. Structure a nd function of A41, a Vaccinia virus chemokine binding protein. PLoS.Pathog. 4, 0001-0014. Beattie, E., Denzler, K. L., Tartaglia, J., Perkus M. E., Paoletti, E., Jacobs, B. L., 1995. Reversal of the interferon-sensitive phenotype of a v accinia virus lacking E3L by expression of the reovirus S4 gene. Journal of Virology 69, 499-505. Beattie, E., Kauffman, E. B., Martinez, H., Perkus, M. E., Jacobs, B. L., Paoletti, E., Tartaglia, J., 1996. Host-range restriction of vaccinia virus E3L-specific deletion mutants. Virus Genes 12, 89-94. Beattie, E., Paoletti, E., Tartaglia, J., 1995. Distinct patterns of IFN sensitivity observed in cells infected with vaccinia K3Land E3Lmutant viruses. Virology 210, 254-263. Beattie, E., Tartaglia, J., Paoletti, E., 1991. Vaccinia virus-encoded eIF-2 alpha homolog abrogates the antiviral effect of interferon. Virology 183, 419-422. Beaud, G., 1995. Vaccinia virus DNA replicati on: a short review. Biochimie 77, 774-779. Borden, E. C., Sen, G. C., Uze, G., Silverman, R. H., Ransohoff, R. M., Foster, G. R., Stark, G. R., 2008. Interferons at age 50: past, current and future impact on biomedicine. Nature Reviews Drug Discovery 6, 975-990. Boyd, A., Fazakerley, J. K., Bridgen, A., 2006. Pat hogenesis of Dugbe virus infection in wildtype and interferon-deficient mice. J Gen Virol 87, 2005-2008. Brandt, T., Heck, M. C., Vijaysri, S., Jentarra, G. M., Cameron, J. M., Jacobs, B. L., 2005. The N-terminal domain of the vaccinia virus E3L-pr otein is required for neurovirulence, but not induction of a protective imm une response. Virology 333, 263-270. Brandt, T. A., Jacobs, B. L., 2001. Both carboxyand amino-terminal domains of the vaccinia virus interferon resistance gene, E3L, are require d for pathogenesis in a mouse model. Journal of Virology 75, 850-856. Broyles, S. S., 2003. Vaccinia virus transcrip tion. Journal of General Virology 84, 2293-2303.

PAGE 175

175 Brum, L. M., Lopez, M. C., Varela, J. C., Baker, H. V., Moyer, R. W., 2003. Microarray analysis of A549 cells infected with rabbitpox virus (RPV): a compar ison of wild-type RPV and RPV deleted for the host range gene, SPI-1. Virology 315, 322-334. Carroll, K., Elroy Stein, O., Moss, B., Jagus, R., 1993. Recombinant vaccinia virus K3L gene product prevents activation of double-stranded RNA-dependent, initiation factor 2 alphaspecific protein kinase. Journal of Biological Chemistry 268, 12837-12842. Chang, H. W., Jacobs, B. L., 1993. Identification of a conserved motif that is necessary for binding of the vaccinia virus E3L gene produc ts to doublestranded RNA. Virology 194, 537547. Chang, H. W., Uribe, L. H., Jacobs, B. L., 1995. Rescue of vaccinia virus lacking the E3L gene by mutants of E3L. Journal of Virology 69, 6605-6608. Colamonici, O. R., Domanski, P., Sweitzer, S. M., Larner, A., Buller, R. M., 1995. Vaccinia virus B18R gene encodes a type I interferonbinding protein that blocks interferon alpha transmembrane signaling. Journal of Biological Chemistry 270, 15974-15978. Colby, C., Duesberg, P. H., 1969. Double-stranded R NA in vaccinia virus infected cells. Nature 222, 940-944. Condit, R. C., 2007. Vaccinia, Inc.Probing the f unctional substructure of poxviral replication factories. Cell Host and Microbe 2, 205-207. Condit, R. C., Moussatche, N., Traktman, P., 2006. In a nutshell: structure and assembly of the vaccinia virion. Adv.Virus Res. 66, 31-124. Dar, A. C., Sicheri, F., 2002. X-ray crystal struct ure and functional analysis of vaccinia virus K3L reveals molecular determinants for PKR subversion and substrate recognition. Molecular Cell 10, 295-305. Davies, M. V., Chang, H. W., Jacobs, B. L., Kaufman, R. J., 1993. The E3L and K3L vaccinia virus gene products stimulate translation th rough inhibition of the double-stranded RNAdependent protein kinase by different mechanisms. Journal of Virology 67, 1688-1692. Davies, M. V., Elroy Stein, O., Jagus, R., Mo ss, B., Kaufman, R. J., 1992. The vaccinia virus K3L gene product potentiates translation by i nhibiting double-stranded-R NA-activated protein kinase and phosphorylation of the alpha subunit of eukaryotic initiation factor 2. Journal of Virology 66, 1943-1950. Domingo-Gil, E., Perez-Jimenez, E., Ventoso, I ., Najera, J. L., Esteban, M., 2008. Expression of the E3L gene of Vaccinia virus in transgenic mice decreases host resistance to Vaccinia virus and Leishmania major infections. The Journal of Virology 82, 254-267. Duesberg, P. H., Colby, C., 1969. On the biosynth esis and structure of double-stranded RNA in vaccinia virus-infected cells. Pr oceedings of the National Academ y of Sciences of the United States of America 64, 396-403.

PAGE 176

176 Fenner, F., 2000. Adventures with poxviruses of vertebrates. Fems Microbiology Reviews 24, 123-133. Fenner, F., 1990. Wallace P. Rowe lecture. Poxvir uses of laboratory animals. Lab.Anim.Sci. 40, 469-480. Gale, M., Jr., Katze, M. G., 1998. Molecular mech anisms of interferon re sistance mediated by viral-directed inhibition of PK R, the interferon-induced protein kinase. Pharmacol.Ther. 78, 2646. Garcia, M. A., Meurs, E. F., Esteban, M., 2007. The dsRNA protein kinase PKR: Virus and cell control. Biochimie. Grinde, B., Gayorfar, M., Hoddevik, G., 2007. Modulation of gene expression in a human cell line caused by poliovirus, vaccinia virus and interferon. Virol.J. 4, 24. Guerra, S., Lopez-Fernandez, L. A., Conde, R., Pascual-Montano, A., Harshman, K., Esteban, M., 2004. Microarray Analysis Reveals Characteristic Changes of Host Cell Gene Expression in Response to Attenuated Modified Vaccinia Viru s Ankara Infection of Human HeLa Cells. Journal of Virology 78, 5820-5834. Guerra, S., Najera, J. L., Gonzalez, J. M., Lope z, L., Climent, N., Gatell, J. M., Gallart, T., Esteban, M., 2007. Distinct gene expression prof iling after infection of immature human monocyte-derived dendritic cells by the attenua ted poxvirus vectors MVA and NYVAC. Journal of Virology. Guidotti, L. G., Morris, A., Mendez, H., Koch, R., Silverman, R. H., Williams, B. R. G., Chisari, F. V., 2002. Intertlueron-regulated pathways that control hepa titis B virus replication in transgenic mice. Journal of Virology 76, 2617-2621. Haga, I. R., Bowie, A. G., 2005. Evasion of inna te immunity by vaccinia virus. Parasitology 130, S11-S25. Haller, O., Kochs, G., Weber, F., 2007. Interfer on, Mx, and viral countermeasures. Cytokine & Growth Factor Reviews 18, 425-433. Harte, M. T., Haga, I. R., Maloney, G., Gray, P., Reading, P. C., Bartlett, N. W., Smith, G. L., Bowie, A., O'Neill, L. A., 2003. The poxvirus prot ein A52R targets toll -like receptor signaling complexes to suppress host defense. J Exp.Med. 197, 343-351. Ho, C. K., Shuman, S., 1996. Mutational analysis of the vaccinia virus E3 protein defines amino acid residues involved in E3 binding to doubl e-stranded RNA. Journal of Virology 70, 26112614. International Committee on Taxonomy of Vi ruses, 2002. 8th ICTV Report. ICTV. Isaacs, A., Lindenmann, J., 1957. Virus interferen ce. I. The interferon. Proc.R.Soc.Lond.[Biol] 147, 258-267.

PAGE 177

177 Jewell, N. A., Vaghefi, N., Mertz, S. E., Akter, P ., Peebles, R. S., Jr., Bakaletz, L. O., Durbin, R. K., Flano, E., Durbin, J. E., 2007. Differential type I interferon induction by respiratory syncytial virus and influenza a virus in vivo. J Virol. 81, 9790-9800. Kakuta S., Shibata, S., Iwakura, Y., 2002. Genomic structure of the mouse 2',5'-oligoadenylate synthetase gene family. J In terferon Cytokine Res 22, 981-993. Katze, M. G., He, Y., Gale, M., 2002. Viru ses and interferon: A fight for supremacy. Nat.Rev.Immunol 2, 675-687. Kibler, K. V., Shors, T., Perkins, K. B., Ze man, C. C., Banaszak, M. P., Biesterfeldt, J., Langland, J. O., Jacobs, B. L., 1997. Double-strande d RNA is a trigger for apoptosis in vaccinia virusinfected cells. Journal of Virology 71, 1992-2003. Kim, Y. G., Lowenhaupt, K., Oh, D. B., Kim, K. K., Rich, A., 2004. Evidence that vaccinia virulence factor E3L binds to Z-DNA in vivo: Im plications for development of a therapy for poxvirus infection. Proc.Natl.Acad.Sci.U.S.A 101, 1514-1518. Laassri, M., Meseda, C. A., Williams, O., Me rchlinsky, M., Weir, J. P., Chumakov, K., 2007. Microarray assay for evaluation of the genetic st ability of modified vaccinia virus Ankara B5R gene. J.Med.Virol. 79, 791-802. Langland, J. O., Jacobs, B. L., 2002. The role of the PKR-inhibitory genes, E3L and K3L, in determining vaccinia virus host range. Virology 299, 133-141. Ludwig, H., Mages, J., Staib, C., Lehmann, M. H. Lang, R., Sutter, G., 2005. Role of viral factor E3L in modified Vaccinia virus Ankara infecti on of human HeLa cells: Regulation of the virus life cycle and identification of differentially expressed host ge nes. Journal of Virology 79, 25842596. Maher, S. G., Romero-Weaver, A. L., Scarzello, A. J., Gamero, A. M., 2007. Interferon: Cellular executioner or white knight? Curre nt Medicinal Chemistry 14, 1279-1289. Martinez, M. J., Bray, M. P., Huggins, J. W., 2000. A mouse model of aerosol-transmitted orthopoxviral disease Morphology of experiment al aerosol-transmitted orthopoxviral disease in a cowpox virus-BALB/c mouse system. Archiv es Of Pathology & Laboratory Medicine 124, 362-377. McShane, L. M., Shih, J. H., Michalowska, A. M., 2003. Stastistical issu es in the design and analysis of gene expression mi croarray studies of animal models. Journal of Mammary Gland Biology and Neoplasia 8, 359-374. Moss, B., 2001. Poxviridae: The viruses and their replication. In: D. E. Knipe, P. M. Howley (Eds.), Fields Virology. Lippincott Willia ms & Wilkins, Philadelphia, pp. 2849-2884. Moss, B., 2006. Poxvirus entry and membrane fusion. Virology 344, 48-54.

PAGE 178

178 Moss, B., 1996. Poxviridae : The viruses and their replication. In: B. N. Fields, D. M. Knipe, P. M. Howley (Eds.), Fields Virology. Lippincott-Reven, Philadelphia, pp. 2637-2672. Moss, B., 1990. Regulation of vaccinia virus transcription. Annu.Rev.Biochem. 59, 661-688. Mossman, K., Upton, C., Buller, R. M., McFadde n, G., 1995. Species specificity of ectromelia virus and vaccinia virus interferon-gamma binding pr oteins. Virology 208, 762-769. Payne, L. G., 1980. Significance of extracellular enveloped virus in the in vitro and in vivo dissemination of vaccinia. Journal of General Virology 50, 89-100. Randall, R. E., Goodbourn, S., 2008. Interferons and viruses: an interplay between induction, signalling, antiviral responses a nd virus countermeasures. Journal of General Virology 89, 1-47. Reading, P. C., Smith, G. L., 2003. A kinetic analys is of immune mediators in the lungs of mice infected with vaccinia virus and comparison wi th intradermal infection. Journal of General Virology 84, 1973-1983. Samuel, C. E., 2001. Antiviral actions of inte rferons. Clinical Microbiology Reviews 14, 778809. Schramm, B., Locker, J. K., 2005. Cytoplasmi c organization of POXvirus DNA replication. Traffic 6, 839-846. Seet, B. T., Johnston, J. B., Brunetti, C. R., Barre tt, J. W., Everett, H., Cameron, C., Sypula, J., Nazarian, S. H., Lucas, A., McFadden, G., 2003. Poxviruses and immune evasion. Annual Review Of Immunology 21, 377-423. Seet, B. T., Singh, R., Paavola, C., Lau, E. K ., Handel, T. M., McFadden, G., 2001. Molecular determinants for CC-chemokine recogniti on by a poxvirus CCchemokine inhibitor. Proc.Natl.Acad.Sci.U.S.A 98, 9008-9013. Shors, S. T., Beattie, E., Paolet ti, E., Tartaglia, J., Jacobs, B. L., 1998. Role of the Vaccinia virus E3L and K3L gene products in resuce of VSV and EMCV from the effects of IFN-alpha. J Interferon Cytokine Res 18, 721-729. Shors, T., Jacobs, B. L., 1997. Complementation of deletion of the vaccinia virus E3L gene by the Escherichia coli RNase III gene. Virology 227, 77-87. Silverman, R. H., 2007. Viral Encounters with OAS and RNase L during the IFN antiviral response. Journal of Virology 81, 12973-12978. Silverman, R. H., SenGupta, D. N., 1990. Transl ational regulation by HIV leader RNA, TAT, and interferoninducible enzymes. J.Exp.Pathol. 5, 69-77. Smith, G. L., 1999. Vaccinia virus imm une evasion. Immunology Letters 65, 55-62.

PAGE 179

179 Smith, S. A., Kotwal, G. J., 2002. Immune response to poxvirus infections in various animals. Critical Reviews in Microbiology 28, 149-185. Stern, R. J., Thompson, J. P., Moyer, R. W., 1997. Attenuation of B5R mutants of rabbitpox virus in vivo is related to im paired growth and not an enhanced host inflammatory response. Virology 233, 118-129. Stewart, M. J., Blum, M. A., Sherry, B., 2003. PKR's protective role in viral myocarditis. Virology 314. Stojdl, D. F., Abraham, N., Knowles, S., Marius R., Brasey, A., Lichty, B. D., Brown, E. G., Sonenberg, N., Bell, J. C., 2000. The murine doubl e-stranded RNA-dependent protein kinase PKR is required for resistance to vesicular stomatitis virus. Journal of Virology 74, 9580-9585. Streitenfeld, H., Boyd, A., Fazakerley, J. K., Bridgen, A., Elliott, R. M., Weber, F., 2003. Activation of PKR by Bunyamwera virus is independe nt of the viral interferon antagonist NSs. Journal of Virology 77, 5507-5511. Symons, J. A., Alcami, A., Smith, G. L., 1995. Vacci nia virus encodes a soluble type I interferon receptor of novel structure and broa d species specificity. Cell 81, 551-560. Takoaka, A., Yanai, H., 2006. Interferon signal ling network in innate defence. Cellular Microbiology 8, 907-922. Thompson, J. P., Turner, P. C., Ali, A. N., Cren shaw, B. C., Moyer, R. W., 1993. The effects of serpin gene mutations on the distinctive pa thobiology of cowpox and rabbitpox virus following intranasal inoculation of Ba lb/c mice. Virology 197, 328-338. Tscharke, D. C., Smith, G. L., 1999. A model fo r vaccinia virus pathogenesis and immunity based on intradermal injection of mouse ear pinnae. Journal of General Virology 80, 2751-2755. Turner, G. S., 1967. Respiratory infection of mice with vaccinia virus. Journal of General Virology 1, 399-402. Upton, C., Slack, S., Hunter, A. L., Ehlers, A., Roper, R. L., 2003. Poxvirus orthologous clusters: toward defining the minimum essent ial poxvirus genome. J Virol 77, 7590-7600. Vijaya, S., Jentarra, G., Heck, M. C., Mercer, A. A., Mc Innes, C. J., Jacobs, B. L., 2008. Vaccinia viruses with mutations in the E3L gene as potential replication-competent, attenuated vaccines: Intra-nasal vaccination. Vaccine 26, 664-676. Vijaysri, S., Jentarra, G., Heck, M. C., Mercer, A. A., McInnes, C. J., Jacobs, B. L., 2008. Vaccinia viruses with mutations in the E3L gene as potential replication-competent, attenuated vaccinies: Intra-nasal vaccination. Vaccine 26, 664-676. Vijaysri, S., Talasela, L., Mercer, A. A., McInnes, C. J., Jacobs, B. L., Langland, J. O., 2003. The Orf virus E3L homologue is able to complement deletion of the vaccinia virus E3L gene in vitro but not in vivo. Virology 314, 305-314.

PAGE 180

180 Wilcock, D., Duncan, S. A., Traktman, P., Zha ng, W. H., Smith, G. L., 1999. The vaccinia virus A4OR gene product is a nonstructu ral, type II membrane glycoprot ein that is expressed at the cell surface. Journal of General Virology 80, 2137-2148. Wolffe, E. J., Isaacs, S. N., Moss, B., 1993. Deletion of the vaccinia virus B5R gene encoding a 42-kilodalton membrane glycoprotein inhibits extracellular virus envelope formation and dissemination. Journal of Virology 67, 4732-4741. Xiang, Y., Condit, R. C., Vijaysri, S., Jacobs, B., Williams, B. R., Silverman, R. H., 2002a. Blockade of interferon induction and action by the E3L double-stranded RNA binding proteins of vaccinia virus. J Virol 76, 5251-5259. Xiang, Y., Condit, R. C., Vijaysri, S., Jacobs, B., Williams, B. R., Silverman, R. H., 2002b. Blockade of interferon induction and action by the E3L double-stranded RNA binding proteins of vaccinia virus. Journal of Virology 76, 5251-5259. Xu, R., Johnson, A. J., Liggitt, D., Bevan, M. J., 2004. Cellular and humoral immunity against vaccinia virus infection of mi ce. Journal Of Immunology 172, 6265-6271. Yang, Y., Reis, L. F. L., Pavlovic, J., Aguzzi, A., Shafer, R., Kumar, A., Williams, B. R. G., Aguet, M., Weissmann, C., 1995. Deficient sign aling in mice deviod of double-stranded RNAdependent protein kinase EMBO Journal 14, 6095-6106. Zaucha, G. M., Jahrling, P. B., Geisbert, T. W., Swearengen, J. R., Hensley, L., 2001. The pathology of experimental aerosolized monke ypox virus infection in cynomolgus monkeys (Macaca fascicularis). Lab Invest 81, 1581-1600. Zhou, A., Paranjape, J. M., Der, S. D., Williams B. R., Silverman, R. H., 1999. Interferon action in triply deficient mice reveals the existence of alternative antiviral pathways. Virology 258, 435-440. Zhou, A. M., Paranjape, J., Brown, T. L., Nie, H. Q., Naik, S., Dong, B. H., Chang, A. S., Trapp, B., Fairchild, R., Colmenares, C., Silverman, R. H., 1997. Interferon ac tion and apoptosis are defective in mice devoid of 2' ,5'-oligoadenylate-dependent RNase L. EMBO J. 16, 6355-6363.

PAGE 181

BIOGRAPHICAL SKETCH Amanda Rice was born in Macon, Georgia, on October 19, 1979. She grew up with her parents, CW4 (U.S. Ar my Retired) and Mrs. Davi d Rice, and her two siblings, Anna and Ian. She attended Western Kentucky University ma joring in both chemistry and recombinant genetics, earning her Bachelor of Science in 2002, Cum Laude. In the fall of 2002 she entered the Ph.D. program in the Interdisciplinary Program in Biomedical Sciences at the University of Florida, joining the laborator y of Dr. Richard Moyer in May 2003. She currently lives in Gainesville, Florida, with her husband, Brian, and daughter, Jasmine.