Elucidation of the Tissue Tropism Determinants of Jaagsiekte Sheep Retrovirus and Enzootic Nasal Tumor Virus
by María Carla Rosales Gerpe
A Thesis Presented to The University of Guelph
In partial fulfillment of the requirements for the degree of Doctor in Philosophy in Pathobiology
Guelph, Ontario, Canada © María Carla Rosales Gerpe, August 2018
ABSTRACT
TROPISM DETERMINANTS OF JAAGSIEKTE SHEEP RETROVIRUS AND ENZOOTIC NASAL TUMOR VIRUS
María Carla Rosales Gerpe Advisor: University of Guelph, 2018 Sarah K. Wootton
Jaagsiekte Sheep Retrovirus (JSRV) and Enzootic Nasal Tumor Virus (ENTV) cause adenocarcinomas of the lung and nasal tract, respectively, in sheep and goats.
Nevertheless, JSRV and ENTV share a high degree of nucleotide (nt) (89%) and amino acid (a.a.) (80%) identity. They also enter cells via the same cell receptor,
Hyaluronoglucosaminidase 2 (Hyal2), an enzyme involved in epithelial, endothelial and chondrocytic hyaluronan metabolism. Finally, they transform tissues using their envelope (Env) glycoprotein. JSRV and ENTV genome sequence comparisons reveal two regions of dissimilarity: the U3 of the Long Terminal Repeats (LTRs) (35% nt), and the cytoplasmic tail of the Envelope glycoprotein (50% a.a.). To uncover genomic regions important for tropism, we developed JSRV-ENTV chimeras within a JSRV backbone, containing ENTV’s LTRs, structural proteins and Env. Because JSRV and
ENTV lack a reporter gene and have difficulty propagating in vitro, we also employed lentivectors (LV) pseudotyped with the JSRV Env (Jenv) and ENTV Env (Eenv), and developed a tissue slice ex vivo model to study entry of JSRV, ENTV and the chimeras.
Our data demonstrated that, unlike JSRV and Jenv LV, ENTV and Eenv LV could not infect or transduce ovine lung tissue slices. Furthermore, JSRV LTRs were statistically significantly less active than ENTV LTRs in primary ovine chondrocyte cells. We also
observed strong staining against Env in ENTV-infected ovine nasal turbinate slices, particularly in chondrocytes. Our in vitro data also showed that lentivectors pseudotyped with Eenv best transduced ovine primary chondrocytes, similarly to cells overexpressing
Hyal2, revealing new aspects of ENTV pathogenesis. These results suggest that JSRV and ENTV tissue selectivity requires both Env and the LTRs. This marked tissue specificity allowed us to explore Jenv as an LV pseudotype in lung gene therapy. We generated two novel Jenv mutants to pseudotype LV, yielding titers similar to Vesicular
Stomatitis Virus glycoprotein pseudotyped LV, capable of efficiently transducing both ovine and Hyal2-expressing murine lung slices. Overall, this thesis proposes that JSRV and ENTV tropism is orchestrated by both their promoters and Env glycoproteins, and that their tissue specificity can be re-purposed for gene therapy to the lung.
iv
“In 2003, Dolly the sheep, the first mammal cloned from an adult cell, died after being diagnosed with an incurable pulmonary adenocarcinoma caused by JSRV.”
Leroux et al. 20061
v
DEDICATION
Para mamá, papá, Mary, y Ryan
vi
ACKNOWLEDGEMENTS
This thesis’ efforts would not have been possible without the support, encouragement and empowerment that my supervisor Dr. Sarah Wootton provided. I have both grown as a scientist and as a person under her supervision. For these reasons, I am deeply thankful to her. I would also like to thank my Thesis Advisory Committee members, Dr.
Peter Krell, Dr. George Harauz and Dr. Alicia Viloria-Petit, whose support, guidance and critical input on my research allowed me to make informed assessments on my thesis work. Scott Walsh, the staff at Ponsonby General Animal Facility, Pathobiology Isolation
Unit, and the Food Science’s Abattoir, all deserve the deepest gratitude for their help to this thesis’ research. Furthermore, I would like to show my appreciation to the Ontario
Veterinary College who supported me financially throughout my PhD with an OVC PhD
Scholarship, and who showed me the benefits of being surrounded by a great research and graduate student community. Finally, I am indebted to the NSERC PGS-D and
University of Guelph Tri-Council Top-up scholarships that allowed me to pursue this degree.
I am also grateful for my peers, Laura van Lieshout, Lisa Santry, Jake van Vloten, Joelle
Ingrao, Thomas McAusland, Betty-Anne McBey, Jondavid de Jong, Jake Domm and
Jodre Datu. I give extra thanks to the latter two for showing the meaning of teaching and proving that mentoring is full of rewards. I sincerely appreciate everyone’s technical assistance and advice during my time at the Wootton lab. Although not in this lab, I would still like to thank Kasandra Bélanger, whose mentorship and friendship during my
Masters cemented the foundations of my scientific journey.
vii
Finally, I would like to thank my parents and sister, my love, Ryan Kim, and friends – especially, Carmen Wong, Stephanie Coit, Eleni Armenakis, Joe Burns, and Kasandra
Bélanger – for pushing and motivating me when times were tough.
This work was funded by an NSERC grant awarded to Dr. Sarah K. Wootton.
viii
LIST OF CONTRIBUTIONS
I performed the work outlined in this thesis under the guidance of Dr. Sarah K. Wootton and my advisory committee (Drs. Peter Krell, George Harauz, and Alicia Viloria-Petit), with the following exceptions:
Jacob van Vloten conducted the resazurin assays and titrated the MG1-eGFP, AdGFP and VACV-GFP viruses in Chapter 2.
Lisa Santry produced and titrated NDV-GFP for Chapter 2.
Laura van Lieshout helped with the mouse delivery and produced AAV-FT-Hyal2 virus for Chapters 3 and 4.
Jakob Domm also produced the AAV-FT-Hyal2 virus for Chapters 3 and 4.
ix
TABLE OF CONTENTS ABSTRACT ...... ii
DEDICATION ...... v
ACKNOWLEDGEMENTS ...... vi
LIST OF CONTRIBUTIONS ...... viii
LIST OF FIGURES ...... xiv
LIST OF TABLES ...... xvi
LIST OF ABBREVIATIONS ...... xvii
Chapter 1: Literature Review ...... 1
Introduction ...... 1
Biology of Betaretroviruses ...... 2
Betaretroviral Life Cycle ...... 9
Pathogenesis of Betaretroviruses ...... 14
Ovine Pulmonary Adenocarcinoma and Enzootic Nasal Adenocarcinoma Pathology 15
JSRV and ENTV Tissue Selectivity ...... 17
The Long Terminal Repeats and their Role in Tropism ...... 17
The Env Protein and its Role in Tissue Specificity ...... 19
Statement of Rationale ...... 24
Hypotheses ...... 26
Experimental Approach ...... 27
x
Chapter 2: Use of precision-cut lung slices as an ex vivo tool for evaluating viruses and viral vectors for gene and oncolytic therapy ...... 30
Abstract ...... 31
Background ...... 32
Reagent Setup ...... 37
Equipment Setup ...... 37
Experimental Procedure ...... 39
1. Preparation of low melting point agarose (30 min) (Day 0) ...... 39
2. Preparation of lung tissue for sectioning with a vibratome (2 h) (Day 0) ...... 39
3. Sectioning with the Vibratome (1 – 4 h) (Day 0) ...... 41
4. Acclimation Period (Day 0 – 3) ...... 42
5. Viability Measurement (Day 3 and Day 28) ...... 42
6. Transduction and infection of lung slices (15 min) (Day 3-4) ...... 43
7. Detection of Viral Transduction/Infection (Day 5-7) ...... 44
8. Immunohistochemistry (4 days) (Week 3-4) ...... 44
Timeline ...... 46
Troubleshooting ...... 47
Potential pitfalls during tissue preparation ...... 47
Transduction and infection of lung slices ...... 49
Anticipated outcomes ...... 50
xi
Chapter 3: Both the U3 and Env protein of Jaagsiekte Sheep Retrovirus and Enzootic
Nasal Tumor Virus contribute to tissue tropism ...... 64
Abstract ...... 65
Introduction ...... 66
Results ...... 69
Jenv is important for mediating ovine lung transduction of lentivectors and chimeric
viruses ...... 69
ENTV but not JSRV is capable of infecting ovine nasal turbinate tissue slices ...... 73
Eenv-facilitated lentivector entry is enhanced in chondrocytes and cells
overexpressing hHyal2 ...... 74
The ENTV LTR is active in ovine primary chondrocytes ...... 76
Discussion ...... 78
Materials and Methods ...... 82
Cell Culture ...... 82
Animals and Tissue Slices Generation ...... 82
DNA constructs and PCR...... 83
Virus Production, Transduction and Infection...... 86
Immunohistochemistry (IHC) ...... 87
Statistical Analyses ...... 88
Chapter 4: Optimized pre-clinical grade production of two novel lentiviral vector pseudotypes for lung gene delivery ...... 95
xii
Abstract ...... 96
Introduction ...... 97
Results ...... 100
LVs pseudotyped with envelope glycoproteins from VSV, EBOV and JSRV differ in
their ability to be concentrated by ultracentrifugation and polyethylene glycol (PEG)
precipitation ...... 100
Pressure use in tangential flow filtration generates high yields in lentivector production
for VSVg and ΔGP Jenv LV pseudotypes ...... 101
Introducing point mutations into the carboxy terminal domain of the JSRV Env protein
increases LV titers ...... 104
Development of a mouse model to evaluate JSRV Env-pseudotyped LV-mediated
transduction ...... 106
Discussion ...... 108
Materials and Methods ...... 112
Cell lines ...... 112
Animals ...... 113
Lung tissue slices ...... 113
DNA constructs ...... 114
Lentiviral vector production ...... 115
AAV6.2FF-FT-Hyal2 production ...... 116
Lentivector purification and concentration ...... 116
xiii
Ultracentrifugation ...... 116
PEG precipitation ...... 117
TFF ...... 117
Downstream TFF Purification and Concentration ...... 118
Vector Titration ...... 119
Transduction of Lung Slices ...... 119
Immunohistochemistry (IHC) ...... 120
Western Blotting ...... 120
Statistical Analysis ...... 120
Chapter 5: General Discussion ...... 132
References ...... 138
xiv
LIST OF FIGURES
Figure 1. Schematic of the genetic structure of a general retroviral provirus and genome...... 4
Figure 2. Representation of the JSRV and ENTV provirus DNA...... 5
Figure 3. General life cycle of retroviruses...... 10
Figure 4. Mechanism of reverse transcription...... 12
Figure 5. Comparison of the long terminal repeats (LTRs) of ENTV–1, ENTV–2 and
JSRV...... 18
Figure 6. Set-up for the perfusion of sheep lungs...... 55
Figure 7. Experimental overview and timeline for generating precision-cut lung tissue slices for viral transduction or infection...... 56
Figure 8. Equipment setup for use of the vibratome...... 57
Figure 9. Sheep lung slices remain viable for a month...... 58
Figure 10. Ovine and murine lung slices can be transduced with gene therapy vectors and murine tumor-bearing lung tissue slices can be infected with oncolytic viruses. .... 59
Figure 11. Sheep and murine lung slices can be infected and virus production can be detected...... 60
Figure 12. Unlike adult ovine tissue, young ovine tissue (lambs) expresses a heat stable alkaline phosphatase (AP)...... 61
Figure 13. Immunohistochemistry for GFP detection and alkaline phosphatase staining as alternatives to circumvent autofluorescence in the lung...... 62
Figure 14. Decreased metabolic activity in murine lung tissue infected with oncolytic vector MG1-eGFP...... 63
xv
Figure 15. Production of JSRV-ENTV chimeric viruses...... 89
Figure 16. Production of lentivirus vectors pseudotyped with VSVg, EBOV, GP64, Jenv and Eenv envelope glycoproteins...... 90
Figure 17. Eenv pseudotyped lentivirus vector (LV) transduces cell lines and ovine lung slices at a lower efficiency than Jenv pseudotyped LV...... 91
Figure 18. Jenv is crucial for mediating viral entry into ovine lung tissue...... 92
Figure 19. Ovine nasal turbinate slices can be infected with ENTV but not with JSRV. 93
Figure 20. Overexpression of Hyal2 facilitates Eenv-LV cell entry and the long terminal repeats from ENTV are highly active in primary ovine chondrocytes...... 94
Figure 21. Comparison of ultracentrifugation and PEG precipitation concentration methods for VSVg, EBOV and ΔGP Jenv lentivirus vectors (LVs)...... 124
Figure 22. Representation of the equipment used in this study...... 125
Figure 23. Optimization of tangential flow filtration as a purification and concentration method for ΔGP Jenv LV...... 126
Figure 24. Toxic effects from tangential flow filtration (TFF) concentrated vector...... 127
Figure 25. Schematic of the methodology used in this study...... 128
Figure 26. Lentivectors pseudotyped by ΔGP Jenv N592T, and ΔGP Jenv M593E have increased Env expression and produce similar titers to VSVg...... 129
Figure 27. Murine lung tissue slice transduction with VSVg, Ebov and ΔGP Jenv lentivirus vectors...... 130
Figure 28. Ovine lung tissue slice transduction with VSVg, EBOV, and ΔGP Jenv lentivirus vectors...... 131
xvi
LIST OF TABLES
Table 1. Lung slice wash and maintenance media composition...... 53
Table 2. Troubleshooting tips for preparing ovine and murine lung tissue slices...... 54
Table 3. Titers and percent recovery of neat and concentrated VSVg, EBOV and ΔGP
Jenv pseudotyped LVs after either ultracentrifugation or PEG precipitation...... 121
Table 4. Titer values and percent recovery of neat and concentrated VSVg LV from low and high pressure TFF conditions...... 122
Table 5. Titer values and recovery yield for neat and concentrated ΔGP Jenv LV using low and high pressure TFF conditions...... 123
xvii
LIST OF ABBREVIATIONS
-sDNA Negative stranded DNA
+ssRNA positive single stranded RNA
µg Microgram
µL Microliter aa amino acid
AAV Adeno-associated virus vector
ALV Avian leukosis virus
AP Alkaline phosphatase
BAC Bronchioloalveolar carcinoma
BSA Bovine serum albumin
B-type Beta type capsid
C/EBV CCAATT enhancer binding protein
CA Capsid cDNA Complementary DNA
CH Cys-His
CMV Cytomegalovirus
CpG Cytosine-phosphate-Guanine
CT Cytoplasmic tail
D-type Delta-type capsid
DMSO Dimethylsulfoxide dNTP deoxynucleotide thymidine triphosphate
xviii dsDNA Double stranded DNA dUTP deoxyuridine thymidine triphosphate dUTPase deoxyuridine thymidine triphosphatase
ECM Extracellular matrix
EDTA Ethylenediaminetetraacetic acid eGFP Enhanced green fluorescent protein
EM Electron microscopy
ENA Enzootic nasal adenocarcinoma enJSRVs Endogenous Jaagsiekte Sheep Retrovirus-like retroviruses
ENTV Enzootic nasal tumor virus
Env Envelope glycoprotein
ESCRT Endosomal sorting complexes required for transport
FITC Fluorescein isothiocyanate
FLAG FLAG peptide (DYKDDDK)
FrMLV Friend murine leukemia virus
FT FLAG tag
FWD Forward
G2 Gap 2 phase of cell cycle
Gag Group antigen gene protein
GFP Green fluorescent protein
GPI glycophosphatidylinositol h Hour
xix
HBSS Hank’s Buffer Saline Solution
HEK 293T Human epithelial kidney 293 transformed with SV40 T antigen cells
HF High fidelity hHyal2 Human hyaluronoglucosaminidase-2
HIV Human immunodeficiency virus
HNF3β Hepatocyte nuclear factor 3 β hPLAP Human placental alkaline phosphatase kb Kilobase kDa Kilodalton iMET Initiation methionine for translation
IN Integrase
IHC Immunohistochemistry
JSRV Jaagsiekte sheep retrovirus
L domain Late domain
LMPa Low melting point agarose
LSMM Lung slice maintenance medium
LSWM Lung slice wash media
LTR Long terminal repeat
LV Lentivirus vector
MA Matrix min Minute
MLV Murine leukemia virus
MMTV Mouse mammary tumor virus
xx
MPMV Mason Pfizer monkey virus
MTOC Microtubule organization center
NC Nucleocapsid
NIH 3T3 National Institute of Health 3-day transfer 3x105 cells nt Nucleotide
OE PCR Overlap extension PCR
OPA Ovine pulmonary adenocarcinoma
ORF Open reading frame
ORF-X Open reading frame X p4, p12, p15, p27 Proteins 4, 12, 15 and 27
PBS Phosphate buffered solution
PCR Polymerase chain reaction
PEG Polyethylene glycol
PIC Pre-integration complex
Pol Polymerase
PPT Polypurine tract
Pro Protease
PT Preprotrypsin
R Repeat
R peptide Membrane fusion regulatory amino acid sequence
RIPA Radioimmunoprecipitation assay buffer
REV Reverse
RNAse H Ribonuclease H
xxi
RTC Reverse transcriptase complex s Second
SRV Simian betaretrovirus
SU surface domain of Env glycoprotein
SV40 Simian virus 40
TFF Tangential flow filtration
TM Transmembrane domain of Env glycoprotein
TNE Tris-NaCl-EDTA buffer
TfR-1 Transferrin receptor 1 tRNA Transfer RNA
Tsg101 Tumor susceptibility gene 101 protein
U3 Untranslated region 3
UTR Untranslated region
YXXM Tyrosine X amino acid X amino acid Methionine motif
W/R Width and Range w/v Weight by volume
Chapter 1: Literature Review
Introduction
In a way, the study of retroviruses is tightly linked with the study of life, the way that cellular processes embark to make a living organism function. Indeed, retroviruses are interlinked with how organisms function, and in many instances retroviral proteins are instrumental to reproduction2, 3. The syncytium protein necessary during placental development evolved from the envelope protein of now endogenous, once exogenous, retroviruses2, 3. Interestingly, oncogenic retroviruses were the first to give us a glimpse into the process of cellular transformation. The first oncogenic retrovirus discovered, the alpharetrovirus Rous Sarcoma Virus (RSV), was discovered as a filterable agent that reproducibly resulted in the formation of tumors. It was later discovered that they encoded potent host-derived oncogenes [reviewed in 4,5-9]. This thesis focuses on two particular oncogenic retroviruses: Jaagsiekte Sheep Retrovirus (JSRV) and Enzootic
Nasal Tumor Virus (ENTV). Studies on JSRV and ENTV have also greatly contributed to research on cellular replication and transformation.
JSRV and ENTV are unique among retroviruses in that they infect sheep (Ovis aries) and goats (Capra hircus) through the airborne route and rapidly generate tumors in the lung and nasal tract of these animals, respectively. Tumor induction occurs as a result of envelope (Env) protein expression through an as–yet–unidentified mechanism 10-14.
Incidentally, JSRV and ENTV are the only replication competent retroviruses known to use the Env protein as the key inducer of tumorigenesis. In addition to these attributes, these viruses have been reported to cause serious economic problems in South Africa,
Scotland and North America 15, 16, with ENTV being much more common in North
1
America than JSRV. Despite both viruses using the Env protein to induce cancer, JSRV causes Ovine Pulmonary Adenocarcinoma (OPA) and leads to lung cancer in sheep whereas ENTV causes Enzootic Nasal Adenocarcinoma (ENA) and results in nasal tumors in sheep and goats 12, 14. Interestingly, OPA shares many similarities with lung adenocarcinoma, a type of human lung cancer previously called Bronchioloaveolar carcinoma (BAC)17, which disparately affects women and particularly, never–smokers18,
19. Moreover, cellular signaling cascades that are normally turned on in lung and in many other types of cancers are also found to be activated in the tumors caused by
JSRV and ENTV 20, 21. As a result, JSRV and ENTV can act as crucial tools in furthering lung and nasal cancer research and their use can also contribute to the collective knowledge of cellular processes. Moreover, studying these two betaretroviruses is of importance to the general public as they can also help us better understand the pathogenesis of retroviral diseases.
Biology of Betaretroviruses
Betaretroviruses belong to the Retroviridae family of viruses, order Ortervirales22, which is divided into two subfamilies: Spumaretrovirinae and Orthoretrovirinae. The former gets its name because of the foamy appearance found in cells infected by viruses in this subfamily. The Orthoretrovirinae encompasses six genera: Alpharetrovirus,
Betaretrovirus, Deltaretrovirus, Epsilonretrovirus, and Gammaretroviruses, as well as
Lentivirus, so–called due to their slow disease progression (reviewed in 22-27). The betaretroviral virions, like other retroviruses, contain 2 copies of positive sense, single, stranded RNA (+ssRNA). Retroviral genomes range between 7–12 kb. For instance,
MMTV is approximately 9 kb long, whereas JSRV and ENTV are approximately 7.5 kb
2
and 7.4 kb, respectively 2, 27-29. Betaretroviral genomes are surrounded by a proteinaceous structure called the capsid (CA), whose diameter ranges between 80–
100 nm. All betaretroviruses have in common a capsid with an eccentric shape.
Virion morphology is based largely on capsid structure and can be determined using electron microscopy (EM). EM studies have shown that betaretroviruses are composed mainly of B–type virion particles, which denote enveloped capsids that, when mature, become eccentric cores. The envelope (Env) of B–type particles also contains protruding structures composed of trimeric surface glycoproteins. In addition, betaretroviruses can also be seen as D–type particles, which, when mature, still contain the protruding Env spikes, but the capsid remains concentric with a cylindrical shape.
JSRV, ENTV and MMTV can be observed as B–type virion particles, yet other betaretroviruses, such as Mason–Pfizer Monkey Virus (MPMV), can be seen as D–type virion particles 23, 25-27, 30.
The +ssRNA of retroviruses is flanked by repeat (R) sequences as well as 5’ and 3’ untranslated regions (UTR), and a 5’ 7–methyl–guanosine cap and a 3’ polyadenylated tail. The genes gag, pro, pol, and env, in that order, are found between the 5’ and 3’
UTRs. The promoter, enhancer sequences and tRNA–primer binding site needed for replication and transcription are found within the 3’ UTR 27. In addition to these canonical 3–4 open reading frames (ORFs), other retroviruses may contain additional ones. Interestingly, for most retroviruses, the Gag–Pro–Pol polyprotein is encoded in the same ORF27 (Figure 1). However, in the case of JSRV and ENTV, the Gag, Pro, and Pol are encoded in three different reading frames and are translated using two ribosomal frameshifts 25.
3
Figure 1. Schematic of the genetic structure of a general retroviral provirus and genome.
A) Retrovirus proviral DNA showing where accessory genes might be found along the genome. The provirus is flanked by Long Terminal Repeats (LTR) encoding untranslated region 3 (U3), repeat region (R), untranslated region 5 (U5), and encodes the gag, pro, pol and env genes. The gag gene encodes the matrix (MA), capsid (CA), and nucleocapsid (NC) proteins. The pro gene encodes the protease (PR). The pol gene encodes the reverse transcriptase (RT) polymerase, and the integrase (IN). The env gene encodes the surface (SU) glycoprotein and transmembrane (TM) subunits of the envelope protein. B) Retrovirus RNA genome. The 5’ leader region starts with the 5’ 7–methyl–guanosine cap and contains the Primer Binding Site (PBS) upstream of the packaging signal (Ψ). The polypurine tract (PPT) is shown in the 3’ as well as the poly-A tail (-AAA). Figure adapted from 31. Only retroviruses with the designated gag–pro–pol–env genomic structure are termed
“simple” retroviruses, whereas those encoding additional proteins are referred to as
“complex” retroviruses. MMTV is an example of a complex betaretrovirus as it encodes an additional gene, the proto–oncogene sag, which stands for superantigen and is encoded within the 5’ long terminal repeat (LTR) 23. Although JSRV and ENTV are normally referred to as simple retroviruses, they also encode an extra ORF, orf–x, whose putative protein Orf–x does not seem to contribute to the replication cycle 32, 33
(Figure 2). They both might also encode an additional accessory protein by the name of
4
Rej, spliced out of the Env mRNA transcripts, that is responsible for the regulation of mRNA export and translation 34, 35.
Figure 2. Representation of the JSRV and ENTV provirus DNA.
Figure adapted from 36. The essential genes in betaretroviruses, like other retroviruses, encode proteins responsible for different aspects of the betaretroviral replication cycle. The gag or group antigen gene, usually encodes the matrix (MA), capsid (CA) and nucleocapsid (NC) proteins. The latter is actually a protein subunit of the CA 24, 25, 27, 36. The NC of lentiviruses directly binds the genomic RNA in HIV–137, 38. The genomic RNA is thought to bind via two Cys–His (CH) motifs present in the NC. These motifs are also found in
JSRV’s NC 25. Using antibodies against the CA of gammaretroviruses, studies have also shown that the genomic RNA binds the gammaretroviral CA 39-42. It is generally assumed that the genomic RNA–NC interaction holds true for all retroviruses, but like with gammaretroviruses, this has yet to be conclusively shown for betaretroviruses due to the lack of an NC–specific antibody. In fact, in the case of alpharetroviruses, the NC
CH–boxes normally found to interact with the genomic RNA were found to be dispensable for genomic RNA packaging. Deletion of the CH motif still permitted genomic RNA binding to occur. Even more interesting, other areas of the CA were also responsible for genomic RNA binding 40.
5
In addition to the Gag subunits mentioned above, gammaretroviruses contain an additional protein, termed p12, which has been linked to various aspects of the life cycle
43, such as integration as well as assembly, and in particular the structural integrity of the virion. Betaretroviruses also contain other additional Gag subunits, which continue to be characterized. In their 1999 study, Palmarini and Fan showed Western blot data of lysed viral particles blotted with a p27 CA antibody and found proteins sized 23, 15, 10 and 5 kDa 11, 44. These results showed that unlike other retroviruses, such as gammaretroviruses, whose MA proteins are 15 kDa, the MA of JSRV is 23 kDa in size.
The gammaretroviral p12 is perhaps similar to JSRV’s p15, the 10 kDa protein corresponds to the NC, and the additional 4–5 kDa protein has an as–of–yet unidentified role. Though the p4 protein has not been fully characterized, it has been speculated to act like the MPMV p4 protein, which interacts with the cytoplasmic chaperon protein, TriC, and could have a role in the assembly of the Gag protein subunits 25, 45. Moreover, the p15 protein is thought to have a role in budding and release because mutation of its L (late domain), markedly diminishes viral budding; this has also been observed with the p12 of gammaretroviruses 43, 44, 46. As mentioned previously, the Gag protein subunits and Pro proteins are not encoded in the same reading frame. In fact, the Pro ORF overlaps the carboxy terminus of Gag.
Unlike the Gag gene, which encodes multiple protein subunits, the pro gene encodes only two proteins, a dUTPase and an aspartic protease (reviewed in 27). The dUTPase has been shown to inhibit uracil from being added to the nascent DNA strand as reverse transcription begins in the newly infected cell (reviewed in 27, 47). This role is crucial since the RT does not efficiently discriminate between uracil and thymidine. As a result,
6
in retroviruses that do not encode a dUTPase, such as HIV–1, when deoxyuridine triphosphate (dUTP) levels are high (due to lack of conversion to dTTP), the RT’s function can be drastically impaired 47. The other protein expressed from the pro gene, the retroviral protease, is responsible for cleaving the subunits of the Gag polyprotein and ensuring that the virus undergoes maturation after egress. Like the Pro ORF, the pol gene begins within the carboxy terminus of the previous gene, pro 25.
The pol gene encodes the reverse transcriptase (RT) polymerase and the integrase
(IN). The discovery of RT shattered the “central dogma”; initially, scientists thought that proteins were translated from mRNA that was transcribed from DNA. However, some
RNA viruses, later called retroviruses, such as RSV, managed to persist and generate chronic infections in their hosts, as well as quickly transform the cells they infected. If the information flow followed the DNA–RNA–protein order, managing persistent chronic infections and quick transformations would require a massive viral titre provided by extensive and quick viral replication. Dr. Howard Temin, who discovered RT, noticed that RSV did not fit this criterion as the virus appeared mostly dormant. The gammaretrovirus Murine Leukemia Virus (MLV), responsible for causing lymphomas in mice, is another transmissible agent with a slow viral replication rate 24, 27. With these observations, Temin argued that perhaps retroviruses used a different route to maintain an infection or transform cells. Indeed, retroviruses would have to use a DNA replication intermediate, which he called the provirus, to produce viral proteins and continually generate retroviral genomic RNA. Thanks to the RT, this provirus can be generated and then, with the use of the IN, be inserted into the host’s chromosome, guaranteeing persistent infection27. Following the Pol ORF, the Env polyprotein is encoded. As is
7
customary with betaretroviruses, the env gene also commences within the ORF of Pol, specifically in the C–terminus of the IN.
The env gene encodes the Env protein subunits, surface (SU) and transmembrane (TM) glycoproteins (Figure 1), which represent the spikes observed protruding from the envelope of B–type particles using EM. The SU and TM subunits are translated from a singly spliced mRNA, as mentioned earlier. The SU protein is supposedly responsible for the interaction with the cell entry receptor, which happens to be the hyaluronoglucosaminidase 2 (Hyal2) receptor in the case of JSRV and ENTV–12, 48, 49.
The TM subunit ensures the SU is anchored to the cellular membrane. The C–terminus of the TM contains a region called the cytoplasmic tail (CT), which faces the inner portion of the virion or the cytoplasm of the cell prior to budding and egress. The CT is thought to play a major role in the fusogenicity of retroviruses in general50. Interestingly, the CT of betaretroviruses JSRV and ENTV is ~45 amino acids (aa), which is longer than many other retroviruses. A particular portion of the CT, the R peptide, is in fact cleaved during virion maturation by the viral protease for some retroviruses, like gammaretroviruses, thereby increasing the membrane fusion properties of their envelope glycoprotein51. In contrast, this does not occur with JSRV nor with ENTV, and concordant with the previous observation, this lack of processivity of the CT is detrimental to JSRV’s fusogenic properties. Indeed, artificial truncation of JSRV’s CT increases transduction of JSRV Env (Jenv) pseudotyped retroviral vectors into various cell lines, whereas it is normally restricted to sheep lung cells (Wootton lab, unpublished data)52, 53. The most peculiar characteristic of the envelope protein of sheep betaretroviruses, JSRV and ENTV, is that it functions as a potent oncogene. In fact,
8
Jenv and Eenv are the only known retroviral oncogenic proteins that do not act in a trans–activating fashion or that require mutation to activate their transformation activity, with the exception of the alpharetrovirus Avian Leukosis Virus (ALV) and the gammaretrovirus Friend Murine Leukemia Virus (FrMLV) 25, 54-56.
Betaretroviral Life Cycle
Betaretroviruses, like most retroviruses, utilize their Env glycoproteins to bind cellular receptors to entercells (Figure 3). For example, MMTV enters murine cells via the
Transferrin receptor 1 (TfR–1)57 and both JSRV and ENTV–1 are thought to use the
Hyal2 receptor48, 49, 58, which belongs to the hyaluronan–glucosaminidase family of glycophosphatidylinositol (GPI)–anchored membrane proteins responsible for hyaluronic acid degradation in the extracellular matrix59. Interestingly, ENTV–1 has a lower binding affinity for Hyal2 than JSRV does, but the reason for this has yet to be fully understood48, 58, 60.
Upon entry, the viral envelope fuses with the plasma membrane, exposing the core to the cytoplasm. In this step during entry, the core is referred to as the reverse transcriptase complex (RTC). The RTC becomes activated by cytoplasmic factors, which could include free deoxynucleotide triphosphate molecules (dNTPs). It is also thought that this pool of dNTPs signals the RT to begin reverse transcription of the genomic RNA to make the proviral double stranded DNA (dsDNA). As mentioned previously, the dUTPase also plays an important role during reverse transcription ensuring dUTP is not incorporated within the newly generated DNA strand thus avoiding it from being degraded. Reverse transcription is accomplished via template switching between the two genomic RNAs and following a series of steps that will be outlined
9
below 24, 27, 36, 61, 62. The RT of retroviruses is notorious for having poor fidelity. However, both template switching and poor fidelity contribute to contiguous genomic evolution of the retrovirus. Oddly enough, Walsh et al. previously showed that ENTV–1’s RT might actually be very efficient compared to those of other retroviruses. The lack of marked genetic diversity amongst ENTV–1 isolates could be a reflection of RT fidelity; though copying of the ENTV genome through tumor cell division is also a possibility16.
Figure 3. General life cycle of retroviruses.
A) Early Steps of the Retroviral Replication Cycle: entry, reverse transcription and integration. B) Late Steps of the Retroviral Replication Life Cycle: Transcription, Translation, Assembly and Budding. Figure adapted from 63. Reverse transcription begins with the use of a tRNA primer. A tRNA primer is an adaptor molecule, normally used during translation within the host cell. However, for retroviruses, in the RTC, it is used instead to prime production of the retroviral dsDNA intermediate by binding to a complementary site defined as the primer–binding site.
Commonly used tRNAs are Lys, Trp, Pro and iMet tRNAs (reviewed in 27, 64). As the RT polymerase generates the new negative sense DNA (-sDNA) strand in a 5’ to 3’ fashion, the RNaseH subunit of the polymerase degrades the RNA template. The directionality
10
of the RT polymerase does not permit it to reverse transcribe the whole genome starting from the tRNA binding site since this is located in the 5’ of the genomic RNA; it must switch templates. The repeat region, where RT has stopped, acts as primer for the second round of reverse transcription. The RNaseH continues to degrade the genomic
RNA but spares a region of the 3’ end of the genomic RNA, called the polypurine tract
(PPT), which is resistant to degradation. The PPT then serves as the primer for the generation of the complementary positive sense DNA. The RT once more stops reverse transcribing as it reaches the end of the newly-generated -ssDNA. This portion serves to prime the final DNA intermediate strand. As a result of the template switching, the proviral DNA intermediate becomes surrounded by the LTRs composed of the untranslated region 3 (U3), repeat sequence (R), and the untranslated region 5 (U5)
(Figure 4) 27.
Once the proviral DNA has been generated, the core undergoes more changes and becomes known as the pre–integration complex (PIC). Depending on the retrovirus, the
PIC may shuttle to the nucleus or remain in the cytoplasm until cell division occurs.
HIV–1 uses its accessory protein Vpr, which can halt cells in the G2 phase and has the capacity to shuttle into the nucleus, to create pores in the nuclear membrane and allow passage of the HIV–1 PIC into the nucleus 65. For gammaretroviruses, however, integration occurs only once the nuclear envelope has dismantled as part of cell division
24. It is thought that the p12 protein plays a major role at this step by tethering the PIC to the host’s chromosome 66. Particularly for betaretroviruses and specifically for JSRV and ENTV, the PIC also does not shuttle to the nucleus, but rather awaits the dissolution of the nuclear membrane during mitosis 25. Thus, unlike HIV–1, which can
11
replicate in both dividing and non–dividing cells, gammaretroviruses and betaretroviruses are both limited to the former. Within the PIC, the IN then processes the ends of the proviral DNA and utilizes the free hydroxyl groups to tether the provirus to the host’s chromosome, becoming a permanent feature of the genome of the cell.
The retroviral genome can now be transmitted horizontally. If the retrovirus reaches germline cells, however, the transmission can be passed vertically to the offspring 27.
Figure 4. Mechanism of reverse transcription.
PBS stands for primer-binding site in this figure. Figure adapted from 27. Other than being efficiently transmitted to daughter cells, the advantage of integration is that it allows for the virus to utilize the cellular machinery to synthesize viral mRNA and translate its protein subunits to assemble new virus particles that can infect neighbouring cells. To produce the protein subunits, the Gag–Pro–Pol polyprotein is
12
translated from unspliced mRNA, whereas the Env protein is transcribed from a singly spliced mRNA. The polyproteins are later processed by the viral aspartic protease and a cellular furin protease in the case of the Env proteins 67. During assembly, the Gag subunits undergo polymerization thanks to scaffolding domains present within the subunits. The scaffolding domains in MPMV, for instance, are found within its p12 protein. Myristoylation of the MA protein has been determined to be an important step in the multimerization of the Gag. The NC Cys–His motif also binds the packaging signal located in the 5’ of the genomic RNA nascently transcribed, allowing packaging of the retroviral genome 25, 27, 68, 69. Most retroviruses assemble at the membrane, but betaretroviruses actually assemble in the cytoplasm and in proximity to the microtubule organization center (MTOC) 36. Budding occurs after the assembled retrovirus acquires some of the cellular membrane as its own envelope, with the help of the L domain of the p15 protein in betaretroviruses 68, 70, 71. Interestingly, the L domain, which seems to act independent of the Gag polyprotein, is crucial for virion budding and interacts with
ESCRT pathway proteins Tsg101, AP–2 and Nedd4 family proteins, responsible for trafficking and endocytosis 72. Of note, Tsg101 localizes to the MTOC during cell division 73. Enlisting such protein to help in budding and release might explain why betaretroviruses assemble intracellularly rather than at the plasma membrane.
Finally, once budding occurs, maturation takes place extracellularly. In betaretroviruses, like all retroviruses, the protease is released via self–cleavage from the Gag–Pro–Pol polyprotein transcript, after conformational changes presumably allow catalytic residues to come into proximity. A reduction–oxidation reaction has been postulated as the possible catalysis that occurs during self–cleaving. Once the protease is released, it can
13
then process the Gag polyprotein into each active subunit. In doing so, the core of most retroviruses transforms into a different form than their immature concentric shape; in the case of betaretroviruses, the new core is eccentric. The maturation step is vital for the retroviral replication cycle to continue. A malfunctioning protease unable to modify the core results in a decrease in infectious particles 25, 27.
Pathogenesis of Betaretroviruses
JSRV and ENTV are found both exogenously and endogenously. In fact, multiple endogenous retroviruses of sheep are genetically similar to exogenous JSRV and
ENTV-12, 29, 69, 74. Ovine endogenous betaretroviruses, also known as enJSRVs, presently share a symbiotic relationship with their host species, the sheep. Interestingly, the promoter of enJSRVs is influenced by progesterone. As a result, enJSRVs transcripts, particularly the singly spliced env transcript, accumulate during pregnancy.
The Env protein of human endogenous retroviruses plays a major role in placental development by creating syncytia. It is possible that the enJSRVs Env could be doing the same in sheep2.
Notably, sheep are immunotolerant toward ENTV and JSRV. Since enJSRVs are of major significance during gestation in sheep, it can be expected that retroviruses bearing striking resemblance to JSRVs would not elicit an immune response. Tolerance to exogenous ovine betaretroviruses, such as ENTV and JSRV, may be developed by having the enJSRVs expressed in the thymus to ensure that the immune system maintains the placenta as an immunosuppressive environment. As a result, exogenous retroviruses may not be recognized as non-self antigens 75.
14
Betaretroviruses, in general, cause a variety of illnesses, which range from chronic, persistent infections that lead to autoimmunity to cancer as they can transform different cell types and tissues 23, 27, 76, 77. For instance, the simian betaretrovirus (SRV) is responsible for a strong immunosuppression in macaques, leading to diarrhea, marked weight loss, inflammation of the spleen, anemia as well as other symptoms. In rare instances, SRV can also develop into B-cell lymphomas (reviewed in 76, 78). On the other hand, MMTV develops mainly into breast cancer in mice and, in some cases, also causes T-cell lymphomas (reviewed in 23, 77). In particular, JSRV and ENTV cause adenocarcinomas of the respiratory tract of sheep. JSRV causes OPA, while ENTV induces ENA (reviewed in 36).
Ovine Pulmonary Adenocarcinoma and Enzootic Nasal Adenocarcinoma
Pathology
OPA was first reported in the 19th century, where it was first described with the name jaagsiekte, an Afrikaans word that translates to “chasing sickness”79. This chasing sickness described the excessive dyspnoa that is observed in animals suffering from both ENA, and OPA, which can decimate up to 80% of flocks1. OPA is most often described in sheep, but it can also occur in goats and mouflon79. Unlike ENA, OPA may affect animals at all ages and the clinical symptoms are distinct for both lambs and adult sheep19. However, OPA is most commonly reported in adult sheep 2 – 4 years of age.
Unlike ENA, OPA tumors are polyclonal in nature, and though they also do not metastasize, there might be more than one lesion in the lung79. Tumors typically arise after many months or years; however, once clinical symptoms are apparent, animals may succumb to the disease as early as 90 days19, 79. OPA is caused by JSRV, which
15
can also be found in ewe’s milk and colostrum. This is a potential route of infection; nevertheless, most commonly, the virus is transmitted through heavy nasal exudate.
The virus preferred cell types, type II pneumocytes and cuboidal cells, are found in the distal lung, although lymphocytes may also be infected. However, propagation of the virus is mostly observed in type II pneumocytes and lymphocytes typically do not produce virus79.
ENA was first described in the 1950s80 and is characterized as a low-grade carcinoma that originates in the ethmoid region of the nasal turbinates of both sheep and goats81.
ENA is also characterized as a viral disease that is easily transmitted through nasal secretions. ENA typically occurs in young animals, around 2 – 4 years old, but sex and breed have not shown to correlate with the onset of disease. Typically, 0.5 – 2% of animals in a flock are affected by ENA, but the numbers can rise to 15%. Animals suffering from ENA typically display difficult breathing, nasal discharge, eye bulging and skull deformations as a result of tumor formation in the ethmoid turbinates80, 81. The tumor may grow into the pharynx, paranasal sinuses and skull cavity, but the tumor is clonal in nature, and no metastases are typically observed80. ENTV particles have been found in multiple studies in ENA tumors80-82, and experimental ENTV-induced ENA has been reported14. The preferred cell type of ENTV has not yet been reported; however, it is thought that an epithelial cell is the most plausible cell type for initial ENTV infection as the carcinomas tend to arise from an epithelial source80, 81. Despite JSRV and ENTV being highly similar at a genomic and protein level, the reasons for their differences in tropism remain unanswered.
16
JSRV and ENTV Tissue Selectivity
The Long Terminal Repeats and their Role in Tropism
The LTRs contain the untranslated region 3 (U3) which harbours the promoter and enhancer sequences that are required for viral replication and transcription — reviewed in 27. The LTRs are the primary driving force behind tropism for gammaretroviruses, and in particular for Murine Leukemia Viruses (MLVs). The LTR’s interaction with T–cell– specific transcription factors allows MLVs to spread in T–cells and eventually, to generate lymphomas83. Another example is that of the murine betaretrovirus Mouse
Mammary Tumor Virus (MMTV). The tissue tropism of MMTV is strongly influenced by its LTRs, which are known for being rather long compared to other retroviruses. In addition to the Repeat (R), untranslated region 5 (U5) and U3, MMTV’s LTRs also encode the super antigen (sag) gene. They also happen to contain sequences that can bind hormones and other transcription factors that aid the virus spread in mammary tissue, resulting in breast cancer, and help MMTV replicate in T–cells, generating lymphomas — reviewed in 23. Furthermore, the LTR of the betaretrovirus MPMV also plays a significant role in its tropism 84.
Interestingly, the U3 region of the LTRs is the genomic region that differs the most (by
35%) between JSRV and ENTV genomes 16. The differences between the U3 sequences could dictate whether a virus replicates in one cell or another, i.e. its tropism.
Lung epithelial cell–specific transcription factors can influence JSRV’s tissue specificity
85-88. In fact, though some of these transcription binding sites are present in JSRV, they are absent in ENTV 87, 89. Transcription factors that potentially play a role in JSRV infection are the lung– and liver–specific Hepatocyte Nuclear Factor (HNF)–3β and the
17
CCAAT Enhancer Binding Protein (C/EBP)–α and β, responsible for driving the expression of lung–specific genes 86, 88. Interestingly, the HNF–3β binding site is found only in JSRV isolates whereas the C/EBPα– and β–binding sites are conserved among both JSRV and ENTV isolates (Figure 5) 87. Remarkably, HNF–3β interacts with LTRs in type II pneumocytes, one of the cell types that JSRV infects. However, HNF–3β does not have a role in the spread of JSRV in club cells, which are also thought to be permissible to JSRV’s infectivity 86, 88.
Figure 5. Comparison of the long terminal repeats (LTRs) of ENTV–1, ENTV–2 and JSRV.
Figure adapted from 87. Unfortunately, though the role of the LTRs in JSRV and ENTV infectivity has been further clarified thanks to the latter studies, it remains unclear whether the differences in transcription factors that interact with JSRV but not with ENTV is what determines their
18
distinct tropism. Part of the problem with tackling the LTR question is the lack of appropriate cell lines to study JSRV or ENTV. Most studies have been performed in rodent cells, particularly in NIH 3T3 murine fibroblasts. These cells have become pivotal in the study of JSRV’s and ENTV’s Env as an oncoprotein since the sheep betaretroviral LTRs are essentially transcriptionally silent in these cells, and NIH 3T3s do not detach when overgrown, and yet are able to develop transformed foci 90, 91. In
2011, the Wootton lab published a study where they looked at whether the LTRs in fact were involved in determining tissue–selectivity, and found promising results showing expression of AP in the liver and lung when under the control of the JSRV LTR but in the nasal tract when under the control of the ENTV LTR. However, the study found no significant difference in the development of tumors induced by expression of Jenv or
Eenv under the control of either JSRV or ENTV’s LTRs in mice. Since it remains unclear whether this is due to the study being performed in a species other than sheep, it would therefore be interesting to repeat this study in sheep, the preferred host of these viruses
92.
The Env Protein and its Role in Tissue Specificity
The role of the Env protein is complex as it is also an oncoprotein; Jenv’s oncogenic activity was first discovered in NIH 3T3 mouse fibroblasts. The LTRs of JSRV and
ENTV are transcriptionally silent in these cells, allowing for a more accurate assessment of the other retroviral genes 33. Eenv was later shown to transform rodent fibroblasts in a study published in the Journal of Virology in 200260. Another study by
Wootton et al. proved that indeed Jenv was solely responsible for tumorigenesis, and that tumor development did not require the Hyal2 receptor in mice 93, 94. This was
19
surprising, as Hyal2 has been implicated in tumorigenesis. For instance, the ECM undergoes major structural changes during breast cancer and the chromosome encoding Hyal2 is usually deleted during breast and lung cancers 95-97. As a result,
Hyal2 was regarded as a plausible tumor suppressor 48, 96, 98.
The process by which the Env protein of both of these retroviruses transforms cells remains largely misunderstood. It was previously postulated that a Y590XXM593 motif found in the cytoplasmic tail (CT) of the transmembrane (TM) domain of the Env protein was responsible for the downstream activation of a PI3K/Akt signaling cascade 99, 100.
The tyrosine residue of this motif has also been found to be indispensable for transformation 101. None of the endogenous ovine betaretroviruses of JSRV and ENTV are transformation–competent nor do any harbour a Y590XXM593motif in their Env’s TM domain (Figure 5) 102, 103. Furthermore, the cytoplasmic tail of the TM, where the
Y590XXM593 motif is located, is thought to be crucial for oncogenesis. A detailed study by
Fan’s group showed that a deleted or mutated CT severely hindered Jenv’s ability to transform cells 104. As a result, the exact mechanism by which Env induces tumorigenesis remains incompletely explained to date.
Other than being an oncoprotein, Env is the most obvious causative agent for spread of the virus since it interacts with the cellular receptor that allows viral entry. Both JSRV and ENTV are thought to bind the constitutively expressed Hyal2 cell–surface receptor
49, 59, 105. This enzyme belongs to a family of GPI–anchored cell–surface proteins, responsible for degrading hyaluronic acid (HA), a major component of the extracellular matrix (ECM) — reviewed in 98. Notably, Hyal2 plays significant roles during clotting and fertilization. During the former, for instance, Hyal2 helps break down HA in platelets to
20
trigger an inflammatory cytokine response 106-108. Recently, Hyal2 has also been shown to play a role during fertilization where it helps sperm break through the ECM to achieve fusion with the oocyte 109. Despite these roles Hyal2 has a lower hyaluronidase activity than other members of its family, suggesting that its ubiquitous expression may be differentially regulated depending on the cell type. For example, in platelets, Hyal2 requires activation before it can be shuttled to the cell surface and begin fragmentation of HA 108. Others have also suggested that Hyal2 plays a different role from the canonical HA degradation attributed to members of its family 98. Furthermore, as shown in platelets, Hyal2 may not only be present at the plasma membrane but rather, it may also be found intracellularly 110, and in particular, in lipid rafts 107.
Interestingly, only the ovine and the human Hyal2 receptors interact with the JSRV Env
(Jenv) and ENTV Env (Eenv) proteins; the murine Hyal2 does not permit entry of JSRV or of ENTV into murine cells, despite the occurrence of transformation 48. Of note, the presence of Hyal2 at endogenous levels does not appear to be sufficient for the spread of ENTV. However, overexpression of Hyal2 does boost ENTV infection. Indeed, Eenv has a lower binding affinity for Hyal2 than Jenv has for this receptor. In addition, ENTV also requires a lower pH than JSRV in order to gain entry to cells 111. The differences between JSRV and ENTV’s Env and JSRV’s and ENTV’s ability to enter a cell also extend to the Eenv cytoplasmic tail (CT), a region of retroviruses known for membrane fusogenicity 51, 52. The CT of JSRV and ENTV differs by 50% at the amino acid level, which could explain why Moloney MLV pseudotyped with Eenv is unable to transduce as many cell lines as JSRV is capable of transducing 16, 58, 60, 103.
21
Jenv and Eenv may be more different than previously thought, and the mechanism behind tissue tropism is possibly more complex than initially expected. Therefore, it is important to point out that, though the Env protein may play a significant role in tropism, the ability of these viruses to enter cells does not necessarily ensure a cell’s permissibility to their replication. In other words, an ENTV or JSRV virion may be able to enter a host cell and yet, be incapable of properly replicating due to the lack of factors that normally bind to the LTR promoter and enhancer sequences. Therefore, tissue– selectivity remains a definitive joint effort between these two viral elements.
Furthermore, studying the underlying factors responsible for tropism has been difficult due to the lack of a cell line permissible to JSRV and ENTV replication11, 25, 36, 112. As such, to understand the factors responsible for tropism, studies in the past have relied on pseudotyping lentivectors or retroviral vectors with the envelope proteins of JSRV and ENTV or on infecting small ruminants with JSRV and ENTV52, 53, 58, 111, 113, 114. In vitro studies have been unable to paint a complete picture of the mechanism behind the tissue specificity of JSRV and ENTV. Moreover, though in vivo studies with small ruminants have thus advanced our knowledge of JSRV and ENTV2, 12, 19, 29, 36, 54, 90, 94,
113, 115, 116, they are costly endeavours. An alternative to in vivo studies is the generation of tissue slices, which maintain the morphology and architecture of the organ112, 117-123.
This advent has gained widespread use in the scientific community for performing studies on physiology and pathogenesis112, 117-124. As a proof of principle, in 2005,
Cousens et al. successfully infected ovine lung tissue slices with JSRV112. This study introduced another tool with which to scrutinize the problem of tropism in JSRV and
ENTV. Understanding the mechanism behind their tissue selectivity could help us
22
develop future gene therapeutics capable of targeting specific areas of the respiratory tract affected by disease or cancer. Furthermore, studying both of these betaretroviruses can help us comprehend additional aspects of the replication cycle of retroviruses and improve the efficiency of current transduction methods for gene therapy vectors.
23
Statement of Rationale
JSRV and ENTV are highly similar betaretroviruses with different disease manifestations and great significance in small ruminant health. JSRV, in particular, is capable of decimating up to 80% of sheep in a flock. ENTV, on the other hand, is becoming more prevalent in Canada and is responsible for most of the nasal adenocarcinomas encountered in post-mortem. Working with JSRV and ENTV has relied mainly on the use of large animal models (sheep and goats), which can be costly.
Unfortunately, the lack of a completely permissive Hyal2 receptor to JSRV and ENTV infection in mice and rats makes it difficult to use small animal models. There is no study in the literature that details the generation of a permissive Hyal2-expressing murine lung model. Furthermore, there is currently no cell line that can support virus infection by JSRV or ENTV.
The similarities of, and different tissues affected by JSRV and ENTV make these viruses unique for the study of retroviral pathogenesis. Recently, Cousens et al. reported using ovine lung slices as a tool to study JSRV infection in vitro. These lung slices will have the cell types (type II pneumocytes and cuboidal cells) that have previously been identified as being permissive to JSRV12, 86, 90, 92, 112. Therefore, these lung slices would be an effective tool to investigate the difference in tropism between
JSRV and ENTV. Furthermore, utilization of nasal turbinate slices could further validate the difference in tissue selectivity. To date, there is no study that has utilized hybrid viruses between JSRV and ENTV to infect ovine lung and nasal turbinate slices. Finally, this is the first work where utilization of a novel murine model expressing human Hyal2
24
was employed to produce tissue slices to test the tropism of these viruses and the potential to use JSRV envelope pseudotyped LV in gene therapy to the lung.
25
Hypotheses
1. Ovine lung tissue slices are capable of being kept viable and can be transduced
or infected by lentivectors or JSRV and ENTV betaretroviruses, respectively.
2. The distinct tissue tropism of JSRV and ENTV–1 is due to differences in the U3
region of the LTR or differences in the overlapping cytoplasmic tail of Env.
3. C57BL/6 mice transduced with an AAV vector expressing the human FLAG-
Hyal2 receptor can be used as a novel mouse model to test the utility of Jenv-
and Eenv-pseudotyped LV for lung gene delivery
4. Jenv pseudotyped LV can efficiently transduce Hyal2-expressing murine lung
tissue slices
26
Experimental Approach
Objective 1: Develop an ex vivo system of ovine tissue slices to study JSRV and ENTV
Key Steps Involved:
• Perfusion of lung tissue with 2% low melting point agarose
• Generation of tissue slices with a vibratome
• Maintenance of slices in DMEM supplemented with growth factors and antibiotics
• Viability determination with a live/dead stain and a metabolic resazurin assay
• Transduction of tissue slices with LV expressing green fluorescence protein
(GFP) reporter gene
• Visualization of transduction using fluorescence microscopy
• Infection of tissue slices with JSRV
• Fixation and paraffin embedding of tissue slices
• Visualization of JSRV infection by immunohistochemistry (IHC) against Env
Objective 2: Determine the factors responsible for the distinct tissue tropism of JSRV and ENTV
Key Steps Involved:
• Development of JSRV-ENTV chimeras using overlap extension polymerization
chain reaction (OE PCR)
• Infection of lung tissue slices with chimeric viruses
• Testing of chimeras for viral replication via Western blot, electron microscopy
(EM) and immunohistochemistry of tissue slices
27
• Utilization of Jenv and Eenv pseudotyped GFP-expressing LV to transduce lung
tissue slices
• Visualization of transduction via fluorescence microscopy
• Testing of Jenv and Eenv pseudotyped LV in vitro under untransfected and
FLAG-tagged (FT) Hyal2 transfected conditions
• In vitro assessment of alkaline phosphatase (AP) expression under the control of
JSRV and ENTV LTRs
Objective 3: Exploit JSRV tissue selectivity for gene therapy
Key Steps Involved:
• Transduction of C57BL/6 mice with adeno-associated virus (AAV) vector
encoding a preprotrypsin signal peptide (PT) followed by FT human Hyal2
(hHyal2) (FT-hHyal2) (AAV-PT-FT-hHyal2)
• Testing for FT-Hyal2 expression in respiratory tract via IHC against FT
• Generation of lung tissue slices from mice stably transduced by AAV-PT-FT-
hHyal2
• Generate Jenv mutants able to pseudotype LV and yield titers similar to those
obtained with vesicular stomatitis virus glycoprotein (VSVg) pseudotyped LV
• Testing of mutant Jenv protein expression via Western blot
• Development of optimized production, purification and concentration methods for
the Jenv mutant pseudotyped LVs by testing and comparing ultracentrifugation,
polyethylene glycol (PEG) precipitation and tangential flow filtration (TFF)
28
• Transduction of lung tissue slices with lentiviral vector encoding AP and
pseudotyped with Jenv, Jenv mutants and VSVg
29
Chapter 2: Use of precision-cut lung slices as an ex vivo tool for evaluating viruses and viral vectors for gene and oncolytic therapy
María C. Rosales Gerpe 1, Jacob P. van Vloten1, Lisa A. Santry1, Jondavid de Jong1, Robert C. Mould1, Adrian Pelin2,3, John C. Bell2,3, Byram W. Bridle1, and Sarah K. Wootton1*
1Department of Pathobiology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada 2Ottawa Hospital Research Institute, Centre for Innovative Cancer Research, Ottawa, Ontario, K1H 8L6, Canada 3Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Ontario, K1H 8M5, Canada
Short Title: Tissue Slices as an Oncolytic and Gene Therapy Tool
*Correspondence should be addressed to Dr. Sarah Wootton, Department of Pathobiology, University of Guelph, Guelph, Ontario, Canada, N1G 2W1, 519-824-4120 ext. 54729 [email protected]
Submitted to journal of Molecular Therapy - Methods and Clinical Development. Currently in revision.
30
Abstract
Organotypic slice cultures recapitulate many features of an intact organ including cellular architecture, microenvironment and polarity, making them an ideal tool for ex vivo study of viruses and viral vectors. Here, we describe a procedure for generating precision-cut ovine and murine tissue slices from agarose-perfused, normal and murine melanoma tumor-bearing lungs. Furthermore, we demonstrate that these precision-cut lung slices can be maintained up to a month and can be used for a range of applications, which include characterizing the tissue tropism of viruses that cannot be propagated in cell monolayers, evaluating the transducing properties of gene therapy vectors, and finally, investigating tumor specificity of oncolytic viruses. Our results suggest that ex vivo lung slices are an ideal platform for studying the tissue-specificity and cancer cell-selectivity of gene therapy vectors and oncolytic viruses prior to in vivo studies, providing justification for pre-clinical work.
31
Background
The preparation of tissue microslices for ex vivo experimentation is a technique that produces precision-cut organotypic slices with the use of a vibratome or tissue slicer that operates with a vibrating blade. The use of a vibrating blade allows for greater accuracy and reproducibility when generating precision-cut lung slices (PCLSs)125.
Compared to in vivo work, tissue slices are cost-effective and less time-consuming, while still possessing many of the attributes of the intact organ123, 125, 126. In fact, PCLSs have been utilized in the study of lung anatomy since they maintain the structural framework at both the tissue and cellular level, while also retaining all relevant cell types117, 119, 121, 123, 125, 127. PCLSs are often utilized in toxicological and anatomical studies regarding contractility in relation to asthma and other respiratory illnesses such as emphysema117, 125. Recently, lung slices have been used in immunological studies, where they have been shown to retain their immune cell populations and functions, which mimic that of the organ127-130.
With regard to infectious diseases, lung slices have been used as a model system to study respiratory pathogens, including adenoviruses, influenza viruses, respiratory syncytial virus, and the bacterium Clamydophila pneumoniae, ultimately contributing to our understanding of the pathogenesis and replication cycles of these infectious agents128-132. Additionally, lung slices have been used to evaluate the transducing activity of various viral vectors120, 124, 133-137. For example, in a study evaluating lentiviral vectors (LV) pseudotyped with the Ebola virus (EBOV) glycoprotein (G) as a potential gene delivery vector133, 134, the EBOV-G-bearing LV did not infect cell monolayers as efficiently as Vesicular Stomatitis Virus (VSVG)-G-pseudotyped LVs; however, the
32
EBOV G-pseudotyped LV was more efficient at transducing cells in three-dimensional culture than VSV-G LV was134. Finally, PCLSs can be used as a platform to study viruses, such as Jaagsiekte sheep retrovirus (JSRV), that are unable to replicate in cell monolayers due to their requirement for cell membrane polarity133, 134, 138-147. Testing viruses in tissue slices would also expedite and help to refine virus-driven gene therapy and oncolytic virus (OV) studies, which ultimately require the use of in vivo models to test their efficacy. Thus, the use of lung slices that preserve many of the in vivo conditions can be used to predict gene and oncolytic therapy outcomes in animal models.
To this end, we present a detailed protocol for generating lung tissue slices from both small (mice) and large animals (sheep) and from normal and tumor-bearing tissue.
Generating slices from the latter preserves the tumor macro- and microenvironment that can never be recapitulated in conventional cellular monolayers or three-dimensional cultures and allows tumor selectivity of an OV to be evaluated relative to adjacent normal tissue. We show that this protocol can be used to: successfully propagate viruses with cell polarity issues (JSRV); allow for efficient transduction with gene therapy vectors (VSV-G-pseudotyped LV and adenovirus (Ad-Green Fluorescent
Protein (GFP)), and support infection with oncolytic viruses (Maraba (MG1-enhanced
GFP (eGFP), vaccinia (VACV-GFP) and Newcastle disease virus (NDV-GFP)). Notably, this protocol could readily be adapted to a broad array of viruses and tissue types.
33
Required Materials
Reagent List (supplier, catalogue #)
Low Melting Point Agarose Ultrapure (Fisher Scientific, 16520100)
1X Hank’s Buffered Salt Solution (HBSS) with Ca2+/Mg2+ (Fisher Scientific, SH3026802)
Industrial Strength Glue (Loctite Super Glue Ultra Gel Control)
Phosphate-Buffered Saline without Calcium/Magnesium/Phenol Red (Fisher Scientific,
SH3037803)
Dulbecco’s Modified Eagle Medium (DMEM) High Glucose with L-Glutamine (Fisher
Scientific, SH30022FS)
Recombinant Human Hepatocyte Growth Factor (Cedarlane, 100-39-10UG(PE))
Recombinant Human Keratinocyte Factor (Gibco Life Technologies, PHG0094)
L-Glutamine (Fisher Scientific, SH3003402)
Penicillin/Streptomycin (Fisher Scientific, SV30010)
Amphotericin B (Fisher Scientific, SV3007801)
Gentamycin (Fisher Scientific, 15710064)
Dexamethasone (Sigma-Aldrich, D4902-25MG)
8-bromo-adenosine 3’,5’-cyclic monophosphate (Sigma-Aldrich, B5386-5MG)
3-isobutyl-1-methylxanthine (Sigma-Aldrich, I5879-100MG)
Hexadimethrine bromide (Polybrene) (Sigma-Aldrich, H9268-5G)
34
Fetal Bovine Serum (Fisher Scientific, 12483020)
10% Formalin (Fisher Scientific, SF100-4)
Ethanol (Fisher Scientific, A962F-1GAL) to make 70% Ethanol
Isopropanol (Fisher Scientific, BP2618-4)
Xylene (Fisher Scientific, HC7001GAL)
Hematoxylin (Fisher Scientific, SH26-500D)
Superfrost Plus Slides (ThermoScientific, 22-037-246)
C57BL/6 Mice (Charles River)
Specific-Pathogen-Free (SPF) Cornell Star Sheep (neonates – 6-month-old lambs)
Live/Dead Viability Dye (Life Technologies, LSL7013)
Resazurin dye (Sigma-Aldrich, R7017-5G)
Equipment List (supplier, catalogue #)
Styrofoam box lids, Saran wrap and wooden skewers
8-mm tissue puncher (TedPella Biopunch, 15111-80)
50-mL Centrifuge Tubes (Fisher Scientific, 14955240)
Petri Dishes 100-mm x 15-mm (Fisher Scientific, FB0875712)
48-well plates (VWR, 10062-898)
96-well plates (VWR, 10062-900)
35
Ted Pella Inc. Microslicer DTK-3000W (Vibratome)
Nutating Mixer Fixed Speed 120V (Fisher Scientific, 88861041)
5- and 20-mL syringes
18-gauge needles
Fine-point forceps
Scalpel
Surgical scissors
Clamps (hemostats)
P1000 Gilson pipette
P1000 pipette filter tips
10-mL serological pipettes
KimWipes (Sigma-Aldrich, Z188956)
CO2 incubator
Fluorescent Microscope
Foam Biopsy Sponges 25-mm x 31-mm (Thermo Fisher Scientific, 8453)
Micromesh Loose Biopsy Cassettes (Thermo Fisher Scientific, B1000735WH)
Microscope Cover Glass (Fisher Scientific, 12-545-C 22X40-1)
Superfrost Plus Microscope Slides (Fisher Scientific, 12-550-15)
36
Dakocytomation Pen (Dakocytomation, S2002)
Thermo Scientific Excelsior ES Tissue Processor
Thermo Scientific HistoStar Tissue Embedding Station
Thermo Scientific Finesse ME+ Microtome
Reagent Setup
Low melting point agarose (LMPa) gel. Prepare 2% (w/v) LMP-a in 1X HBSS 1 h prior to perfusion of the lungs, by adding 4 g of LMPa to 200 mL of 1X HBSS and allowing it to dissolve through heating via a microwave by pulsing every 2 min or until boiling is observed, for approximately 6 min until the LMPa has solubilized. Once the
LMPa has solubilized, distribute the 200 mL among four 50-mL centrifuge tubes and place in a water bath pre-heated to 42˚C to avoid solidification until ready to perfuse.
Wash and maintenance media solutions. Please see Table 1 for recipes on how to make the appropriate medium. Make sure to have these solutions ready prior to cutting.
Medium and polybrene. For experiments involving lentiviral vectors, use a final concentration of 8 µg/mL of polybrene (stock: 8 mg/mL; 1 µL polybrene to 1 mL medium) to aid in transduction148. This can be prepared the day of transduction/infection with LV or JSRV, respectively.
Equipment Setup
Perfusion of ovine lungs. Set up a platform to maintain lungs in an upright position using styrofoam lids for insulation. See Figure 6.
37
Vibratome settings. Remove the buffer tray platform and add ice to the reservoir. Turn on the lamp and set the blade settings to 100% frequency at a speed of 8 µm/s. Set the tissue cutting settings to 300-µm thickness, and a blade travelling range of 20 – 40 mm, depending on the number of blocks in the platform.
Thermo Scientific Excelsior ES tissue processor settings. Use the following program: Six 23-second cycles of exposure to alcohols (70%, 85%, 90%, 95%, and 2 cycles of 100% isopropanol) at 30˚C, three 23-second cycles in xylenes at 30˚C and 3 cycles of 23 seconds in wax at 62˚C.
38
Experimental Procedure
Please consult Figures 6 and 7 for an overview of the procedure. Note that the steps in this protocol should all be performed in a containment level/biological safety level-2 laboratory within a type II biological safety cabinet because the viruses used in this protocol are designated as Risk Group 2 pathogens, and because the lungs should be kept as sterile as possible.
1. Preparation of low melting point agarose (30 min) (Day 0)
Prepare 2% LMPa in 1X HBSS 1 h prior to procuring lungs (see the Reagent Setup section). Depending on the lung size, the volume may range from 60 mL to 200 mL for lamb-derived lungs, or less than 5 mL for all five lung lobes from a mouse.
2. Preparation of lung tissue for sectioning with a vibratome (2 h) (Day 0)
It is important to note that this step must be performed quickly, shortly after euthanasia of the animal, while lungs are warm and the agarose is still in its liquid state in order to ensure that it can completely perfuse lung tissues. In the case of mice, remove the lungs, separate each lobe and inject LMPa gel using a 5-mL syringe directly into each lobe by inserting the needle into the tissue and perfusing slowly and carefully so as not to damage the lung architecture. Stop once noticeable inflation is observed; this typically requires about 1 mL of LMPa per murine lung lobe.
In the case of sheep, set up the lungs as shown in Figure 6A and 6B. Set up two styrofoam box lids by covering them with paper towels and plastic (Saran) wrap. Spray the Styrofoam box lids with 70% ethanol and prop them against each other with wooden skewers. Cut off the sheep trachea two inches from the primary bronchi and hold it in an
39
upright position with a clamp. Use another clamp to secure the trachea to the vertical
Styrofoam box. This will ensure the lungs remain upright throughout the entire procedure. Remove one of the 50-mL conical tubes containing LMPa gel from the 42˚C water bath and set aside at room temperature to cool for 5 min to avoid burning the tissue. Then, using a 20-mL syringe filled with LMPa, inject the liquefied gel down the trachea into the lungs.
The rate of flow of gel going down the trachea into the lungs should be relatively fast so as to avoid solidification of the LMPa due to over-cooling (Table 2). Once there is an observable increase in lung volume as evidenced by a slight inflation in the lungs, it can be assumed that the lungs have been properly perfused (Figure 6B). Clamp the lungs shut with surgical forceps and cool them by submerging them in cold (4˚C) 1X HBSS on ice for 45 min to 1 h for sheep, and 30 min for mice to allow the LMPa to solidify.
Proceed to cut the ovine lungs with a scalpel into smaller 2 – 3 mm thick sections
(Figure 6C) and use an 8-mm biopunch to core each section (Figure 6D). The final core should be approximately 2 mm thick with an 8-mm diameter. For mice, each individual lung is the appropriate size and will not need to be cored.
Pour a thin layer of LMPa over a Petri-dish that is on ice to provide a base to completely surround the cores with LMPa. Once this layer has solidified, transfer each sheep lung core or murine lung lobe into this Petri-dish. Next, pour another thin layer of LMPa over them, leaving part of the core exposed; do not cover the sections completely yet. This will secure each core or lobe and prevent them from floating. Once this layer has solidified, pour enough LMPa to completely cover the cores or lobes (Figure 6E). Leave
40
the Petri dish with the LMPa-embedded lung tissue on ice for 30 min to allow the agarose to completely solidify.
3. Sectioning with the Vibratome (1 – 4 h) (Day 0)
Fill the vibratome reservoir with ice. Replenish the ice to ensure the buffer tray platform remains cool at all times. Set the vibratome blade frequency to 100% and the speed to
8 µm/s, using the blade settings panel (Figure 8A). Spread glue over the buffer tray platform (Figure 8B). Using a scalpel, cut out a block containing a core or lobe from the
Petri dish with around 2 mm of agarose still surrounding the tissue, dab the block dry with a Kimwipe, and place the block on top of the glue to fasten the tissue block to the vibratome platform. We recommend fitting as many blocks side by side as possible to increase yield (e.g., 4 to 6 blocks can fit in the platform) (Figure 8C and 8D). Bring surgical scissors to cut cartilage when necessary (Table 2).
Set the vibratome thickness parameter to 300 µm and adjust the blade travel range to cover the distance the blade will travel (40 mm should cover the platform) using the width and sectioning range (W/R) function in the control panel (Figure 8B). Once everything is set up, insert the ceramic blade into the blade holder and position the blade above and behind the blocks by selecting the Manual function and then pressing the Down button to lower the platform, and the reverse (“Rev”) button to place the blade behind the blocks. Begin cutting by selecting the Auto function and pressing the Start button, ensure that the blade is cutting and not simply pushing the block (Table 2)
(please refer to Troubleshooting section). You may press Stop at any time.
Once the blade has sectioned through the tissue and a slice has been generated, remove the slice and place it in lung slice wash medium (LSWM) (Table 1). Once all
41
lung slices have been collected, use forceps to place the lung slices in a 48-well plate (1 slice per well). To reduce the chances of contamination, wash each slice with 200 µL of
LSWM five times using a P1000 pipette and incubate them in 200 µL of the appropriate maintenance media, including antibiotics penicillin/streptomycin, gentamycin, amphotericin (refer to Table 1 for exact composition) and change the media every other day (Figure 6F). The slices will typically last up to 4 weeks under these conditions.
Note: Avoid suction with Pasteur pipettes and a vacuum aspirator, which can readily aspirate the lung tissue slices, suctioning them right to the waste.
4. Acclimation Period (Day 0 – 3)
Maintain the lung slices by continuing to change the media for fresh media (200 µL) every day for three days prior to infection.
5. Viability Measurement (Day 3 and Day 28)
Viability of slices can be assessed using a live-dead viability stain and results visualized using an inverted fluorescence microscope (Figure 9A) or using a resazurin viability assay (Figure 9B) and can be done during the acclimation time window (Day 3) to select viable slices for infection and during and after infection (Day 28). Live-dead staining has been previously detailed112. Briefly, the stain is composed of two dyes, a green dye that easily transverses the cell membrane and a red dye that can enter cells only if there is damage to the cell membrane. Remove the media from the lung slices with a P1000 pipette and wash twice with 1X PBS before adding the stain. Lung slices can be treated with 10% Triton-X and left in the fridge for 24 h, to be used as a positive control for cell death, before being treated with the stain. Visualize lung slices with a fluorescence microscope according to the manufacturer specifications.
42
The resazurin assay has also been described149. Add LSMM-B media containing resazurin to lung slices in 48-well plates. Incubate plates in a 37˚C, 5% CO2 incubator.
After two hours, move each lung slice into a 96-well plate with fresh LSMM-B media without resazurin for quantification of fluorescence using a plate reader. Resazurin is converted to a fluorescent compound by metabolically active cells, and, therefore, higher fluorescence values correlate with greater viability. We arbitrarily defined lung slices as unviable when they had fluorescence values equal to or lower than one standard deviation below the mean of untreated lung slices and recommend excluding those from experiments (Figure 9). Typically, 93-99% of slices remain viable up to a month after the procedure (Figure 6F and 9) when using the maintenance media recipe detailed in this paper.
6. Transduction and infection of lung slices (15 min) (Day 3-4)
Production of JSRV, lentiviral vectors, Ad-GFP, MG1-eGFP, VACV-GFP, and NDV-GFP has been extensively described in the literature11, 12, 112, 150-156, and we have used similar methods. For retroviruses and lentivectors, remove medium from the lung slices and add fresh medium containing polybrene at a final concentration of 8 µg/mL. Next, infect with 2 x 106 U (Reverse Transcriptase units) of JSRV concentrated by ultracentrifugation or 1 x 106 transducing units (TU)/mL for the VSV-G GFP LV. For the other viruses, polybrene is not needed; simply replace with fresh media and add 1 x 108 plaque forming units (PFU)/mL for MG1-eGFP, 5 x 106 PFU/mL for VACV-GFP, 1 x 107
PFU for NDV-GFP or 1 x 107 PFU Ad-GFP per lung slice dropwise. Place the 48-well plate on a rotating platform (Nutating Mixer Fixed Speed 120V) inside a tissue culture
43
incubator at 37˚C with 5% CO2 overnight. Change media of the lung slices the next day to remove remaining virus in the supernatant.
7. Detection of Viral Transduction/Infection (Day 5-7)
At 48 – 72 h post-transduction, visualize the lung slices transduced with VSV-G GFP LV
(48h) (Figure 10A), or infected with Ad-GFP (72h) (Figure 10B), VACV-GFP (72h),
NDV-GFP (72h) (Figure 10C) or MG1-eGFP (48h) (Figure 11C-D) using fluorescence microscopy. For JSRV, continue to maintain the lung slices as described previously for
3 – 4 weeks prior to fixing and performing immunohistochemistry (IHC) (Figure 11A-B).
Collect media from lung slices infected with JSRV to measure reverse transcriptase activity as described previously14.
8. Immunohistochemistry (4 days) (Week 3-4)
Remove media from the lung slices and wash them twice with 1X PBS. Fix the lung slices in 10% formalin for approximately 16 h. After fixation, wash the lung slices three times with 1X PBS before placing them in 70% ethanol. Transfer the lung slices to an embedding cassette by spreading them on top of two foam pads and then store in 70%
Ethanol. Place the cassettes in an automated tissue processor (Thermo Scientific
Excelsior ES) and begin fixation, dehydration, and paraffin wax infiltration of the tissues by following the program described in the Equipment Setup section of this paper.
CAUTION: Do not use conventional overnight processing programs as leaving the thin tissue slices in high temperatures might lead to slices being easier to shear during cutting with the microtome.
44
Once finished, proceed to embed tissues in paraffin blocks using an embedding work station (Thermo Scientific HistoStar) and protocols as described in the instruction manual. The lung slice paraffin blocks should be stored at -20˚C since blocks should be cold during sectioning. The next day, use a microtome (Thermo Scientific Finesse ME+) to prepare 5 µm sections from the blocks. Keep blocks on ice while waiting to use the microtome.
CAUTION: Do not use the trim function; instead use “section” as you risk losing the lung slices.
Using fine-point tweezers, place tissue sections in a 42˚C water bath, and using clean tweezers, place the section onto a Superfrost Plus slide. Charged slides improve tissue adhesion to the glass. Leave the slide on a heat block at 37˚C to dry overnight (>16 h) and the next morning, proceed to IHC as previously described 93, 112.
45
Timeline
Day 0: Prepare tissue for vibratome and slice tissue. Wash slices.
Day 0-3: Maintain and culture lung tissue slices in appropriate maintenance medium.
Day 3-4: Infect and transduce lung slices.
Day 5: Change media of lung slices.
Day 5-7: Check for fluorescence of GFP reporter gene. Take pictures. Collect media from slices to measure virus titer.
Weeks 2 – 3: Continue culturing lung slices for JSRV. Collect media from lung slices infected with JSRV to measure reverse transcriptase activity.
For long-term experiments:
Day 28: Collect lung slices and fix 16 h in 10% formalin.
Day 29: Place lung slices in cassettes and store in 70% ethanol overnight and proceed to tissue processing.
Day 30: Paraffin-embed lung slices into blocks and leave in freezer overnight
Day 31: Cut specimen blocks, mount on slides and leave baking overnight.
Day 32: Perform IHC to visualize virus transduction.
46
Troubleshooting
Tissue preparation and cutting
Please consult Table 2 for troubleshooting during tissue preparation and vibratome use.
Lung slice architecture
Although sheep lung tissue slices can be kept for up to four weeks in culture (Figures 6 and 9), there is a reduction in size over time (Figure 9A). This loss of structure could be due to the dissolution of agarose125. To reduce this effect, we recommend making wider slices (> 8 mm) for longer experiments.
Variability in lung slice viability
We recommend generating large numbers of tissue slices from each murine lung sample to obtain a more representative mean fluorescence for the resazurin assay.
Potential pitfalls during tissue preparation
This protocol requires an uninterrupted day and depends on the experimenter’s speed and efficiency. We recommend obtaining lungs that are as fresh as possible. The shorter the time the lungs are outside the body prior to perfusion with LMPa gel, the better the quality of the lung slices (Table 2). In general, PCLSs range in size between
150 µm to 500 µm, although most studies use sizes between 250 and 300 µm119, 121, 123,
125-128, 131, 132.
With respect to the vibratome, two main parameters need to be optimized: the frequency of vibration to create an easier cutting path for the blade, and the speed of the blade to generate slices in a timely manner; these parameters are specific to each
47
type of tissue. The recommended manufacturer frequency and speed settings for tissues such as lung are 80-100% and 10 µm/s, respectively, in the case of the vibratome we used. In this study, we preferred using 100% and 8 µm/s, for frequency and speed, respectively, for lung tissue slices. With these settings, we were able to generate lung tissue slices at a rate of one slice per minute. Though it was possible with certain cores to increase the speed, prolonged speed increments had the potential of warping and shearing lung tissue slices. In fact, with a speed greater than 10 µm/s, the blade would move the block off the platform, instead of cutting the tissue.
The vibratome procedure can take between 1 to 4 h depending on how well the tissue is being cut. The vibratome used in this study comes equipped with a very sharp ceramic blade; however, we found that cutting heavily depends on the tissue architecture. The amount of time spent with cutting depends on the variation in perfusion and presence of cartilage, which sometimes needs to be removed. We recommend adding up to 6 blocks into the buffer tray platform to yield as many lung slices per minute as possible and decrease the time it takes to complete the procedure. We also suggest keeping a watchful eye during the entire procedure to ensure that: 1) the blade is cutting, not pushing, the tissue; 2) cartilage is not preventing efficient cutting; and 3) the platform is not heating up. Increasing the frequency of the vibration can introduce heat. Adding ice to the platform and the buffer tray platform can ensure consistently low temperatures.
Lung Tissue Slices Preparation Time
To shorten tissue preparation times, we considered freezing tissues post-LMPa perfusion; however, freezing at this point significantly reduced tissue viability (data not shown). Freezing tissues after cutting with the vibratome, however, did preserve
48
viability. In this case, tissues were frozen in 10% DMSO-LSMM-C (Table 1), after two days of culturing. Upon addition of the freezing media, the lungs were frozen overnight at -80˚C before being transferred to a liquid nitrogen tank for long-term storage.
Thawing was rapidly executed at 37˚C and the slices were allowed to culture for three days to one week prior to experimental use. Although thawed slices did not seem to have reduced viability, freezing drastically reduced transduction efficiency. Therefore, we do not recommend freezing for lung slice work that involves viral infections or transductions. Recent studies have looked at cryopreservation of lung slices and have also noted similar drawbacks121, 123. Perhaps a longer acclimation period could resolve this problem.
Transduction and infection of lung slices
After generating the slices with the vibratome, we recommend acclimating the lung tissue slices in culture for three days prior to transduction or infection with the lung slices. We found that this acclimation period was important for transduction and infection efficiency. We also found that transducing with GFP-expressing viruses yielded punctate foci indicative of efficient transduction of the lung slices (Figure 10A-C).
However, it is important to note that a strong promoter should be used to drive expression of the GFP reporter gene. Lung slices grown for more than three weeks exhibited large amounts of auto-fluorescence that were difficult to discern from positively transduced foci, compared to negative untreated tissue. Other studies have also noted auto-fluorescence in the lungs as a potential drawback from using GFP as a reporter gene157, 158. Therefore, we recommend using other reporter genes such as
49
mCherry and heat stable human placental alkaline phosphatase (hPLAP) for long-term studies.
However, we found that for sheep lung slices from lambs less than 6 months of age, the hPLAP reporter gene is not a viable option since lambs express a form of heat-stable
AP in their lungs (Figure 12). Therefore, other reporter genes such as mCherry, nuclear
β-galactosidase or firefly luciferase might be useful for ovine lung tissues. Alternatively, the use of immunohistochemistry could circumvent the autofluorescence problems observed. We were able to detect the presence of GFP using a rabbit anti-GFP
(Invitrogen, A11122), while no GFP was detected in the no virus control (Figure 13).
Furthermore, using murine lung tissue slices that naturally do not express heat-stable
AP, we were able to effectively transduce these lung tissue slices with VSVg hPLAP LV, and Ebola virus glycoprotein pseudotyped hPLAP LV (Figure 13).
Anticipated outcomes
The use of tissue slices is gaining widespread use117, 121, 123, 125, 127-132 and becoming an attractive alternative to in vitro cell culture experiments due to the presence of a cellular framework and microenvironment similar to that observed in vivo117, 123, 125, 127. In this study, we optimized the maintenance of ovine and murine lung tissue slices in culture to evaluate small ruminant betaretroviruses (JSRV), gene therapy vectors VSV-G GFP LV and Ad-GFP, and oncolytic viruses MG1-eGFP, VACV-GFP and NDV-GFP. With this protocol, we were capable of transducing and maintaining infection with all of the aforementioned viruses in lung slices (Figures 10-11, 13). The media recipes we used to maintain infection with these viruses (Table 1) can serve as a reference for future
50
virus work in organotypic slices but might need to be optimized depending on the transcription factors important for the virus. For instance, JSRV’s promoter region has two lung and liver-specific hepatocyte transcription factor (HNF-3) binding sites36, 85-88,
90, and we found that addition of these two transcription factors was better for infection.
We also measured viability using two different assays: a live-dead stain for ovine lung slices and a well-established resazurin assay for murine lung tumor slices112, 159. Using the live-dead viability assay, we were able to determine that the sheep lung slice tissue could remain viable for approximately one month (Figure 9A), longer than reported for other studies, regardless of the species and tissue-type112, 128, 129, 132, 137. Interestingly, we found that the sheep lung tissue slices that were viable and capable of harbouring transduction or infection (Figure 10A and 11A), would often be accompanied by migrating lung fibroblast-like cells that would adhere to the bottom of the plate. We ascertained that this observed characteristic was a marker of tissue slice viability.
Interestingly, these adherent cells also had the capacity of being transduced by the
VSV-G GFP LV (Figure 11A). These cells could be harvested as a primary cell line and constitute another reason for the use of tissue slices. We also took advantage of the resazurin assay to determine murine lung slice viability without compromising the ability to subsequently infect them with viruses. Resazurin can be added to lung slice media for two hours, and then washed and replaced with normal media without significantly affecting lung slice viability. Thus, this protocol provides two alternative methods for measuring viability that future studies could benefit from. We also found that the OVs were found in mostly tumor tissues or the peri-tumoral region in the lung slices (Figure
12A and B), as has also been well established in the literature160-165. Moreover, we
51
observed a decrease in tumor tissue metabolic activity when in the presence of oncolytic vector MG1-eGFP, using the resazurin assay, a surrogate tool for viability
(Figure 14). Therefore, this protocol is a convenient tool too to assess the activity of other gene and oncolytic virus vectors prior to and to complement in vivo work. We expect that this protocol can also be applied to other pathogens and other tissues.
52
Table 1. Lung slice wash and maintenance media composition.
Four types of media were used to treat the ovine and murine lung slices. A Lung Slice Wash Media (LSWM) was used for both types of lung tissue slices to prevent contamination. Three different types of Lung Slice Maintenance Media (LSMM) were used. Both the LSWM and LSMM media contained the following antibiotics, penicillin, streptomycin, gentamycin, and amphotericin. LSMM-A contained the bare minimum standard media reagents such as 10% FBS and 1% L-glutamine and was used to culture the VSV-G GFP LV-transduced ovine lung slices. LSMM-B contained the same reagents as LSMM-A in addition to the recombinant epithelial growth factor. LSMM-B was used to culture murine lung tumor slices infected with MG1-eGFP, VACV-GFP, NDV-GFP and Ad-GFP. Finally, LSMM-C was composed of the same reagents as LSMM-B but also contained a human recombinant hepatocyte factor required for propagation of JSRV in sheep lung slices.
50U/mL Penicillin/Streptomycin
0.2µg/mL Gentamycin
1.25µg/mL Amphotericin
10µM 8-bromo-adenosine 3’,5’-cyclic Lung Slice Wash Medium (LSWM) monophosphate
100µM 3-isobutyl-1-methylxanthine
100nM Dexamethasone
DMEM with High Glucose
Lung Slice Maintenance Medium (LSMM) Reagents in LSWM
for Infection with LV system (A) AND 10% FBS; 2mM L-Glutamine
Lung Slice Maintenance Medium for Reagents in LSMM-A
infection with Maraba virus, Vaccinia virus, AND 10 ng/mL Recombinant Human
NDV and Adenovirus (B) Keratinocyte Factor
Reagents in LSMM-B Lung Slice Maintenance Medium for AND 5 ng/mL Recombinant Human infection with JSRV (C) Hepatocyte Factor
53
Table 2. Troubleshooting tips for preparing ovine and murine lung tissue slices.
Observation Possible Explanation Troubleshooting
Dried-up, leathery Lungs are getting colder Work fast; if LMPa gel does not go easily, then directly re-perfuse texture individual cores with LMPa gel Avoid if possible
Bloody, dark spots Lungs too hot Do not pour LMPa gel. Cut lung lobes, and directly perfuse the cores in lung lobes Time in hot room exceeded instead Avoid if possible
LMPa gel rate too Agarose too hot Wait at least 1 hour after preparing the LMPa gel fast Avoid if possible
LMPa gel rate too Lungs are cold Ensure that the lungs are fresh so that core temperature remains warm. slowly If delayed, lungs can be left within bag inside Styrofoam container at 37˚C room for no more than 30 min Directly perfuse the cores instead
Lungs appear Agarose poured too hot, or Check that the conical tube containing LMPa gel feels warm and not hot mushy when Waited less than 45 min for to touch; avoid this if possible cutting sections agarose to solidify Leave lungs incubating for longer to allow them to solidify
Blade is pushing Lungs not properly perfused, Re-infuse the core directly using warm LMPa gel with 5mL syringe and the core or a 30G ½ needle Blade is in contact with Have surgical scissors at hand to cut the cartilage cartilage Glue core-gel block again or use another one Core-gel block not fastened to platform
Tissue is sheared Agarose poured too hot Check that the conical tube containing LMPa gel feels warm and not hot Lung tissue is over-perfused to touch; avoid this if possible Ice melting in reservoir Select another core according to Figure 2 criteria Add more ice to reservoir. Ensure that vibratome platform remains cold throughout procedure. Drop a small ice cube on top of the tissue if possible.
Lung tissue media Not enough washes Wash the lung slices at least 5 times prior to culturing them and make appears cloudy sure to change the media with fresh media every other day
54
Figure 6. Set-up for the perfusion of sheep lungs.
(A) Two styrofoam box lids are lined with paper towels and enclosed with plastic wrap prior to being positioned 90˚ to each other with the use of wooden skewers at room temperature. Clamps (hemostats) are then attached to the trachea and used to position the lungs upright. Another set of hemostats is put through the clamps holding the lungs and adhered to the vertical Styrofoam cover. The initial state of the sheep lungs is characterized by the lobes appearing deflated and folded onto themselves (A); however, upon perfusion (B) they no longer fold and appear engorged. (C) A section approximately 2 mm thick is generated using a scalpel and then (D) an 8 mm puncher is used to core the slice to generate 8 mm- wide, 2 mm-thick cores. The cores may vary in perfusion level, as the lungs will be differentially perfused throughout. The colored bars under the lung cores denotes the difference in perfusion level that can be observed in one section, with the orange bar being the highest leaving the core spongy, and the light blue bar showing a flat section, lacking perfusion. The appropriate lung cores (pink bar or 2nd from the left) can then be moved into a Petri dish and embedded within LMPa (E). (F) The use of antibiotics extends culturing of sheep lung slices. Greater survival was observed when lung slices were cultured over a period of 1 to 4 weeks throughout 6 months with antibiotics, compared to without antibiotics. Those maintained in media with antibiotics were also subjected to 5 washes prior to culturing.
55
Figure 7. Experimental overview and timeline for generating precision-cut lung tissue slices for viral transduction or infection.
56
Figure 8. Equipment setup for use of the vibratome.
The vibratome contains two setting panels: (A) a blade settings panel on the left, and (B) a tissue cutting settings panel on the right. The blade settings panel can adjust the frequency and speed of setting, while the tissue cutting settings regulates the thickness of the slice and the blade travel range along the buffer tray platform. The reservoir should be filled with ice throughout the procedure. (C) and (D) show representative tissue blocks for ovine and murine lung tissues, respectively.
57
Figure 9. Sheep lung slices remain viable for a month.
(A) Sheep lung slices were cultured in Lung Slice Maintenance Media-C (Table 1) and harvested after 1 and 4 weeks. At both time points, the lung slices were tested for viability using a live-dead viability stain and the results were visualized using a Zeiss Leica inverted fluorescence microscope using the FITC (for live cells) and Texas Red (for dead cells) channels. Lung slices treated with 10% Triton-X 100 and left in the fridge for 24 h were used as positive controls for cell death. After 4 weeks, slices which where observed as redder than green throughout the entire tissue slice were deemed as dead. (B) Mouse lung slice viability as assessed by a metabolic assay. Resazurin was added to mouse lung slices in LSMM-B for two hours and then fluorescence intensity of each sample was read by a plate reader (530/25 nm; 595/5 nm). Lung slices with fluorescence intensity of two or more standard deviations below the mean were excluded from experimentation (indicated by red arrows).
58
Figure 10. Ovine and murine lung slices can be transduced with gene therapy vectors and murine tumor-bearing lung tissue slices can be infected with oncolytic viruses.
VSV-G-LV-transduced ovine lung cells (A) and Ad-GFP-transduced mouse lung cells (B) can be seen as green fluorescent foci in the center and edges of the lung slices. Transduction was possible in standard cell culture medium supplemented with additional antibiotics (LSMM-A; Table 2). Migrant cells from the lung slices are also susceptible to transduction (A). Lung slices were imaged by fluorescence microscopy to detect infection with NDV-GFP or VACV-GFP 72 h post-infection (C). B16-F10 tumors are outlined in white dotted lines in the BF images. GFP expression can be observed as punctate spots in and around the tumors. Uninfected lung slices show dark B16-F10 tumors (arrows). GFP puncta localized in and/or around the tumors can be identified in NDV-GFP- and VACV-GFP-treated slices.
59
Figure 11. Sheep and murine lung slices can be infected and virus production can be detected.
Lung slices grown in LSMM-C were infected with JSRV and incubated for four weeks at 37˚C. (A) Robust immunohistochemical staining of a lung slice infected with JSRV using a monoclonal antibody against the JSRV Env oncoprotein. An uninfected lung slice was used as a negative control and lacked staining. Tissue from the lung of an adult sheep suffering from ovine pulmonary adenocarcinoma was used as a positive control (B) Reverse transcriptase (RT) activity in the supernatant of mock and JSRV-infected lung slices. Murine B16-F10 melanoma tumor-bearing lung slices grown in LSMM-B were infected with oncolytic Maraba virus (MG1-eGFP). (C) Top: Uninfected lung slices show dark B16-F10 melanoma tumors (arrows). Bottom: GFP puncta localized around and within the tumor (arrows) indicates infection by MG1-eGFP. (D) MG1-eGFP virus is detectable in supernatant from infected lung slices. Viral titers (1 x 105 PFU/mL) from infected lung slice supernatants (N=6) were determined by TCID50 48 h post-infection.
60
Figure 12. Unlike adult ovine tissue, young ovine tissue (lambs) expresses a heat stable alkaline phosphatase (AP).
(A) Ovine lung tissue slices from a 9-month old lamb shows dark staining for the presence of AP post heat-inactivation at 65˚C for 1 h. (B) Heat-inactivated paraffin-embedded tissue from adult sheep (left) does not show purple staining characteristic of AP in contrast to young lamb tissue (right).
61
Figure 13. Immunohistochemistry for GFP detection and alkaline phosphatase staining as alternatives to circumvent autofluorescence in the lung.
(A) IHC targeting GFP for ovine lung tissue slices not transduced (left) or transduced (right) with GFP- expressing VSVg pseudotyped lentivector. Transduced lung tissue cells expressing the GFP reporter gene are stained brown. (B) Alkaline phosphatase (AP) staining of murine lung tissue slices not transduced (left) or transduced with VSVg (middle) and Ebola virus (EBOV) glycoprotein (right) pseudotyped lentivectors expressing the human placental AP reporter gene. Purple foci denote the presence of AP.
62
Figure 14. Decreased metabolic activity in murine lung tissue infected with oncolytic vector MG1- eGFP.
(A) Metabolic activity measure via the resazurin assay of murine, lung normal and tumor tissue two hours prior to infection. (B) Metabolic activity measured via resazurin 48 h post-infection with MG1-eGFP. Unpaired t-tests were conducted for each data set (A and B) using GraphPad Prism 7 software (GraphPad Software, LaJolla, CA, USA). The p values were 0.7751 (A) and 0.0124 (B).
63
Chapter 3: Both the U3 and Env protein of Jaagsiekte Sheep Retrovirus and Enzootic Nasal Tumor Virus contribute to tissue tropism
María C. Rosales Gerpe 1, Scott R. Walsh2, Darrick L. Yu, Laura P. van Lieshout1, Jakob M. Domm1, Joelle C. Ingrao1, Jodre Datu1, Jondavid de Jong1, Michelle Beaudoin-Kimble3, Emma Lamoure3, Mark B. Hurtig3, and Sarah K. Wootton1*
1Department of Pathobiology, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada. 2McMaster Immunology Research Center, Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada . 3Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada.
Running title: U3 and Env Orchestrate JSRV and ENTV Tissue Tropism
*Correspondence should be addressed to S.K.W.: Mailing address: Department of Pathobiology, Ontario Veterinary College, University of Guelph. Building 89, 50 Stone Road East, Guelph, Ontario, Canada, N1G 2W1. Phone: +1 (613) 824 4120, x54729. E-mail: [email protected]
Submitted to Journal of Virology
64
Abstract
Jaagsiekte Sheep Retrovirus (JSRV) and Enzootic Nasal Tumor Virus (ENTV) are small-ruminant betaretroviruses that share high homology at the DNA and protein levels, utilize the same cellular receptor, hyaluronoglucosaminidase 2 (Hyal2) for entry, and drive oncogenesis through their envelope (Env) glycoprotein; yet, they target distinct areas of the respiratory tract, the lung and nose, respectively. Their distinct tissue selectivity makes them ideal viruses with which to study the pathogenesis and tropism of retroviruses. To uncover the genetic determinant for tropism, we constructed
JSRV-ENTV hybrid viruses and generated lentivectors pseudotyped with the Env proteins from JSRV (Jenv) and ENTV (Eenv). We also developed a tissue slice ex vivo model that retains cell polarity thereby enabling infection with JSRV and ENTV. Through infection of lung and nasal turbinate slices, we observed that the viral envelope protein is crucial for entry, but that the promoters of these viruses are also responsible for tissue-specificity. Interestingly, we also found that Hyal2 abundance seems to act as the limiting step for ENTV entry. Furthermore, we show evidence of ENTV Env protein expression in the extracellular matrix of ENTV-infected ovine nasal turbinate tumors, which is naturally enriched with Hyal2. Our work points to a tentative novel ENTV pathogenic mechanism. Finally, our study also shows JSRV and ENTV tropism likely stems from cooperation involving the cellular receptor, envelope glycoprotein, and the
LTR.
65
Introduction
Jaagsiekte sheep retrovirus (JSRV) and Enzootic nasal tumor virus (ENTV) are two small-ruminant betaretroviruses, whose infection of sheep results in ovine pulmonary adenocarcinoma (OPA) and enzootic nasal adenocarcinoma (ENA), respectively12-14, 25,
74, 93. JSRV and ENTV share high nucleotide and amino acid identity and utilize the same cell entry receptor, hyaluronoglucosaminidase-2 (Hyal2)48, 54 for entry. They also utilize their envelope (Env) protein as an oncoprotein to drive transformation, which is rare amongst retroviruses25, 54, 94, 166. Despite these similarities, these two viruses affect two different areas of the respiratory tract. JSRV targets the distal respiratory tract
(lungs), while ENTV affects the nasal turbinates. The mechanism behind this tissue specificity remains incompletely understood. Two areas of dissimilarity which may contribute to tropism include the long terminal repeats (LTRs), particularly the U3 promoter region, and the C-terminus of Env; the U3 differs by 37%, while the cytoplasmic tail (CT) of the C-terminus of Env differs by 50%16, 25, 54, 85, 86, 90, 144. These regions of dissimilarity make infection with ovine betaretroviruses JSRV and ENTV an excellent model to study viral tropism.
Currently, the consensus in the field is that ovine betaretrovirus tropism is dictated by differences in the U3 region85-88, given that ENTV or JSRV virions may be able to gain entry to a host cell and yet, be incapable of properly replicating due to a lack of transcription factors. The lung– and liver–specific hepatocyte nuclear factor (HNF)-3β
(HNF-3β) binding site is present in only the U3 of JSRV, whereas the CCAAT enhancer binding protein (C/EBP)-α and β binding sites are conserved in the U3 of both JSRV and ENTV isolates86, 87. HNF-3β interacts with JSRV LTR in alveolar type II
66
pneumocytes, the target cell for JSRV-induced lung tumorigenesis86, 88. However, in
2011, a study in mice found that there was no statistically significant difference in expression of the alkaline phosphatase reporter protein expressed under the control of either the JSRV or ENTV LTR when delivered using AAV vectors92. It would therefore be interesting to repeat this study in sheep, the natural host for JSRV and ENTV.
The Env protein is perhaps the most obvious component for spread of the virus since it interacts with the cellular receptor Hyal2 that facilitates viral entry49, 54, 60, 98, 105, 147, 167.
Hyal2 belongs to a family of glycosylphosphatidylinositol–anchored cell–surface proteins, responsible for degrading hyaluronic acid (HA), a major component of the extracellular matrix (ECM), and enriched in epithelial, endothelial and chondrocyte cells59, 96, 106-108. Hyal2 is enriched in the fetal lung, which is preferentially targeted by
Jenv115, 168. Notably, Hyal2 plays significant roles during clotting and fertilization. Hyal2 helps break down HA in platelets to trigger inflammatory cytokine responses and facilitates fertilization by helping the sperm break through the ECM to achieve fusion with the oocyte109. Despite these roles, it has also been demonstrated that Hyal2 has a lower hyaluronidase activity than other members of its family, suggesting that its ubiquitous expression may be differentially regulated depending on the cell type2, 49, 58,
59, 96, 98, 105, 107-109, 167. Interestingly, only the rat, ovine and human Hyal2 receptors interact with Jenv and Eenv proteins, with the rat Hyal2 having a lower affinity for Eenv than for Jenv; the murine Hyal2 does not permit entry of JSRV nor of ENTV into murine cells48, 58, 105, 147, 167. Furthermore, the presence of Hyal2 at endogenous levels does not seem to be sufficient for Eenv binding, unlike Jenv; overexpression of Hyal2 boosts
Eenv-pseudotyped retroviral transduction58.
67
The Jenv and Eenv cytoplasmic tail (CT) is the region with the lowest amino acid identity between JSRV and ENTV16, 20, 35. The CT of retroviruses is important for tumorigenesis, viral assembly and replication as well as cellular membrane fusogenicity51-53, 111, 169, 170. CT-truncation of Jenv and Eenv results in increased transduction efficiency of lentivectors pseudotyped with Jenv and Eenv CT-truncation mutants52, 53, 111; however, this enhancement is much more pronounced in the case of
Eenv53. The differences between Jenv and Eenv are far from discrete and the mechanism behind tissue tropism is possibly more complex than initially expected. In fact, the tissue tropism of other betaretroviruses, including mouse mammary tumor virus and Mason-Pfizer monkey virus is strongly influenced by both their envelope glycoprotein and LTRs23, 57, 84, 171-174.
In this study, we used chimeric viruses and lentivectors pseudotyped with Eenv and
Jenv, as well as ex vivo ovine lung and nasal turbinate tissue slices and found that
JSRV and ENTV tissue tropism is orchestrated by the envelope and U3 viral elements.
Furthermore, we found evidence that points to nasal turbinate chondrocytes as being a potential target for ENTV infection. Finally, we show that the JSRV LTR is not as active as the ENTV LTR in ovine primary chondrocytes.
68
Results
Jenv is important for mediating ovine lung transduction of lentivectors and chimeric viruses
To ascertain the genetic region responsible for tissue specificity, six JSRV-ENTV-1 hybrid viruses were constructed within the context of a JSRV molecular clone using overlap extension PCR: Chm1 (ENTV-1 R+U5), Chm2 (ENTV-1 gag), Chm3 (ENTV-1 env), Chm4 (ENTV-1 U3), Chm5 (ENTV-1 full LTR), and Chm6 (ENTV-1 env and U3)
(Figure 15A). Electron microscopic analysis of supernatant from HEK 293 cells transfected with the various chimeric molecular clones (Figure 15B) showed evidence of retrovirus like particles, in some cases with eccentric cores and an average size of 100-
150 nm, within the range of betaretroviral particles175. Western blot analysis of viral and cellular lysates from cells transfected with the chimeric molecular clones also demonstrated production of mature virus particles as evidenced by the presence of Env and fully processed capsid protein (Figure 15C). Western blot on cell lysates and purified and ultra-centrifuged and lysed virus supernatant was performed using an anti-
SU antibody to detect Env93 and anti-p27 antibody116 to detect the JSRV and ENTV capsids. Imaging after blotting with anti-SU showed the presence of bands around 63 kDa, the size of Jenv and Eenv16, 93, 116, 146, indicating a strong specificity for the Env of
JSRV and ENTV. We also observed a difference in size between Eenv and Jenv in the
Western blot of lysates of cells transfected with JSRV, chimeras encoding Jenv
(chimeras 1-2, and 4-5), ENTV, and chimeras encoding Eenv (chimeras 3 and 6) 48 h post-transfection; however, no difference was observed in viral lysates of supernatants collected, purified and ultra-centrifuged 48 h post-transfection. Similarly, bands were
69
also observed between 22 and 29 kDa, after anti-p27 blotting, showing specificity of the anti-p27 for the 27 kDa capsid protein of ENTV and JSRV. Bands indicative of p27 were seen for all viruses used in both cell and viral lysates, although weakly for chimera 1
(viral lysate) and chimera 2 (cell lysate).
To address the role of the envelope protein in determining tropism, we produced lentiviral vectors (LV) expressing GFP and pseudotyped with Jenv wt (Jenv GFP) and
Eenv wt (Eenv GFP). In addition, LV pseudotyped with VSVg (VSVg GFP) was used as a positive control for widespread tropism176-184, with baculovirus GP64, as a positive control for tissue specificity to the nasal tract113, 142, 185, and with EBOV (EBOV GFP), as a positive control for tissue specificity to the lungs133, 134, 178 (Figure 16A). Western blot analysis of HEK 293T producer cells confirmed expression of each of the envelope pseudotypes as well as the GFP reporter protein (Figure 16B). Western blots were loaded with equal protein amounts verified via Bradford assays, and membranes were cut post-transfer to blot with appropriate antibodies. Envelope expression of the different pseudotypes was observed 48 h post-transfection in transfected cell lysates using antibodies targeting the VSVg C-terminus, the baculovirus GP64 entry glycoprotein, the
EBOV envelope glycoprotein133, and anti-SU (Jenv and Eenv) antibodies (see methods for details on antibodies used). All the sizes of the envelope proteins corresponded to their molecular weights and none of the untransfected cell lysates showed the presence of bands, demonstrating the specificity of the antibodies to each envelope protein and
GFP. GFP expression was variable amongst transfected cell lysates of GFP encoded lentivectors pseudotyped by the envelopes described above, with VSVg GFP LV
70
expressing the most GFP, followed by GP64, then Jenv, EBOV and finally Eenv pseudotyped GFP LV.
We first focused on investigating the role of Env in mediating tissue tropism by utilizing the previously mentioned GFP-expressing lentivectors. This allowed us to visualize transduction via the GFP reporter and to circumvent the inability of JSRV and ENTV to efficiently replicate in cellular monolayers11, 112, 145. Since all lentivectors are produced using the same GFP-expressing vector (Figure 16A), all lentivectors express GFP from the same promoter in transduced target cells, allowing us to measure transduction efficiency based on transduced GFP DNA expression. Initially we wanted to investigate the transduction efficiency of Jenv and Eenv pseudotyped LV in vitro in different cell lines from different tissues, including GFP-expressing lentivectors pseudotyped with
EBOV and GP64 glycoproteins as controls. For this purpose, we employed human HEK
293T and ovine OA3.Ts fetal lamb testis cells186, as well as murine lung MLE-12 cells transfected with a FLAG-tagged human Hyal2 (FT hHyal2) (Figure 17A). MLE-12 cells were transfected with FT hHyal2 because the murine Hyal2 receptor is not permissive to Jenv or Eenv entry, while the human Hyal2 is98, 147, 167. Expression of hHyal2 was observed only at 60 kDa in Western blots of cell lysates of MLE-12 cells transfected with
FT hHyal2 48 h post-transfection, demonstrating the specificity of the anti-FLAG antibody to the FLAG tag present in the N-terminus of FT hHyal2. However, MLE-12 cells transfected with FT hHyal2 and transduced with Jenv pseudotyped GFP- expressing LV did not show uniform transduction efficiency as observed by the sparsely scattered green fluorescent cells compared to FT hHyal2 untransfected cells transduced with EBOV pseudotyped GFP LV, indicating a non-uniform transfection of FT hHyal2 in
71
MLE-12 cells (Figure 17B). While Jenv and EBOV pseudotyped LV transduced all three cell lines at similar high levels, GP64 and Eenv pseudotyped LV consistently transduced cells at a much lower level, as evidenced by the percentage of GFP-expressing cells imaged by flow cytometry (Figure 17B-C). We found no statistically significant difference in the level of transduction by Eenv pseudotyped LV in HEK 293T and MLE-12 cells
(p=0.995), as measured by GFP-expressing transduced cells. We also found no statistically difference between transduction by GP64 pseudotyped LV and Eenv pseudotyped LV within all cell lines (HEK 293T and MLE-12 each p>0.9999 and OA3.Ts p = 0.6539). Interestingly, Eenv LV showed the highest transduction in ovine OA3.Ts testis cells (Figure 17A), which was found to be statistically significant (HEK 293T vs
OA3T.s p=0.012; MLE-12 vs OA3.Ts p=0.0273).
To assess the role of Env in mediating transduction of tissues ex vivo, we then investigated the transduction efficiency of VSVg, GP64, Jenv and Eenv pseudotyped lentivectors in ovine lung tissue slices shown in the bright field (BF) images (Figure
18A). We did not transduce these lung tissue slices with EBOV LV because we had previously found that LVs pseudotyped with the EBOV glycoprotein do not transduce ovine lung tissue well (Chapter 2). Green punctate foci, representing cells transduced with GFP-encoded LV (FITC and FITC-BF channels), showed that Eenv pseudotyped
LV did not transduce ovine lung cells since no GFP expression was observed (Figure
18A). In contrast, VSVg and Jenv pseudotyped LV were capable of transducing ovine lung tissue slices as evidenced by the presence of green puncta (Figure 18A). Having examined the role of the Env-mediated transduction via the use of GFP-expressing lentivectors, we then decided to infect lung tissue slices with JSRV, ENTV and the
72
various chimeras we constructed to address the role of individual genomic regions in mediating infection. Four weeks post-infection, we used immunohistochemistry with an anti-SU antibody to detect for infection of ovine lung tissue slices with JSRV, ENTV or the chimeras. Brown staining representative of Env was observed only in ovine lung tissue slices infected with JSRV and other chimeras expressing the envelope of JSRV
(chimeras 1, 2, 4 and 5) (Figure 18B) but not in ENTV or chimeras encoding Eenv
(chimeras 3 and 6).
ENTV but not JSRV is capable of infecting ovine nasal turbinate tissue slices
Based on the results we observed in vitro and with the ovine lung tissue slices, we decided to investigate infection of ovine nasal turbinate tissue slices with JSRV, ENTV and chimeras 2, 3 and 4. We decided to first focus only on chimeras 3 and 4 to separately look at the role of Eenv (chimera 3) and ENTV U3 (chimera 4) in the nasal tract, given that the Env and U3 share the lowest nt and aa identity, respectively16; we used chimera 2 (ENTV gag), which shares >90% aa identity with JSRV gag16, 116, 187, as a negative control for nasal turbinate tissue infection. After infecting nasal turbinate slices with JSRV, ENTV and these chimeras, we collected the supernatant of the nasal turbinate slices 3 weeks post-infection, centrifuged and purified it with a syringe filter to remove cell debris, and lysed it prior to measure reverse transcriptase (RT) activity. As expected, supernatants of JSRV and chimera 2-treated ovine nasal turbinate slices had statistically significantly lower RT activity than the supernatant of ovine nasal turbinate slices infected with ENTV, and chimeras 3 (ENTV env), and 4 (ENTV U3) (Figure 19A).
Moreover, IHC staining for the SU of Env in the fixed nasal slices demonstrated strong expression of Env in ovine nasal turbinate slices infected with ENTV and chimera 3,
73
while no obvious staining was observed in ovine nasal slices treated with JSRV (Figure
19B). Even more interestingly, this figure suggests that cartilage tissue is infected by
ENTV and chimera 3 as evidenced by the brown staining in chondrocytes. Finally, using
IHC to detect the Env SU, we were able to observe some chondrocyte staining in a case of naturally occurring enzootic nasal adenocarcinoma (ENA) (Figure 19C). Most of this staining was observed in the extracellular matrix of chondrocytes which is known to be enriched with HA and also serves as a reservoir for HA degradation enzymes such as Hyal2110, 188, 189.
Eenv-facilitated lentivector entry is enhanced in chondrocytes and cells overexpressing hHyal2
Given our results in ex vivo ovine nasal turbinates, we were interested to determine if overexpression of Hyal2 would facilitate Eenv-LV entry. We employed HEK 293T cells since they endogenously express Hyal258, 98, 105, 111, 147, and transfected them in duplicate with a construct encoding FT hHyal2 to generate a cell line overexpressing hHyal2. We also simultaneously transfected the FT hHyal2 transfected cells and non- transfected cells with a construct encoding Luciferase, as a measure of transfection efficiency190-192. Forty-eight hours post-transfection, one set of untransfected or transfected cells with FT hHyal2 were transduced with GFP-expressing LV pseudotyped with EBOV (EBOV GFP), Jenv (Jenv GFP) and Eenv (Eenv GFP), and the others were used to run a Western blot. Western blot results showed expression of FT hHyal2 at varying levels, among transfected cells using anti-FLAG antibody as mentioned previously (Figure 20A). When compared to the loading control actin, FLAG seemed to be overexpressed in cells also transduced with Jenv and Eenv pseudotyped GFP LV.
74
We found that Eenv pseudotyped LV transduction was less efficient than Jenv in HEK
293T cells (black bars; Figure 20B) with no FT hHyal2, as we had previously observed
(Figure 17). However, in FT hHyal2-expressing 293Ts, Jenv LV transduction efficiency seemed similar to that of Eenv LV (light grey bars; Figure 20B). However, given the differences in transfection efficiency of FT hHyal2 observed from comparison with the loading control (Figure 20A), we decided to compare the flow cytometry GFP expression data from Eenv and Jenv LV transduced HEK 293T cells with the luciferase activity also measured in all cells. This resulted in ratio of GFP-expressing cells over luminescence activity output in these cells. We found that upon adjustment of transfection efficiency using this ratio, Eenv LV transduced FT hHyal2-expressed HEK
293T cells were just as efficiently transduced as those FT hHyal2 expressing 293T cells transduced with Jenv LV (Figure 20 C).
We also transduced primary ovine chondrocytes, representing a Hyal2-enriched cell line110, 188, 189, and similarly found that Eenv and Jenv transduction of these cells was not statistically different; however, we did observe a statistically significant drop in transduction efficiency in chondrocytes transduced with Jenv pseudotyped LV compared to Jenv LV transduced HEK 293T cells, regardless of FT hHyal2 expression
(p=0.0055 chondrocytes vs HEK 293T without FT hHyal2; p<0.0001 chondrocytes vs
HEK 293T with FT hHyal2). We also found that Eenv transduced these cells just as efficiently as it did Hyal2-overexpressing 293T cells (p=0.9999), and slightly higher than
Jenv LV did, although this was not found to be statistically significant (p=0.5622) (Figure
20B). To investigate whether we would see a similar increase in Eenv-LV transduction of murine tissues expressing FT hHyal2, we delivered AAV-FT-hHyal2 intranasally to
75
Balb/c mice and after 3 weeks, harvested lungs to generate murine lung slices stably expressing FT hHyal2. We first confirmed expression of FT hHyal2 in the lung and found that it was widespread and strongly expressed (Figure 20D) unlike previously seen in vitro in murine lung cells transfected with FT hHyal2 (Figure 17B). After transducing the FT-hHyal2-expressing murine lung tissue slices with Eenv, Jenv and
EBOV pseudotyped LV expressing human placental alkaline phosphatase (hPLAP), we observed dark staining indicative of exogenous AP activity, showing that Eenv LV expressing hPLAP was capable of transducing these slices similarly to Jenv LV expressing hPLAP (Figure 20E).
The ENTV LTR is active in ovine primary chondrocytes
Given that Jenv-LV was better at mediating entry into lung tissue than Eenv-LV, that
ENTV and ENTV Env or U3 encoding chimeras were capable of gaining entry and replicating in ovine nasal tissue slices at much higher levels than JSRV and chimera 2
(Figure 19), and that high Hyal2 expression facilitates entry via Eenv in vitro, we concluded that Env was important for entry. Having assessed the role of Env in tropism, we turned our attention to the LTRs of these viruses, which have been postulated to dictate tissue specificity28, 85-88, 90, 92. We transfected all primary ovine chondrocytes only with a GFP plasmid as a transfection control, or with a GFP plasmid and plasmids encoding hPLAP (as a control for AP activity) under the control of a CAG promoter, or under the control of the JSRV and ENTV LTRs (AJwtAP and AE1wtAP), or the full LTRs and enhancers (AJEnhancerAP and AE1enhancerAP)92. We measured GFP expression in cells via flow cytometry and AP expression by counting dark purple cells using the BF channel of a fluorescence inverted microscope. We then generated a ratio of these two
76
values to represent normalized AP expression. As measured by the level of normalized
AP expression in transfected cells with the aforementioned constructs, the JSRV LTRs and enhancer constructs were not significantly different from the AP control, whereas the wild type ENTV LTR and the ENTV LTR with the enhancer showed the highest reporter gene expression (Figure 20F).
77
Discussion
To understand which genomic region is responsible for the difference in tissue specificity between JSRV and ENTV, six chimeric viruses were generated using JSRV as a backbone and swapping JSRV genes or sequences for the equivalent sequences from ENTV-1. The chimeras were designed to contain sequences responsible for important aspects of the betaretroviral replication cycle, such as: replication and packaging (Chm1 – ENTV-1 R+U5), packaging and structural proteins (Chm2 – ENTV-1 gag), cell entry (Chm3 – ENTV-1 env), transcription (Chm4 – ENTV-1 U3), replication, packaging and transcription (Chm5 – ENTV-1 full LTR), and finally cell entry and transcription (Chm6 – ENTV-1 env and U3)1, 25, 35, 36, 69, 78. We also utilized lentivectors pseudotyped with Jenv and Eenv to better discern the role of the Env glycoprotein in tropism.
The use of ex vivo tissue slices has been remarkably useful in the past to study the pathogenesis of infectious pathogens112, 120, 125, 127, 128. Our use of ex vivo ovine lung and nasal turbinate tissue slices permitted us to bypass the inability of JSRV and ENTV to replicate in cellular monolayers. Furthermore, our study utilizes a novel murine lung model we previously developed that overexpresses a FLAG-tagged Hyal2 to test JSRV and ENTV tissue specificity (Rosales Gerpe et al. unpublished 2018). This is also the first study to show Eenv-pseudotyped lentivector transduction of ovine chondrocytes in vitro, and ENTV infection of ovine nasal turbinate chondrocytes ex vivo and in vivo.
Other studies have focused only on the ovine lung or have employed ovine epithelial cells11, 90, 103, 112, 113, 167. Our data demonstrate that both the Env and U3 are required for
JSRV and ENTV tissue selectivity, and that overexpression of Hyal2 might be crucial for
78
ENTV infection, which may initiate within nasal turbinate chondrocytes, where Hyal2 is enriched110, 188, 189.
The Eenv-pseudotyped lentivectors were lower than Jenv-pseudotyped lentivectors in transduction efficiency in HEK 293T cells and MLE-12 cells transfected with FT hHya2.
Other studies have previously demonstrated that overexpression of hHyal2 in different cell lines enhances transduction by Eenv pseudotyped lentivectors48, 58, 60, 98, 167, with the exception of rat cells58. However, though binding efficiency between Hyal2 and the Eenv surface glycoprotein was measured in the study by van Hoeven and Miller in 2005, the relative abundance or expression of Hyal2 was not measured or compared between cell lines58. In addition, our data showed that when Hyal2 was overexpressed in transfected
HEK 293T cells, Eenv pseudotyped lentivectors were able to transduce these cells to the same extent as Jenv pseudotyped LV. Furthermore, there was no statistically significant difference in transduction efficacy between Eenv and Jenv pseudotyped LV transduction of ovine chondrocyte and testes cells. Studies have shown that JSRV and
ENTV have invaded the germline of sheep in the past, leading to the presence of numerous endogenous betaretroviruses within the ovine genome2, 15, 29, 44, 69, 74, 75, 87, 115,
144. While most of this research has focused on JSRV and the ovine female reproductive tract2, 29, 44, 75, 115, it would be interesting to determine whether ram testes harbor endogenous ENTV-like sequences since HA metabolism is important in the testes and
Hyal2 might be highly expressed in this tissue109.
We also observed uniform overexpression of human Hyal2 in murine lung tissue slices post-delivery of AAV-FT-hHyal2, which allowed similar transduction levels of AP expressing lentivectors pseudotyped by Eenv and Jenv. Experiments with lentivectors
79
allowed us to conclude that high levels of Hyal2 are crucial for cellular entry via Eenv.
Moreover, newborn ovine lung tissue slices were neither transduced by Eenv- pseudotyped lentivectors, nor infected by either ENTV or Eenv-encoded chimeras.
Interestingly, other studies have shown that fetal lungs are enriched with Hyal2168.
However, this level might not meet the threshold level of Hyal2 needed for ENTV infection. Indeed, based on our results, in the case of ENTV, a low endogenous threshold of Hyal2 seems to act as the primary block to infection in the lung. It would be interesting to see if ENTV LTRs also play a significant role in restricting ENTV spread in the ovine lung. Future studies could look at whether sheep lungs overexpressing FT hHyal2 after transduction with an AAV vector, like we did in this study with murine lungs, would be permissive to ENTV infection.
We observed higher RT activity in the supernatant of nasal turbinate tissue slices treated with ENTV, ENTV Eenv-expressing chimera 3, and ENTV U3-encoded chimera
4 than in those nasal slices treated with JSRV and ENTV Gag-expressing chimera 2.
We also noted strong staining in nasal tissue infected with ENTV and chimera 3, compared to the faint staining for Env in the cartilage of JSRV-infected nasal turbinate tissue slices. Moreover, our data showed Env-stained chondrocytes in ENA tissue but not in normal tissue. Our results suggest that ovine nasal tissue is permissive to Jenv- mediated entry of JSRV; yet, JSRV did not seem to replicate in this tissue as efficiently as ENTV or chimera 3. It is important to point out that although chimera 3 encodes only the ENTV Eenv, the cDNA of Eenv extends largely into the U325, 35, 36, 92. It is possible that important ovine nasal tissue transcription factors could bind to this region permitting
ENTV and restricting JSRV viral transcription. Notably, our lab previously demonstrated
80
that the C-terminus of Eenv and Jenv contain enhancer elements92. However, we did not observe a significant difference in expression of hPLAP under the control of the
JSRV LTR or the JSRV LTR and enhancer in primary ovine chondrocyte cells. In contrast, ENTV LTR control of hPLAP resulted in higher hPLAP expression than hPLAP expression under the control of the JSRV LTR. The expression of hPLAP under the control of the ENTV LTR and enhancer was also significantly higher than under the control of the ENTV LTR alone.
Our results suggest that ENTV is capable of infecting chondrocytes in ovine nasal turbinates. However, we cannot yet speculate whether ENTV begins replicating in the nasal tract via chondrocytes, and whether cartilage tissue is initially transformed.
Studies have shown that most of the cartilage is destroyed in nasal turbinates burdened by ENA, but this has been explained by the invasion of ENA into neighboring tissue1, 14,
15, 19, 25, 36. Future in vivo studies using JSRV, ENTV, chimeras 3 and 5 in sheep would be crucial to conclude on this matter.
81
Materials and Methods
Cell Culture
Human Epithelial Kidney (HEK) 293, HEK 293T, and ovine epithelial OA3.Ts cells were cultured in High Glucose Dulbecco’s Modified Eagle Medium (DMEM) (Fisher, Whitby,
Ontario, Canada) supplemented with 10% Cosmic Bovine Calf Serum (BCS) (Fisher,
Whitby, Ontario, Canada), and 2 mM L-Glutamine (Fisher, Whitby, Ontario, Canada),
100 U penicillin/mL and 100 µg/mL streptomycin (P/S) (Fisher, Whitby, Ontario,
Canada) in a humidified 5% CO2 incubator at 37ºC. Murine Lung Epithelial (MLE 12)
(ATCC; CRL-2110TM) cells were cultured in 50:50 DMEM:DMEF12, supplemented with
1 g/L insulin, 0.55 g/L transferrin, 0.67 mg/L sodium selenite (Thermo Fisher,
Mississauga, Ontario, Canada).
Animals and Tissue Slices Generation
All animal experiments were conducted in accordance with the Canadian Council on
Animal Care guidelines and approved by the animal care committee of the University of
Guelph. Six, four-week old female BALB/c-mice were purchased from Charles River
Laboratories (St Constant, QC), and housed in separate cages containing two and four mice each. They were given ad libitum 14% Protein Rodent Maintenance Diet (Teklad
Global, Indianapolis, Indiana, USA), and tap water. The mice were acclimated to the environment for a week prior to being treated with 1 x 1011 vector genomes (vg)/mL of
AAV6.2FF-FTHyal2 as described previously (Rosales Gerpe et al. unpublished 2018); two BALB/c were kept as negative controls. Two weeks post-delivery, mice were euthanized to procure lungs for generating precision-cut lung tissue slices, as described previously (Rosales Gerpe et al., unpublished 2018). Specific-Pathogen Free Cornell
82
Star neonate lambs were purchased from Ponsonby Laboratories (Guelph, Ontario,
Canada) and euthanized via intravenous injection of pentobarbitol (Euthansol, Merck
Animal Health, Quebec, Quebec, Canada) prior to post-mortem and procurement of lungs for precision-cut lung tissue slice generation as previously described in Chapter 2.
These slices were maintained for up to a month as well.
DNA constructs and PCR
The pSinCASI-hPLAP-WPRE construct was generated using the InFusion kit (Takara
Clontech, Mountain View, California, USA) according to manufacturer’s instructions.
Briefly, the EF1α promoter in pSIN-EF1α was preplaced by the CASI promoter using
AgeI and SpeI restriction sites and the following primers: forward (FWD) 5’- cggcaattgaaccggtggagttccgcgt-3’ and reverse (REV) 5’- gccccagcatactagtctgttcgtcacccaggacct-3’. The heat-stable human placental alkaline phosphatase (hPLAP) gene was directionally cloned using SpeI and EcoRI (Abcam,
Cambridge, Massachusetts, USA) restriction sites and the following primers: FWD 5’- gatccttcgaactagtatgctggggccctgcatgc-3’ and REV 5’ ctcaagcttcgaattccttcagggagcagtggccgtct-3’. Finally, for increased mRNA stability and translation efficiency193-195, the Woodchuck Post-Transcriptional Regulatory Element
(WPRE) was also directionally cloned with the following primers: FWD 5’- ctccctgaaggaattcaatcaacctctggattacaaaatttgtgaaagA-3’ and REV 5’- atccttcgaactagcgctgcggggaggcggc-3’. All primers were generated using the InFusion function in the SnapGene software (GSL Biotech LLC, Chicago, Illinois, USA). The Q5
High Fidelity 2X Master Mix (New England Biolabs (NEB), Oshawa, Ontario, Canada) was employed to generate the CASI, hPLAP and WPRE inserts, according to
83
manufacturer’s specifications, with a few exceptions. For PCR amplification of the CASI and WPRE fragments, betaine196, 197 and dimethyl sulfoxide197 were added to reduce potential secondary structure and the extension time was increased to 1 min. The PCR program used was: 1 cycle at 95˚C for 30 s, 30 cycles of 95˚C for 5 s, 60˚C for 15 s and
72˚C for 20 s, followed by a final cycle of 2 min at 72˚C. Following PCR, the products were subjected to a 1 h DpnI digestion at 37˚C to remove template plasmid DNA. The
NucleoSpin Gel and PCR Clean-up Extract II kit was used to purify the DNA after PCR amplification and gel electrophoresis as indicated by the manufacturer (Fisher
Scientific).
The hybrid molecular clones were generated via Overlap Extension PCR (OE PCR) following a published protocol198. Primers were first used to create amplicons termed
‘inserts’ for the chimeras using Q5 HF DNA polymerase (NEB, Whitby, Ontario,
Canada), that were then inserted via OE PCR using the non-strand displacement
Phusion Polymerase (NEB, Whitby, Ontario, Canada), both according to manufacturer specifications.
The first set of primers used to generate inserts was as follows:
Chimera 1: FWD 5’- gaataaacaagttatgtactttataaatatagcattgtaataaagcaaggtatcagccattcttggtctg -3, REV 5’- acggagcgtcctcgctaagaaaataagagagagaccgcagccagcacggacaaaag -3’;
Chimera 2: FWD 5’- aggaggagtagtaaggtatatagttgagagtataaatatggggcaaacacatagtcgtcaaTTG -3’, REV: 5’- acccagtttcccgaaaccgggggtaaaggattaccttgaacatctgttttagaccggcaatC-3’;
84
Chimera 3: Inserts were generated in sequential order. First set of primers were FWD
5’- catgtttgtgtttttccacagaatgccgaagcaccgcgctggatc -3’, and REV: 5’- gctcttaagacttcctgagggtggcgggaca -3’; The second set of primers was: FWD 5’- atgcaacgcatgacgctgagcgagcccacgagtgacatgtttgtgtttttccacagaatg-3’, and REV: 5’- agcagtgccaaaagcaaacatccgagccttaaga gctcttaagacttcctgagggtggcgggaca -3’;
Chimera 4: FWD 5’- gagatgcgggggacgacccgtgagggggacaacccgcggagggttaa -3’, REV:
5’- tgagaggatcagaccaaaatggctgatactctgctttattgtgcagcagtat-3’;
Chimera 5: Chimera 4 FWD and REV: 5’- gttgagaggatcagaccaagaatggctgataccttgctttattgtgcagcagtat-3’;
Chimera 6: A similar PCR was done as Chimera 3. Both chimera 3 FWD primers were used, but Chimera 4 REV was used for both reactions with each FWD primer.
Two different programs were used to generate inserts 1, 2, 4, and 5 (1 cycle of 30 s at
98˚C, 30 cycles of 5 s at 98˚C, 30 s at 62˚C, 60 s at 72˚C, and 1 cycle of 2 min at 72˚C) and inserts 3 and 6 (1 cycle of 30 s at 98˚C, 30 cycles of 5 s at 98˚C, 30 s at 65˚C, 60 s at 72˚C, and 1 cycle of 2 min at 72˚C). Two additional different programs were used to perform overlap extension PCR for chimeras 1, 2, 4 and 5 (1 cycle of 30 s at 98˚C, 20 cycles of 5 s at 98˚C, 30 s at 60˚C, 18 min at 72˚C, and 1 cycle of 10 min at 72˚C), and for chimeras 3 and 6 (1 cycle of 30 s at 98˚C, 18 cycles of 5 s at 98˚C, 30 s at 60˚C, 36 min at 72˚C, and 1 cycle of 10 min at 72˚C) with the addition of 5 µm betaine and 4%
DMSO. Finally, site-directed mutagenesis was performed on chimeras 3 and 6 to remove the original splice acceptor site for Jenv in the JSRV backbone using Phusion
HF DNA polymerase, the following program 1 cycle of 30 s at 98˚C, 25 cycles of 10 s at
85
98˚C, 30 s at 70˚C, 7 min at 72˚C, and 1 cycle of 2 min at 72˚C, and the following primers: FWD 5’-gtttgtgtttttccacaaaatgccgaagcgccg-3’, and REV: 5’- cggcgcttcggcattttgtggaaaaacacaaac-3’.
Virus Production, Transduction and Infection
Vector and virus particles were produced using 10 million HEK 293T cells (vectors) or
293 cells (viruses) seeded in 10 cm dishes. Twenty-four hours post-seeding, cells were transfected using reporter gene plasmids pSinGFP, pSinCASIhPLAPWPRE, packaging plasmid psPax2 and envelope plasmids (EBOV, GP64, Eenv and Jenv) to generate lentivectors or plasmids pCMVJS21 for JSRV12, pCMVENTV-1NA4 for ENTV16, 116, pCMVJS21-ENTVRU5 for Chm1, pCMVJS21-ENTVgag for Chm2, pCMVJS21-
ENTVenv for Chm3, pCMVJS21-ENTVU3 for Chm4, pCMVJS21-ENTVFullLTR for
Chm5, and pCMVJS21-ENTVenvU3 for Chm6 and 67.5 µL of 1 mg/mL polyethylenimine transfection polymer, and 450 µL of DMEM per dish. Post-production and PEG-8000 precipitation of viruses or vectors was performed as previously detailed, with a few exceptions199, 200. Two hours post-transfection, the media was changed with
DMEM supplemented with 2 mM L-glutamine, and P/S (basal DMEM). Every 24 hrs, for a total of 72 hrs, the supernatant was collected into 50-mL conical tubes and replenished with fresh basal DMEM. After collection, supernatant was distributed into five 50-mL conical, high-speed centrifuge tubes, and PEG-8000 (Sigma-Aldrich,
Oakville, Ontario, Canada) was added in a 1:2 PEG to supernatant ratio. The tubes were kept rocking gently overnight at 4˚C. The next day, the virus-PEG mixture was spun in a swinging-bucket centrifuge at 4000 g for 15 min. Each tube was then decanted into a waste bottle and set upside down on top of paper towels to remove any
86
excess media. Finally, the white pellet was resuspended in TNE buffer at 1/100th the initial supernatant volume (i.e. if 30 mL of supernatant were combined with PEG, then
300 µL of TNE would be used to resuspend the virus or vector pellet).
Lentivector titers were measured via transducing units (TU)/mL151, 201 calculated from
GFP-based flow cytometry or via p24 ELISA for hPLAP lentivectors. JSRV, ENTV and chimera titers were measured via the reverse transcriptase (RT) assay, as outlined in
Chapter 2. We normalized the percent of GFP positive cells representing transduction to those cells transduced with GFP-expressing lentivector pseudotyped with the EBOV133,
134, 178 envelope glycoprotein for Figures 17 and 20.
Post-production of particles, HEK 293T (for lentivectors) or HEK 293s (for JSRV, ENTV and chimeras) were lysed, or their supernatant was harvested for ultracentrifugation as described previously14, 53, 116, prior to lysing with RIPA buffer as mentioned in previous studies116 and processed for Western blot or EM. Protein expression was visualized via
WB and viral and vector particles were imaged using EM as detailed in Walsh et al.
2017116. To visualize VSVg, GP64 and EBOV the following monoclonal antibodies were used respectively: rabbit monoclonal antibody from Abcam (ab1974), mouse monoclonal antibody from Abcam (ab91214), and mouse monoclonal antibody 5E6, which was a gift from Gary Kobinger133.
Immunohistochemistry (IHC)
Lung lobes from PBS-treated control and AAV6.2FF-FT-Hyal2-transduced BALB/c mice were harvested for IHC as described previously146, 202. Briefly, Rabbit α-FLAG (Abcam, cat. #Ab21536) was used at 1:100 in 3% bovine serum albumin in 1X PBS-Tween- containing SignalStain® Boost IHC Detection Reagent (HRP Rabbit) (Cell Signaling,
87
NEB) as recommended by the manufacturer. To detect the Env of JSRV and ENTV,
1:50 mouse DM12 antibody and 1:100 Goat anti-mouse secondary antibody were used, as previously described93. SIGMAFASTTM 3,3’-Diaminobenzidine DAB (Sigma-Aldrich) was used for detection following the manufacturer’s instructions. The slides were counterstained with hematoxylin and mounted with Richard-Allan ScientificTM Cytoseal
XYL (Thermo Fisher Scientific), before being imaged with a Clinical Olympus BX45 microscope (Olympus, Tokyo, Japan).
Statistical Analyses
Two- and three-way ANOVAs, and student t-tests were conducted for Figures 17A (two- way ANOVA), 19A (unpaired t-tests), 20B (two- and three-way ANOVAs), 20C (paired t- tests), and 20F (unpaired t-tests) using GraphPad Prism 7 software (GraphPad, La
Jolla, California, USA).
88
Figure 15. Production of JSRV-ENTV chimeric viruses.
A) Schematic representation of JSRV-ENTV hybrid viruses with JSRV as a backbone. Regions from ENTV (in grey) were swapped for complementary regions in JSRV (white). B) Electron microscopy of chimeric viruses produced in HEK 293 cells. C) Western blot of cell and viral lysates of JSRV, ENTV and chimeric viruses produced in HEK 293 cells showing expression of the envelope and capsid proteins.
89
Figure 16. Production of lentivirus vectors pseudotyped with VSVg, EBOV, GP64, Jenv and Eenv envelope glycoproteins.
A) Schematic representation of lentiviral vector systems and VSVg, EBOV, GP64, Jenv and Eenv pseudotypes. Western blot of cell lysates of HEK 293T cells transfected with GFP-expressing LV pseudotyped with VSVg, EBOV, GP64, Jenv and Eenv, showing expression of envelope (B) and GFP (C).
90
Figure 17. Eenv pseudotyped lentivirus vector (LV) transduces cell lines and ovine lung slices at a lower efficiency than Jenv pseudotyped LV.
A) Western blot of lysates of MLE-12 cells untransfected (untransfected cells and no FT hHyal2) or transfected with FT hHyal2 (with FT hHyal2). B) MLE cells untransfected (no FT hHyal2) or transfected with FT hHyal2 (with FT hHyal2) and transduced with GP64, EBOV, Jenv and Eenv pseudotyped LV expressing GFP imaged in the FITC channel. BF = brightfield channel. C) Cell lines HEK 293T, OA3.Ts, and MLE-12-hHyal 2 (transfected with FLAG-tagged human Hyaluroglucosaminidase-2) were transduced with GP64, EBOV, Jenv and Eenv, and the percent of GFP positive cells was normalized to the EBOV sample per cell line. A Two-way ANOVA was conducted to perform multiple comparisons. Eenv pseudotyped LV of HEK 293T and MLE cells (p=0.995). Eenv LV in HEK 293Ts or MLEs vs OA3.Ts testis cells (HEK 293T vs OA3.Ts p=0.012; MLEs vs OA3.Ts p=0.0273). GP64 LV and Eenv LV comparison amongst all cell lines (HEK 293Ts and MLEs each p>0.9999 and OA3.Ts p = 0.6539).
91
Figure 18. Jenv is crucial for mediating viral entry into ovine lung tissue.
A) Ovine lung tissue slices transduced with GFP-expressing lentivectors pseudotyped with VSVg, GP64, Jenv and Eenv envelope glycoproteins. Jenv and VSVg pseudotyped LV transduction is visualized by the punctate green foci in the FITC channel. B) Immunohistochemical staining (IHC) of paraffin-embedded ovine lung tissue from a field sample of ovine pulmonary adenocarcinoma (OPA) or paraffin-embedded ovine lung tissue slices infected with JSRV, ENTV, and JSRV-ENTV chimeras. Positive staining for the envelope protein of JSRV and ENTV is indicated by the brown stain. The hematoxylin counter stain is purple.
92
Figure 19. Ovine nasal turbinate slices can be infected with ENTV but not with JSRV.
A) Reverse transcriptase (RT) activity from the supernatant of ovine nasal turbinate slices infected with JSRV, ENTV and chimeras 2 – 4. B) IHC staining of ovine nasal turbinate slices for the Env of JSRV and ENTV shows cartilage (chondrocytes) staining brown for the presence of Eenv in ENTV and Chimera 3 (Chm3(env)) infected tissue slices, but not in JSRV-infected tissue slices. C) IHC of nasal turbinate from a normal sheep and a sheep with enzootic nasal adenocarcinoma (ENA) showing staining of chondrocytes and epithelial cells for Env in the ENA sample. Unpaired student t-tests were performed for (A). JSRV vs ENTV: p=0.0002 (*); JSRV vs chimera 2: p=0.4031 (n/s); JSRV vs chimera 3: p=0.0001 (*); JSRV vs chimera 4: p=0.0009 (*).
93
Figure 20. Overexpression of Hyal2 facilitates Eenv-LV cell entry and the long terminal repeats from ENTV are highly active in primary ovine chondrocytes.
A) Western blot analysis of lysates from HEK 293T cells transfected with FLAG-tagged human Hyal2 (FT hHyal2) transduced with GFP-expressing lentivectors pseudotyped with EBOV, Jenv and Eenv. (B) Flow cytometry data showing the percent of cells expressing GFP from the cells described in (A). Values were normalized to those obtained with EBOV LV. A two-way ANOVA was performed for Jenv and Eenv samples of ovine primary chondrocyte cells (not significant p=0.5622) C) Data showing the ratio between flow cytometry values in (B) and luminescence values from HEK 293T cells co-transfected with FT hHyal2 and a plasmid expressing the luciferase gene and, transduced with the lentivectors described in (A), 48 h post-transfection. Values were normalized to those obtained with EBOV LV. A paired two-tailed student t- test was performed between Jenv GFP with FT hHyal2 and Eenv GFP with FT hHyal2 samples (n/s p=0.3181). A paired student t-test was also performed between Eenv GFP no FT hHyal2 and Eenv GFP with FT hHyal2 (* p<0.0001) D-E) Murine lung tissue slices expressing FT hHyal2 (D) transduced with human placental alkaline phosphatase (hPLAP) expressing lentivectors pseudotyped with EBOV, Jenv and Eenv (E). F) Ratio of the count of AP-positive foci and the percent of GFP-expressing cells in primary ovine chondrocytes co-transduced with pSinGFP and pCAGAP (hPLAP), or pCAGJwtAP (JwtAP), pCAGE1wtAP (E1wtAP), or pCAGJenhAP (JEnhancerAP), or pCAGE1enhAP (JE1EnhancerAP). A student unpaired two-tailed t-test was performed between AJwtAP and AE1wtAP (** p=0.0124) and between AE1wtAP and AE1enhancerAP (***p=0.0034).
94
Chapter 4: Optimized pre-clinical grade production of two novel lentiviral vector pseudotypes for lung gene delivery
María C. Rosales Gerpe 1, Laura P. van Lieshout1, Jakob M. Domm1, Jacob P. van Vloten1, Jodre Datu1, Joelle C. Ingrao1, Darrick L. Yu 1, Jondavid de Jong1, Byram W. Bridle1, and Sarah K. Wootton1*
1Department of Pathobiology, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada.
Running title: Production of novel lentiviral vectors for lung gene therapy
*Correspondence should be addressed to S.K.W.: Mailing address: Department of Pathobiology, Ontario Veterinary College, University of Guelph. Building 89, 28 College Avenue West, Guelph, Ontario, Canada, N1G 1R7. Phone: +1 (613) 824 4120, x54729. E-mail: [email protected]
Submitted to Molecular Therapy – Methods and Clinical Development.
95
Abstract
Lung gene therapy requires efficient transduction of slow-replicating epithelia and stable expression of delivered transgenes in the respiratory tract. Lentiviral vectors (LV) have the ideal coding, expression and transducing capacity required for gene therapy. A modified envelope glycoprotein, Jenv, derived from betaretrovirus Jaagsiekte Sheep
Retrovirus, is well-suited for LV-mediated lung gene therapy because of its lung-centric tropism and fetal-stage transducing efficiency. In this study, we generated two novel
Jenv-pseudotyped LVs that effectively transduce airway epithelia and yield titers similar to the gold standard, vesicular stomatitis virus glycoprotein (VSVg)-pseudotyped LVs.
As Jenv-pseudotyped LVs responded poorly to concentration via ultracentrifugation, we developed a large-scale production method tailored for these novel Jenv pseudotypes and determined the most appropriate method of LV concentration. We found that, in contrast to VSVg and Ebola virus glycoprotein-pseudotyped LVs, ultracentrifugation through a sucrose cushion drastically reduced the yield of Jenv LVs, whereas polyethylene glycol precipitation and tangential flow filtration (TFF) proved to be more suitable methods for concentrating Jenv LVs. We also found that the pressure in TFF was crucial for increasing LV recovery. Finally, our work showcases the value of an appropriate model when titrating and testing the efficacy of LVs for potential gene therapy.
96
Introduction
Retroviral and lentiviral vectors (LVs) currently make up almost a quarter of all gene therapy clinical trials worldwide203. They remain attractive tools for gene delivery due to their wide tissue tropism, their ability to modify the host genome, and their low immunogenicity204-207. However, retroviral vectors may cause oncogenic side effects208-
213 and are dependent on cellular replication before they can modify the genome, delaying the onset of transgene expression214-216. In contrast, lentiviruses are capable of infecting non-dividing cells216. This ability of lentiviruses to target quiescent cells is of special interest for gene therapy to the lung where the epithelia generally have low turnover rates (>100 days in mammals)217-219. LVs have also been modified to mitigate the risk of oncogenesis-linked genome integration220-222. Therefore, LVs offer a substantial advantage over retroviral vectors for genetic correction through cellular transduction of both dividing and non-dividing cells148, 193, 223-227.
These new developments have increased the demand for LVs in gene therapy trials; in
2009, they made up 1.4% of clinical trials and their use had tripled by 2016203, 204. Other viral vectors are also being used for gene therapy in the lung, such as AAV. However, the high immunogenic characteristic of adenovirus vectors and the low coding capacity of AAVs represent drawbacks in their potential use217, 228. In contrast, LVs have relatively low immunogenicity and relatively high coding capacity (>8 kb cloning capacity between the LTRs) 185, 206, 221, 229, 230. However, despite these advantages, LVs continue to lag behind adenovirus and AAV vectors for utilization within lung gene therapy trials.
This likely stems from a lack of an efficient envelope to promote efficient transduction of lungs in vivo.
97
Vesicular stomatitis virus glycoprotein (VSVg)-pseudotyped LVs efficiently transduce a wide range of cells and tissues in vitro and in vivo177, 184, 231, 232. However, studies have shown that VSVg cannot readily access the lung epithelia due to the lack of cognate basolateral low density lipoprotein (LDL) receptors available to VSVg181, 233-237.To circumvent this, chelating and surfactant agents are continuously being developed to disrupt the tight junctions in the lung and facilitate delivery176, 234, 238-240. Though this idea seems suitable for the advanced stages of diseases such as cystic fibrosis (CF)206, 241, the use of chelators might not be appropriate for CF because cumulative disruption of tight junctions could lead to inflammation enhancing the pathology of CF206, 242. Efforts to find an alternative envelope to VSVg continue to introduce different LV pseudotypes133, 134, 142, 235, 236, 243, 244.
Recently, the envelope glycoprotein of the betaretrovirus Jaagsiekte Sheep Retrovirus
(JSRV) was identified as a promising LV pseudotype 113, 245. JSRV has restricted tropism to the respiratory tract and targets the lungs at an early age in ruminants113. As such, the JSRV envelope glycoprotein (Jenv) demonstrates great potential as an envelope to pseudotype LV for use in lung gene therapy. In 2005, Sinn and colleagues found that flanking Jenv with promoter regions from JSRV increased expression of the envelope and production of LV pseudotyped by Jenv113, 245; they termed their new Jenv construct delta gag pol Jenv (ΔGP Jenv). Furthermore, in 2012, a study showed that the
Jenv-pseudotyped LV was indeed capable of efficiently transducing the fetal ovine airway, unlike Ebola virus (EBOV) glycoprotein pseudotyped LV, baculovirus GP64 LV and AAV2/6.2113. In an effort to further increase Jenv LV titers, we produced and concentrated two novel ΔGP Jenv LV pseudotypes by individually mutating residues
98
N592 or M593, previously found to enhance or abrogate Jenv’s transformation activity, respectively246, which we hypothesized contributed to reduced LV titers. We found that both ΔGP Jenv N592T and ΔGP Jenv M593E LVs can yield in vivo-grade titers (109 transducing units (TU)/mL)247 similar to VSVg LV. Finally, no pre-clinical grade production has been described for the Jenv LV pseudotype12, 33, 113, 114, 245, 248. Here, we describe a systematic optimization of this purification process and compare methods for use in small and large-scale applications.
99
Results
LVs pseudotyped with envelope glycoproteins from VSV, EBOV and JSRV differ in their ability to be concentrated by ultracentrifugation and polyethylene glycol
(PEG) precipitation
It is well known that LVs pseudotyped with envelope glycoproteins from disparate viruses differ with respect to their stability during purification. Initially we compared two standard methods of concentrating LVs to determine if they would work for ΔGP Jenv pseudotyped LVs expressing GFP. The methods were ultracentrifugation through a
20% sucrose cushion or PEG precipitation. For comparison, we included VSVg- pseudotyped LV, as it is the gold standard, and EBOV-pseudotyped LV as another lung tropic vector133, 134. We measured vector yields using a p24 ELISA and flow cytometry in input vector (neat) and after different concentration methods (ultracentrifugation or ultra,
PEG-8000 or PEG and tangential flow filtration or TFF). Using the p24 ELISA, we were able to quantify capsid protein in pg/mL and calculated infectious units (IFU) based on p24 levels and the number of p24 molecules normally found per virion. We measured transduction efficiency by measuring the percent of GFP expressing cells and the volume of vector used to transduce cells. We found that for VSVg LV, known to have high recovery yields post-ultra-centrifugation114, 179, both ultracentrifugation and PEG-
8000 precipitation yielded vector titers that were suitable for delivery in vivo if administered at a dose of 1x108 TU, with ultracentrifugation yielding the highest titer
(Figure 21A to D and Table 3). In the case of EBOV LV, PEG-8000 precipitation and ultracentrifugation total transducing units were very close to the total transducing units found in the neat vector, resulting in recovery levels that were nearly 100%; however,
100
due to the low titer of the input vector, the titers achieved were not amenable for in vivo delivery. Finally, LVs pseudotyped with ΔGP Jenv were stable during PEG precipitation, yielding 100% recovery, but were destroyed during ultracentrifugation such that only ~30% of LV was recovered.
With regard to titrating LVs, we found that titers obtained using the p24 Enzyme linked immunosorbent assay (ELISA) correlated well with the titers obtained by flow cytometric analysis of transducing units per mL (Figure 21B-D). To convert p24 ELISA values to transducing units, the manufacturer provides two concentration values, 100-fold to
1000-fold Infectious units (IFU)/mL, to account for the number of LV particles that it would take to reach a multiplicity of infection (MOI) of 1. We found that the 100-fold
IFU/mL estimation of titer was most similar to the TU/mL titer determined by flow cytometry (Figures 21B to D). Therefore, conversion of p24 ELISA data to IFU/mL using the 100-fold estimation can be used in place of TU/mL in the case where an LV does not express a reporter gene.
Pressure use in tangential flow filtration generates high yields in lentivector production for VSVg and ΔGP Jenv LV pseudotypes
Large-scale LV concentration is often needed to obtain titers that are high enough for use in vivo. Processing large volumes via ultracentrifugation or high-speed centrifugation, in the case of PEG precipitation is cumbersome and time-consuming, as typical research centrifuges are limited to ~180 mL of LV-containing supernatant per rotor, and often requires multiple spins, which may lead to diminished viral particle recovery. Tangential flow filtration (TFF), on the other hand, is a simple method used to concentrate liter-scale volumes of LV-containing supernatant201, 247, 249. TFF does not
101
appreciably trap LV and allows for buffer exchange and easier downstream processing as volumes are considerably reduced. Therefore, we explored the use of TFF as an alternative to centrifugation for concentrating large volumes of LVs. The VSVg- pseudotyped LV was used in the optimization process due to its relatively stable nature.
Initial TFF experiments to concentrate VSVg LV involved the use of 5 psi of differential pressure as this has been reported in the literature201, 249. Thus, the inlet pressure was set to 5 psi with no pressure on the outlet, which can be modulated with the use of a clamp restricting flow from cassette to the reservoir; this controls how much vector is circulated (Figure 22). This condition was designated as Low Pressure TFF. Throughout the process (Figure 22), we set aside volumes of neat vector (to measure input vector titers), retentate elutions, and filtrate waste to measure transduction efficiency of vector contained within each fraction using TU/mL and a p24 ELISA as detailed in the Methods section (below). Three elutions were collected from the retentate side of the cassette: the first elution (E1) from the remaining volume (8-10 mL) after concentration, and the second (E2) and third elution (E3) (both 8-10 mL) procured after adding TNE buffer to collect remaining vector in the TFF cassettes; a 5 mL aliquot was collected from the waste flowing from the filtrate end of the cassette.
The literature reports near 100% recovery yields with conditions using 5 psi of differential pressure in TFF201, 247, 249. Despite the high TFF yields reported in the literature, we were unable to initially reproduce those results using a 100K molecular weight cut off (MWCO) cassette and 5 and 0 psi of inlet and outlet pressure, respectively (Table 4: Top 100K). We then explored whether this might be due to differences in the MWCO of the membrane used. Therefore, we tested three MWCO
102
sizes: 50K, 100K, and 300K using 5 psi inlet pressure and 0 psi outlet pressure (Figure
23A and Table 4: Top). Nevertheless, we found that with 5 psi inlet pressure and no outlet pressure, TFF purification of VSVg LV with all three membranes still resulted in low titers (104 to 105 TU/mL for all membranes) and very poor recovery (2.5%, 21.0%, and 10.6%, respectively) (Figure 23A and Table 4: Top). We also increased the inlet pressure up to 15 psi again with no outlet pressure and this did not improve yields (data not shown).
Next, we decided to increase the outlet pressure to 10 psi while maintaining a differential pressure of 5 psi, which we designated as High Pressure TFF. In this case, using a 15 psi inlet pressure and a 10 psi outlet pressure by clamping the retentate tubing (Figure 22) resulted in yields increasing by more than 50% when compared to yields obtained when maintaining the outlet pressure at 0, and the inlet pressure at 5 psi
(Table 4: High pressure TFF). Furthermore, with these conditions, the 50K membrane showed a higher titer than with low pressure conditions, and the highest recovery yields of all MWCO membranes. The100K and 300K MWCO membranes output higher titers than the 50K MWCO, though TFF purification with the former membranes also began with higher input vector titers. The 100K and 300K MWCO membranes had lower recovery yields than the 50K MWCO.
We tested transduction efficiency of vectors in the volumes collected by adding them directly to HEK 293T cells at 1:1 DMEM: TFF process collection (0.5 dilution factor or df) using polybrene as described in the methods. Unfortunately, high dilution TFF preparations (df = 0.5) were toxic to cells in vitro, as the addition of the elutions to cells resulted in cell death (Figure 24). To resolve this issue and to further increase titers, we
103
implemented a post-TFF purification sucrose gradient ultracentrifugation dialysis step
(Figure 25) using as a proof of principle the recovered vector from the 50K MWCO membrane. Using these conditions, the vector titer was increased to levels appropriate for in vivo work (108 TU/mL)113, 133, 134, 178, 180, 245, 248, 250 without compromising yield
(99.6% for VSVg LV; Table 4, bottom row). Finally, we did not find any vector in the waste fraction using any of the aforementioned conditions (Figure 23). We thus determined that the optimal conditions for generating LV required the high-pressure conditions described above (15 psi inlet and 10 psi outlet), using a 50K MWCO membrane for best recovery yield. Having optimized the TFF concentration method for
VSVg LV, we then tested it with ΔGP Jenv LV. In contrast to ultracentrifugation, TFF using a 50K membrane in combination with high pressure resulted in a high percentage of ΔGP Jenv LV-pseudotyped vector recovery (94.1%; Table 5), similar to what was observed for PEG. However, despite the fact that we had developed a scalable TFF method to concentrate large volumes of ΔGP Jenv LV without having to use multiple rounds of centrifugation, the fact remained that the starting titers were likely too low to be scaled up for use in vivo.
Introducing point mutations into the carboxy terminal domain of the JSRV Env protein increases LV titers
Production of ΔGP Jenv-pseudotyped lentivector can be toxic to HEK 293T producer cells, which are already highly transformed by the SV40 T-antigen245. The possibility that cytotoxicity or poor expression levels of Jenv were impeding vector yields prompted us to introduce mutations in the cytoplasmic tail (CT) of ΔGP Jenv, where the
36 Y590XXM593 transformation motif is found , to address three possibilities. Previously,
104
mutation of methionine at position 593 to glutamic acid (M593E) was shown to severely abrogate transformation147, whereas an N592T led to increased Jenv expression. To test whether these mutations would increase Env expression, we transfected cells with
GFP-expressing lentivectors pseudotyped by Jenv wt, ΔGP Jenv wt, ΔGP Jenv N592T and ΔGP Jenv M593E, and harvested and lysed cells 48 h post-transfection. Western blot analysis using anti-SU antibody showed strong bands around 63 kDa, absent from the untransfected control, showing the specificity of the antibody. Expression levels of
Jenv in HEK 293T producer cells revealed that both the M593E and N592T mutant versions of ΔGP Jenv were expressed at higher levels than wild-type ΔGP Jenv compared to the GFP transfection control and the actin housekeeping loading control
(Figure 26A). These results translated to higher input vector titers (data not shown). We then proceeded to attempt the TFF conditions optimized above in an effort to see whether similar titers to VSVg LV could be attained. Use of TFF and downstream purification of lentivectors pseudotyped with ΔGP Jenv mutants N592T and M593E yielded average titers of 3 x 109 and 2 x 109 TU/mL, respectively, which were 10-fold higher than ΔGP Jenv (2 x 108 TU/mL) when purified in the same manner (Figure 26B), and comparable to VSVg LV titers (6.8 x 109 TU/mL) concentrated via ultracentrifugation (Figure 26B). Taken together, introducing mutations into the JSRV
Env CT to enhance its expression led to increased LV production, with similar titers to those of VSVg-pseudotyped LV.
105
Development of a mouse model to evaluate JSRV Env-pseudotyped LV-mediated transduction
Next, we wanted to test VSVg, EBOV, ΔGP Jenv, and M593E and N592T ΔGP Jenv
LVs in tissue explants to evaluate transduction efficiencies in both murine and ovine lung tissues. Although ovine and human hyaluronoglucosaminidase-2 (hHyal2) can function as a receptor to mediate ΔGP Jenv LV infection, the murine Hyal2 homologue does not48, 49, 98, 167. Therefore, in order to test the Jenv-pseudoypted LV vectors in mice, we developed a novel murine lung slice model by delivering a FLAG-tagged version of hHyal2 (FT-hHyal2) to mouse lungs using the recently characterized AAV6.2FF capsid251 (Figure 27A). Briefly, 1x1011 viral genomes (vg) of AAV6.2FF-FThHyal2 were administered to 6-week-old BALB/c mice intranasally as described251. Three weeks later, lungs were harvested and used to generate lung tissue slices as previously described (Chapter 2) and tested via immunohistochemistry for FT hHyal2 expression.
Using an anti-FLAG antibody, lungs stained dark for the presence of the FLAG tag present in FT hHyal2, indicating that there was widespread and strong expression of FT hHyal2 (Figure 27B). These widespread expression of a permissive receptor to Jenv allowed us to test Jenv pseudotyped LVs. Having developed a model to evaluate Jenv- pseudotyped LV transduction in murine lung tissue, we prepared LV vectors expressing the human placental alkaline phosphatase (hPLAP) reporter gene and pseudotyped with the VSVg, ΔGP Jenv, ΔGP Jenv-N592T, ΔGP Jenv-M593E or EBOV envelopes.
Tissue slices prepared from the lungs of mice previously transduced with AAV6.2FF-FT- hHyal2 permitted transduction with ΔGP Jenv LV-pseudotyped vectors, as evidenced by the purple staining indicative of AP activity. Mutant ΔGP Jenv LV pseudotypes (N592T
106
and M593E) were also capable of proficiently transducing mouse lungs with greater efficiency than ΔGP Jenv, VSVg and EBOV LV-pseudotyped vectors, as shown by the higher incidence of dark purple staining observed in Mutant ΔGP Jenv LV pseudotypes
(N592T and M593E) transduced murine lungs expressing FThHyal2 (Figure 27C).
We also examined the various LV pseudotypes in ovine lung tissue slices. The greatest expression of GFP was observed in ovine lung tissue slices transduced with ΔGP Jenv
GFP (Figure 27). VSVg LV was also capable of efficiently transducing ovine lung tissue slices to a lesser extent than ΔGP Jenv GFP (Figure 27). Interestingly, although the
EBOV pseudotype was able to transduce murine lung slices (Figure 26C), ovine lung slices were not efficiently transduced by EBOV LV (Figure 28).
107
Discussion
We generated two novel lentiviral pseudotypes that can reach in vivo-grade titers equivalent to VSVg LV and optimized their large-scale production for lung gene therapy.
We also developed a novel murine lung tissue slice model that allowed us to test these
LV pseudotypes ex vivo. We showed that TFF, followed by gradient ultracentrifugation, dialysis and PEG-20,000 concentration, is a scalable and suitable method for producing
ΔGP Jenv LV that does not compromise recovery. Furthermore, we demonstrated that murine and ovine lung tissue slices are appropriate models to test the transduction efficiency of these and other lentivectors for lung gene therapy.
We first evaluated three different methods of concentrating LV: ultracentrifugation, PEG precipitation and TFF. Ultracentrifugation was the most suitable method for VSVg and
EBOV LVs, as previously described114, 179, 183, 200, 252-254. However, this was not the case for ΔGP Jenv-pseudotyped LV. Ultracentrifugation involves great shearing forces255 and not all viral envelopes are able to sustain them179, 200, 252, 253. In fact, a study found that
LV particle stability in ultracentrifugation depends on the particle core stability and the envelope protein used to pseudotype the LV179. Studies have evaluated the resilience of the VSVg glycoprotein and found it to be resistant to the shear forces of ultracentrifugation179, 182, 253. Ultracentrifugation of EBOV LV has also been shown to yield high titers and recovery yield134, which we confirmed. Therefore, it is likely that LV particle stability is maintained in the presence of the EBOV glycoprotein. Conversely,
JSRV envelope-pseudotyped LV has a low recovery yield when ultra-centrifuged, compared to LVs pseudotyped by VSVg and EBOV114, 133, 134, 256, thus we do not
108
recommend the use of ultracentrifugation to concentrate the ΔGP Jenv-pseudotyped
LV.
Unlike ultracentrifugation and PEG-8000 precipitation, which are acceptable concentration methods for pre-clinical testing in mice, TFF is a promising technique for concentrating large volumes of LVs and is better suited for large animal gene therapy and clinical applications due to its scalability. The current literature on the use of TFF for
LV production for gene therapy purposes has predominantly utilized the 100K molecular weight cut-off membrane, though a recent study also compared 100K and 300K membranes247. Furthermore, to our knowledge, another aspect of TFF that has not been well defined is the importance of pressure during TFF purification of LV. The pressure values reported in the literature tend to reflect the differential pressure and do not clearly specify whether or not the outlet pressure is altered201, 249. We found that adding a clamp to increase the outlet pressure was critical for increasing recovery yields for VSVg and for ΔGP Jenv LV pseudotypes. Furthermore, compared to the 100K and
300K membranes, the 50K membrane had the highest recovery yield for VSVg LV and was thus used in the TFF purification and concentration of ΔGP Jenv LV pseudotypes.
Though the titers obtained with the 100K and 300K MWCO membranes were higher than those obtained with the 50K membrane, it is important to point out that the input titers were higher for TFF performed with the100K and 300K membranes. Thus, the
50K MWCO membrane also had the highest concentration fold of all three membranes.
The high recovery yields we observed were consistent with high recovery yields observed with the 100K and 300K membranes in the literature201, 247, 249, 257. Though our yields were high for both 100K and 300K membranes, we could not reach 100%
109
recovery yields for these membranes, as observed by Geraerts and colleagues, though they did not report the exact inlet and outlet pressures they used247. Since no vector was observed in the waste fraction for these two membranes (Figure 22), it is possible that without higher pressure, some vectors were retained in the membrane. It should be noted that because the 50K membrane has a smaller pore size, the purification process took longer than when using the 100K and 300K membranes. Therefore, 100K and
300K membranes might be more favorable if conducting large scale preparation, without severely compromising recovery.
Despite a high recovery yield from TFF, it was necessary to include additional purification steps as the product was found to be toxic to cells in vitro. Given that TFF can be used to concentrate LV down to relatively small volumes (<10 mL), gradient ultracentrifugation followed by dialysis would be an appropriate downstream purification method. Indeed, subjecting the TFF eluant to gradient ultracentrifugation and dialysis yielded increased titers with a high rate of recovery and abolished the toxicity observed with TFF elution fractions.
Having optimized LV purification parameters, we next sought to establish appropriate models for testing the produced lentivectors. The standard VSVg LV is not ideally suited for lung gene therapy. In vivo studies in mice have shown that VSVg glycoprotein is unable to efficiently transduce the lung cells and relies on chelating and surfactant agents for its entry into the distal respiratory tract176, 177, 184, 231-236, 238-240. Indeed, we also observed a reduced ability of VSVg LV to transduce murine lung slices, despite efficiently transducing ovine lung tissue slices. Conversely, the EBOV-pseudotyped LV has shown promising results in mice133, 134. We were able to replicate such results in
110
murine lung slices but were unable to effectively transduce ovine lung tissue slices.
Others have also shown that EBOV and GP64 glycoprotein LV pseudotypes were unable to efficiently transduce the ovine airway113.
Age also appears to be an important factor contributing to transduction efficiency as neither fetal mice nor fetal sheep lung tissue have been found to be susceptible to transduction with VSVg LV142, 180. However, ΔGP Jenv LV has an affinity toward fetal and newborn lung tissues, making it potentially suitable for lung gene therapy in utero or in neonates113. Indeed, our results demonstrated that the ΔGP Jenv LV was superior to
VSVg and EBOV LV in newborn lamb lung tissue slices.
As a first step toward developing a small animal model for testing ΔGP Jenv LVs, we used the lung tropic AAV6.2FF vector251 to deliver hHyal2, the receptor for JSRV, to mouse lungs resulting in high-level expression in the lung parenchyma. This novel murine model is a promising tool for the study of JSRV, which currently relies on large animal models (e.g. sheep and goats) that are naturally infected by JSRV133, 134, 138-147.
This novel murine model also serves as an example of AAV vector-based repurposing of existing animal models lacking pathogen-specific receptors, as in the case of Severe
Acute Respiratory Syndrome (SARS) corona virus (CoV)258. Research has been unable to establish murine models with which to study SARS-CoV, given that murine receptor, for the SARS-CoV S1 protein258, 259, differs substantially from the human receptor and transgenic mice encoding the human receptor have been used to circumvent this258.
Delivering an AAV vector expressing the human receptor for SARS-CoV might be a more effective alternative. Using this murine AAV-FT-hHyal2 model allowed us to compare VSVg, EBOV, and the two ΔGP Jenv mutant LV pseudotypes. Both ΔGP Jenv
111
N592T and M593E LVs were capable of transducing the FT-hHyal2-bearing murine lung tissue slices and with greater efficiency than EBOV or VSVg LV. With JSRV being a lung tropic virus, it would not be surprising that ΔGP Jenv would have a particular specificity to the lung that is not exhibited by VSVg and EBOV LVs113, 133, 176, 178, 180, 205,
206, 236, 240, 245, 248. In addition, production of the mutant ΔGP Jenv LV pseudotypes yielded titers much higher than those observed for EBOV LV in this or previous studies113, 133, 134, 178, and generated similar titers to those achieved with VSVg LV113, 134,
180, 245, 247, the gold standard in the field. Furthermore, the affinity of Jenv toward slow- replicating differentiated type II cells in fetal lung tissue would decrease the possibility of any potential oncogenic activity of the lentivector113, 260. Finally, since the vector driving expression of Jenv is not flanked by Long Terminal Repeats (LTRs)260, Jenv would not be produced in the transduced cells, which might explain the lack of tumors observed post-delivery of Jenv LVs in vivo in the past113.
In summary, our study emphasizes the suitability of ΔGP Jenv LV as a lung gene therapy vector and highlights the importance of having an appropriate model system to test these vectors prior to in vivo work. This study should assist others with the production of lentivectors using TFF and the testing of these vectors in lung tissue slices.
Materials and Methods
Cell lines
Human embryonic kidney (HEK) 293T cells were cultured and maintained with High
Glucose Dulbecco’s Modified Eagle Medium (DMEM) (Fisher Scientific, Whitby, Ontario,
112
Canada) supplemented with 10% heat-inactivated Bovine Calf Serum (BCS) (Thermo
Fisher Scientific, Mississauga, Ontario, Canada), 2mM L-glutamine (Fisher Scientific),
100U/mL penicillin and 100 µg/mL streptomycin (P/S; Fisher Scientific) in a humidified
5% CO2 incubator at 37˚C.
Animals
All animal experiments were conducted in accordance with the Canadian Council on
Animal Care guidelines and approved by the animal care committee of the University of
Guelph. Four-week old BALB/c female mice were purchased from Charles River
Laboratories (St Constant, Quebec, Canada). They were given 14% Protein Rodent
Maintenance Diet (Teklad Global, Indianapolis, USA) and tap water ad libitum. Mice were acclimated to the environment for seven days prior to being euthanized to procure lungs or treated with 1 x 1011 vg of AAV6.2FF-FTHyal2, as described previously251, 261.
Control mice were administered Phosphate Buffered solution (PBS). One specific- pathogen free Cornell Star neonate lamb was obtained from Ponsonby Research
Station (University of Guelph) and euthanized via intravenous injection of pentobarbital
(Euthansol, Merck Animal Health, Quebec, Canada) prior to harvesting tissues.
Lung tissue slices
Lung slices were prepared and maintained as described in Chapter 2. Briefly, mouse and lamb lungs were excised immediately after euthanasia. Each murine lung lobe was separated with surgical scissors and perfused with 2% low melting point agarose
(LMPa) (Fisher Scientific) in 1X Hank’s Balanced Salt Solution (HBSS) (Fisher
Scientific) (LMPa-HBSS), while intact lamb lungs were perfused via the trachea. The murine lung lobes were kept on ice for 5 min to allow the LMPa-HBSS to solidify prior to
113
being transferred to a separate Petri dish. The ovine lungs were kept in cold HBSS for 1 h prior to coring 8-mm-wide sections. These cores were transferred to Petri dishes.
Lung tissues in Petri dishes were then covered with warm LMPa-HBSS. The encased murine lung lobes and ovine lung cores were kept on ice for 30 min to allow the LMPa-
HBSS to solidify. A scalpel was used to excise gel blocks containing each lung lobe or core, which were then glued with Loctite supler glue (ultra-gel control) to a Microslicer
DTK-3000W (PALL, Mississauga, Ontario Canada) platform to cut 300-µm-thick slices.
The tissue slices were then washed with Lung Slice Wash Media (LSWM) (High
Glucose DMEM with 50U/mL penicillin/streptomycin, 0.2 µg/mL gentamycin, 1.25 µg/mL amphotericin, 10 µM 8-bromo-adenosine 3’,5’-cyclic monophosphate, 100 µM 3- isobutyl-1-methylxanthine, 100 nM dexamethasone) and maintained in Lung Slice
Maintenance Media (LSMM) (LSWM supplemented with 10% FBS and 2 mM L-
Glutamine).
DNA constructs
Construction of the pSin-CASI-hPLAP-WPRE plasmid was carried out as described before (Chapter 3 Methods). Lentivirus vector particles were generated with the use of psPAX2 (second-generation packaging plasmid devoid of vif, vpr, vpu, and nef genes)
(Addgene plasmid #12260, Cambridge, Massachusetts, USA), pSinGFP or pSinCASI- hPLAP-WPRE (backbone constructs with reporter gene), and envelope glycoprotein- encoding vectors pCI-Neo-VSVg, pEF1α-EBOVwt or pCMV3JS21 for VSV, Zaire EBOV and JSRV glycoproteins, respectively.
The Jenv N592T and M593E mutants were generated via site-directed mutagenesis
(SDM) using the Phusion SDM Kit according to the manufacturer’s instructions (Thermo
114
Fisher Scientific) and the following primers for N592T: FWD 5'- gctgcatatgaaatatagaaCtatgttacagcaccaacatc-3', REV 5'- gatgttggtgctgtaacataGttctatatttcatatgcagc-3'; and for M593E: FWD 5'- gctgcatatgaaatatagaaatgagttacagcaccaacatcttatg-3', REV 5'- cataagatgttggtgctgtaactcatttctatatttcatatgcagc-3'. The following program was used for the SDM: one cycle at 98˚C for 30 s followed by 25 cycles of 98˚C for 10 s, 70˚C for 30 s, 72˚C for 7 min, and one final cycle of 72˚C for 2 min.
Development of the AAV6.2FF vector has already been described251. The pACAG-PT-
FT-Hyal2 vector was directionally cloned into restriction sites SphI and XhoI (NEB) in the pACAG-MCS vector, as previously described92. The Hyal2 CDS was amplified from the pL-Hyal2-SN plasmid using Q5 High Fidelity 2X Master Mix (NEB) and the following primers: FWD (5’-tctcgcatgcatgtctgcacttctgatcc-3’), and REV (5’- cacactcgagctacaaggtccaggtaaagg-3’) to include a preprotrypsin (PT) leader sequence that enhances translation of secreted proteins or proteins expressed on the cell surface262, upstream of the FLAG tag in the N-terminus of hHyal2. The following PCR program was used to generate the PCR product: one cycle at 98˚C for 30 s followed by
35 cycles of 98˚C for 5 s, 62˚C for 30 s, 72˚C for 60 s, and one final cycle of 72˚C for 2 min.
Lentiviral vector production
Vector particles were produced in HEK 293T cells. Two million cells were seeded per
10-cm dish for a total of fifteen 10-cm dishes. A day post-seeding, cells were transfected using the following per dish: 5 µg of genome plasmid (pSinGFP or pSinCASI-hPLAP-WPRE), 3.25 µg of helper plasmid (psPAX2) and 1.75 µg of envelope
115
plasmid (pCI-Neo-VSVg, pEF1α-EBOV, pCM3JS21, pCMV3JS21 N592T, or pCMV3JS21 M593E) plus 67.5 µL of 1 mg/mL polyethylenimine transfection polymer, and 450 µL of DMEM per dish. Two hours post-transfection, the media was changed with DMEM supplemented with 2 mM L-glutamine, and P/S (basal DMEM). Every 24 h, for a total of 72 h, the supernatant was collected into 50-mL conical tubes and replenished with fresh basal DMEM. Once collected, the supernatant was spun down in a swinging-bucket centrifuge at 500 g for 5 min to pellet cellular debris. The cleared supernatant was passed through a 0.45-μm polyethersulfone syringe-tip filter (VWR,
Mississauga, Ontario, Canada) and stored at 4˚C until all three vector collections were procured. Ten mL of the neat supernatant were set aside and maintained at 4˚C for vector titrations.
AAV6.2FF-FT-Hyal2 production
In vivo-grade production of AAV6.2FF was described previously251. The AAV6.2FF-FT-
Hyal2 was produced in the same fashion.
Lentivector purification and concentration
Ultracentrifugation
Concentration of LVs by ultracentrifugation has been extensively described12, 94, 150, 247.
Briefly, the volume of supernatant collected over three days of production was distributed among six polypropylene, thin-walled ultra-centrifuge tubes (Beckman
Coulter 360743, Mississauga, Ontario, Canada) (30 mL per tube) suited for a SW32Ti rotor (Beckman Coulter 369650, Mississauga, Ontario, Canada), on top of a 6 mL 20% sucrose cushion. The vector was then spun at 60,000 g for 2 h at 4˚C. Immediately after centrifugation, the vector pellet in each tube was resuspended in 100 µL of Tris-NaCl-
116
EDTA (TNE) buffer, pH 7.4 (100 mM NaCl, 10 mM Tris-HCl, 1 mM EDTA)12, yielding a total of 600 µL of concentrated vector.
PEG precipitation
PEG precipitation with PEG-8000 was performed as described in Chapter 3.
TFF
Our lab recently characterized a TFF protocol156 and this study undertook similar steps.
Three different sized membranes were used in this study: 50K, 100K and 300K. Other than the different membranes, the centramate cassette (PALL) was assembled as specified by the manufacturer and described previously156. The TFF process was run with two different differential pressures: 5 and 0 psi. The process specified below was the same for each membrane (see Figure 28B for set up).
Gauges were placed at the feed line and the retentate line to ensure the pressure at these locations did not exceed 20 psi throughout the procedure. First, the system
(tubing plus membrane) was thoroughly cleansed by running 0.3M NaOH through it for
30 min. The NaOH was then washed off the membrane with 1L of sterile 1X PBS. The lines were monitored for clearance of the NaOH using pH strips. Once the pH was neutralized and the PBS was exhausted, the waste line closest to the feed line was closed. The lentivirus vector supernatant (~225 mL) was then added to the reservoir and allowed to circulate, maintaining a pressure of <10 psi at the feed line and 0 psi at the retentate line. After 10 min of circulation, the waste line was returned to a waste bottle, and the retentate line was clamped shut. For the ‘Pressure’ run, the feed and retentate line gauges were maintained at 15 and 10 psi, respectively, throughout the
117
run, with a differential pressure of 5 psi. For the ‘No-Pressure’ run, <10 and 0 psi were sustained for the feed and retentate line gauges, respectively.
The run was continued until only 5 mL of vector-containing supernatant remained in the reservoir. At this point, we stopped the pump and unclamped the retentate line to set up a buffer exchange. TNE buffer (~300 mL) was added to the reservoir and allowed to briefly re-circulate by returning the waste line to the reservoir. The waste line was then restored to the waste bottle, and the procedure continued by running the TNE buffer until only 1 – 2 mL remained in the reservoir.
At this point, the pump was turned off, and the retentate line unclamped. The remaining waste line was also closed. The retentate line was then disconnected from the reservoir and a 15-mL conical tube was used to collect the first vector-containing eluate from the retentate line. The retentate line was re-connected to the reservoir and 8 mL of TNE buffer was added to the reservoir and allowed to re-circulate for 1 min. Then, another
15-mL tube was used to collect a second elution. Finally, the latter step was repeated a second time to collect a third elution before proceeding to clean the membrane with 0.3
M NaOH. A total of 10 mL was collected from the Waste bottle prior to cleaning for subsequent TU/mL and p24 ELISA assays.
Downstream TFF Purification and Concentration
The concentration of lentivirus particles via a sucrose gradient and PEG-20,000 has been previously described156, 254. Briefly, ultra-clear centrifuge tubes (Beckman Coulter,
344058, Brea, CA, USA) were partially filled with 10 mL of 60% sucrose, followed by 8 mL of 20% sucrose and finally, by 20 mL of pooled TFF elutions. The gradients were spun at 51,000 g for 2 h at 4˚C. Immediately after, the tubes were clamped and a
118
syringe was used as previously described 254 to remove approximately 3 mL of the band between the 20% and 60% sucrose cushions. The vector preparation was then dialyzed and precipitated using PEG-20,000 as defined before156.
Vector Titration
An ELISA kit reactive against the p24 capsid protein of the lentivirus particles was used
(Takara Clontech, Mountain View, California USA), according to manufacturer’s specifications.
The TU/mL assay has been extensively described before and the same approach was employed here151, 254. HEK 293T cells were seeded at 300,000 cells in a 6-well plate and transduced the next day with a serial dilution of vector in six-well plates in the presence of 8 µg/mL polybrene. Twenty-four h later, the percentage of cells expressing the GFP reporter protein was measured with the use of a flow cytometer (Canto II, BD
Biosciences, Mississauga, Ontario, Canada), and analyzed using FlowJo LLC software
(version 10; BD Biosciences). The concentration of the vector was measured as previously stated151..
Transduction of Lung Slices
Lung slices were transduced with 1 x 106 TU per vector in the presence of 8 µg/mL of polybrene. Immediately after transduction, the plate containing the lung slices was placed on a rotator (Nutating Mixer, VWR, Ontario, Mississauga, Canada) overnight in a humidified 5% CO2 incubator at 37˚C. Cells and lung slices transduced with GFP LVs were imaged with a Zeiss AXIO Observer.A1. A Zeiss Stereo Microscope CL1500 ECO
(Zeiss, Toronto, Ontario, Canada) was used to image the murine lung slices following
Alkaline Phosphatase (AP) staining, as described previously263, 264.
119
Immunohistochemistry (IHC)
Immunohistochemistry was performed as described in Chapter 3.
Western Blotting
HEK 293T cells were transfected with Jenv GFP LV, ΔGP Jenv GFP LV, ΔGP Jenv
N592T GFP and ΔGP Jenv M593E GFP as described in the Lentiviral Vector
Production section of the Materials and Methods. Cell lysate preparation and Western blotting were performed as previously described53.
Statistical Analysis
All experiments were performed in triplicate on three separate occasions. We performed a two-way analysis of variance to determine statistical significance when appropriate, using Prism7 software (GraphPad, La Jolla, California, USA).
120
Table 3. Titers and percent recovery of neat and concentrated VSVg, EBOV and ΔGP Jenv pseudotyped LVs after either ultracentrifugation or PEG precipitation.
Volume required to Recovery Vector Method Average (TU/mL) deliver 1x108 TU in vivo (%) (µL)
VSVg LV Neat 4.50 x 106 + 2.2 x 2.22x104 PEG 1.99 x 109 + 1.4 x 50* 78 .0 105§ Ultra 6.80 x 109 + 7.3 x 15* 86.4 109 EBOV LV Neat 5.7 x 104 + 7.3 x 103 1.74x106 108 PEG 1.8 x 107 + 1.6 x 106 5,555 100.0 Ultra 2.1 x 107 + 5.0 x 106 4,761 100.0 ΔGP Jenv Neat 1.1 x 106 + 5.8 x 104 9.09x104 PEG 2.5 x 108 + 7.5 x 107 400 100.0 LV Ultra 1.3 x 108 + 2.3 x 107 769 31.3 *Acceptable volumes for in vivo murine lung delivery
§mean +/- SEM
121
Table 4. Titer values and percent recovery of neat and concentrated VSVg LV from low and high pressure TFF conditions.
Method Membrane Average (TU/mL) Volume (mL) Recovery (%) Low Pressure Neat 2.8 x 105 + 6.6 x 104§ 300 TFF 50K 8.6 x 104 + 1.5 x 104 24 2.5 Neat 2.8 x 105 + 5.9 x 104 300 100K 7.1 x 105 + 2.5 x 105 24 21.0 Neat 2.96 x 105 + 6.7 x 104 100 300K 1.3 x 105 + 1.6 x 104 24 10.6 High Pressure Neat 9.8 x 105 + 4.9 x 104 300 TFF 50K 1.3 x 107 + 2.4 x 106 25.4 100.0 Neat 2.77 x 106 + 6.9 x 105 300 100K 2.9 x 107 + 8.4 x 106 26.4 91.4 Neat 3.69 x 106 + 1.6 x 105 150 300K 2.0 x 107 + 3.2 x 106 25.4 81.3 TFF Downstream Initial 3.5 x 107 + 4.0 x 106 19.5 Purification* (50K) Final 1.3 x 108 + 1.9 x 107 2 99.6 §mean + SEM *Sucrose gradient ultracentrifugation, dialysis, and PEG-20,000 concentration
122
Table 5. Titer values and recovery yield for neat and concentrated ΔGP Jenv LV using low and high pressure TFF conditions.
Volume Recovery Method Sample Pressure Average TU/mL (mL) (%) 50K-TFF Neat Low 6.1 x 105 + 1.1 x 104§ 194 High 1.4 x 106 + 5.7 x 105 184 Final Low 2.2 x 106 + 9.9 x 104 25.9 48.6 High 1.1 x 107 + 1.9 x 106 24.7 97.9
TFF Initial 7 6 Downstream (50K) 1.4 x 10 + 1.2 x 10 19.5 Purification* Final 2.2 x 108 + 2.0 x 107 1.2 94.1 §mean + SEM *Sucrose gradient ultracentrifugation, dialysis, and PEG-20,000 Concentration
123
Figure 21. Comparison of ultracentrifugation and PEG precipitation concentration methods for VSVg, EBOV and ΔGP Jenv lentivirus vectors (LVs).
(A) p24 ELISA values comparing ultracentrifugation (Ultra) and polyethylene glycol (PEG) precipitation methods of concentrating VSVg, EBOV, and ΔGP Jenv pseudotyped LVs. (B – D) Comparison between the infectious units (IFU) and transducing units (TU) per mL calculated from the p24 ELISA and flow cytometry values, respectively, of VSVg (B), EBOV (C), and ΔGP Jenv (D).
124
Figure 22. Representation of the equipment used in this study.
Tangential flow filtration apparatus set up.
125
Figure 23. Optimization of tangential flow filtration as a purification and concentration method for ΔGP Jenv LV.
(A, B) Evaluation of the effect of pressure in TFF using VSVg LV as a prototype LV. (A) A maximum of 5 psi and 0 psi (low pressure) were maintained using the outlet and inlet pressure gauges, respectively. The concentrated vector particles were analyzed via TU/mL. (B) Pressure was increased to 10 psi and 15 psi (high pressure) for the outlet and inlet pressure, respectively, and particles were analyzed once more via TU/mL. (C) Evaluation of the effect of pressure in TFF concentration of ΔGP Jenv LV. Low- and high- pressure conditions were tested and the results are presented as TU/mL units.
126
Figure 24. Toxic effects from tangential flow filtration (TFF) concentrated vector.
Neat, TFF-concentrated LV particles from three elutions (E1, E2 and E3) and TFF waste were added to cells with polybrene at various dilutions, dilution factors (df) of 0.5, 0.05 and 0.005 (not shown here). After 48 hours, transduced cells seen in the brightfield (BF) channel were imaged for GFP expression, observed in the FITC and the merged BF + FITC channels using an inverted fluorescence microscope.
127
Figure 25. Schematic of the methodology used in this study.
Procedures used in this study. PT: post-transfection; AAV: adeno-associated virus vector; TFF: tangential flow filtration.
128
Figure 26. Lentivectors pseudotyped by ΔGP Jenv N592T, and ΔGP Jenv M593E have increased Env expression and produce similar titers to VSVg.
(A) Western Blot of cell lysates of HEK 293Ts transfected with psPax2, pSinGFP and envelopes Jenv, ΔGP Jenv, ΔGP Jenv N592T and ΔGP Jenv M593E, confirming expression of the various envelope proteins. M593E and N592T express higher levels than the wild type Jenv. (B) Comparison of transduction efficiency between VSVg and EBOV LVs concentrated by ultracentrifugation, and TFF- purified and downstream-concentrated ΔGP Jenv, ΔGP Jenv N592T, and ΔGP Jenv M593E pseudotyped LV.
129
Figure 27. Murine lung tissue slice transduction with VSVg, Ebov and ΔGP Jenv lentivirus vectors.
(A) Schematic representation of AAV-FT-hHyal2 vector used to establish murine lungs stably expressing FT-hHyal2. The transgene FT-Hyal2 is preceded by a preprotrypsin tag (PT) and under the control of the CAG promoter (CMV enhancer and Chicken β actin promoter). (B) Immunohistochemical staining showing expression of the FLAG-tagged human Hyal2 (FT-hHyal2) receptor in lung tissue from mice transduced with AAV.2ff-FT-hHyal2 three weeks earlier. (C) Polybrene-treated murine lung slices were transduced with 1 x 106 TU of VSVg, EBOV and ΔGP Jenv, ΔGP Jenv N592T and ΔGP Jenv M593E LV particles bearing the AP reporter gene, respectively. Forty-eight hours post-transduction, the lung tissue slices were stained for AP expression and sandwiched between a coverslip and glass slide and imaged with stereomicroscope.
130
Figure 28. Ovine lung tissue slice transduction with VSVg, EBOV, and ΔGP Jenv lentivirus vectors.
Polybrene-treated ovine lung tissue slices were transduced with 1 x 106 TU of VSVg, EBOV and ΔGP Jenv, ΔGP Jenv N592T and ΔGP Jenv M593E lentivector (LV) particles bearing Green Fluorescent Protein (GFP) reporter genes, respectively. Ovine lung slices were transduced with GFP-bearing LVs, as we had previously found that lung tissue from lambs (<6 months of age) expresses a heat-resistant AP (Rosales Gerpe et al., submitted 2018). Forty-eight hours post-transduction, the lung tissue slices were sandwiched between a coverslip and glass slide and imaged with an inverted fluorescence microscope. The faint GFP-positive foci observed in ovine lung tissue slices transduced by EBOV GFP LV are pinpointed by the white arrows. BF = bright field; FITC = Fluorescein isothiocyanate channel in fluorescence inverted microscope (channel for GFP); Neat = purified, non-concentrated vectors; df = dilution factor (i.e. 1 mL of vector was added and 1 mL of media + polybrene was added for a total volume of 2 mL, so the vector was diluted at a dilution factor of 0.5); E1, E2, E3 = elutions acquired from tangential flow filtration (TFF); Waste = waste liquid collected from TFF.
131
Chapter 5: General Discussion
With this thesis, I set out to understand what determines the tropism difference between
JSRV and ENTV, despite these viruses being so similar. I was also able to exploit the inherent tissue specificity displayed by the JSRV envelope and evaluate it as a pseudotype for a lentivector, proving it could be a promising gene therapy vector. To understand tissue selectivity, I employed respiratory tract tissue slices, and a novel murine lung model stably expressing the JSRV and ENTV receptor, Hyal2.
Tissue slices were advantageous for many aspects of this research. I could generate up to 100 lung tissue slices at a time per mouse, and more with sheep, allowing us to conduct multiple experiments with greater reproducibility (Chapter 2). Furthermore, we were able to extend the maintenance time for these slices up to one month, which is rare in the field112, 118, 121, 122, 125-128. Nevertheless, generating tissue slices requires a considerable amount of time and does not allow for breaks; stopping the procedure by freezing the slices post-perfusion, or prior to wash and maintenance risks death of these slices or decrease in transduction efficiency. To make this procedure more user friendly, it will be important to determine the specific conditions needed to freeze lung slices without compromising their viability.
Our use of lung slices helped us determine that an interplay between the envelope and the LTR, in particular the U3 region of the LTR, was important for JSRV and ENTV in targeting specific tissues. In my experiments with lung slices, I used medium containing hepatocyte growth factor to facilitate infection with JSRV, similarly to Cousens et al.112. I was also able to observe RT activity in the supernatant from lung tissue slices infected
132
with JSRV, demonstrating that virus was being produced from these slices (Chapter 2); as was demonstrated previously112. Furthermore, the supernatant of ovine nasal turbinate slices treated with JSRV also yielded some RT activity, although not to the same extent as ENTV, ENTV Env-expressing chimera 3, or ENTV U3-encoding chimera
4 (Chapter 3). My work suggests JSRV relies mainly on its LTRs to replicate within lung tissue given that lentivectors pseudotyped by Jenv were capable of transducing a variety of different cell lines and lung tissue slices. Furthermore, very faint staining and
RT activity was observed from JSRV-infected ovine turbinate tissue slices, showing that many cell types are permissive to Jenv-mediated entry. Nevertheless, primary ovine chondrocyte cells were less efficiently transduced than other cell lines by LV pseudotyped with Jenv (Chapter 3). Another study has also shown that fetal ovine lung differentiated ciliated columnar and basal cells, where expression of Hyal2 was found in the apical surface below the cilia, are preferred by Jenv pseudotyped lentivectors despite there being other Hyal2 expressing cells in the lung and the use of a promoter that would allow widespread transduction113. The authors did not mention alveolar type
II cells113, which are the preferred cell type of JSRV86, 90, 112. With their 2005 study,
Davey et al. argued that cilia potentially hampered transduction by Jenv pseudotyped
LV at day 65 of the fetal stage and perhaps earlier delivery of Jenv LV would help address whether a more widespread transduction could be possible113.
Conversely, ENTV’s envelope seems to play a greater role in ENTV’s tissue specificity than its LTRs. Eenv LV was not capable of transducing HEK 293T cells at the same level as Jenv LV; however, I did see a level of transduction similar to that of Jenv when
Eenv was used to transduced HEK 293T, which already endogenously express hHyal2,
133
transfected with a vector encoding FT hHyal2 to generate a cell line overexpressing hHyal2. Interestingly, ovine primary chondrocytes showed the same level of transduction as hHyal2-transfected HEK 293T cells by Eenv LV. I also observed strong staining against Env in chondrocyte cells in ovine nasal turbinate slices, indicative of infection by ENTV and chimera 3. Based on my results, there appears to be a minimum threshold amount of Hyal2 required for a cell to be permissive to ENTV. I did not quantify the Hyal2 expression levels in individual cell lines and tissue slices due to the lack of an efficient, commercially available anti-Hyal2 antibody; therefore, I cannot speculate as to the exact amount of Hyal2 necessary for Eenv-mediated entry.
My data also showed that the ENTV LTRs play a crucial role in ENTV’s replication in the nasal tract, which perhaps occurs initially in ovine chondrocytes. However, time-course experiments using chimera 4 (ENTV U3) and 5 (ENTV full LTRs) in vivo would have to be done to be able to reach conclusions on the role of the LTRs, and whether or not
ENTV replication begins in chondrocytes. This is especially the case since this is the first study to suggest that chondrocytes might be the preferred cell type of ENTV infection. In Chapter 2, I presented results from chimera 3 (ENTV env) that tentatively suggest tropism could be attributed not just to Eenv but also to a potential enhancer present in the C-terminus92, given that the ENTV LTR with enhancer was found to have better activity than the ENTV LTR alone. I did not employ CT-truncated envelopes to pseudotype lentivectors in this thesis, but it would have been interesting to see how they would have performed in the nasal tissue slices and primary ovine chondrocytes.
This would have allowed us to confirm whether this enhancer or potential regulatory factor34, 35 was important for envelope-mediated entry.
134
Past studies have painted a complex picture where JSRV and ENTV tropism may rely on concurrent infections1, 18, 19, 46, 94, 112, 144, receptor utilization2, 49, 58, 60, 98, 105, 109, 111, specific pH conditions52, 53, 111 and JSRV and ENTV promoter regions85-90, 92, some of which we have shown in this paper. Particularly, pH seems to be an interesting aspect that connects some of the aforementioned factors that may be important for Env-related tissue selectivity. Fusogenicity to endosomal membranes by Eenv relies on a much lower pH (pH 4.5) than Jenv (pH 6.0)52, 111. Concurrent infections of the lung with bacteria decrease the pH of the lung to approximately pH 6.0265, which suggests that
Jenv would fuse easily to membranes in bacterial concurrent infections1, 18, 19, 46, 94, 112,
144. Interestingly, HA metabolism also relies on low pH levels (pH<5)266. Hyal2 HA- degradation activity is turned on at a low pH (pH 6.0)107, 267, which is similar to the pH of fetal lung extracellular fluid (pH ~ 6.27)113, and Hyal2 activity is upregulated in low pH environments107, 108, 188, 267, 268. These findings further contribute to betaretroviral pathogenesis because they complement data from betaretroviruses such as MMTV, which relies on low pH for entry by the TfR-1 receptor173.
Compared to the lung, the pH of the nose is lower265, 269; this could be because of the physiological structure of nasal turbinates which consist of epithelial tissue intricately linked with cartilage, where chondrocytes maintain a low turnover rate involving extracellular matrix metabolism of HA110, 188, 189, 270-272. It is also of note that low pH promotes a tumorigenic environment by activating HA-degradation enzymes that foster tumor cell invasion268. In the field, ENA is often characterized by tumor invasion and cartilage destruction14, 273. Perhaps, expression of Eenv in ENTV-infected chondrocytes could enable a low pH environment in vivo, precipitating tumorigenesis. Though, in the
135
past, activation of transformation via Hyal2 binding of Jenv has been suggested98, 105,
167, it has not been reproduced to date and a similar mechanism for Eenv remains to be tested.
Finally, there might be other factors associated with tropism which I did not explore, such as a signal peptide in the N-terminus of Env, which has been recently shown to shuttle to the nucleus and potentially affect the replication and oncogenic capacity of
JSRV and ENTV274. Env was also shown to be expressed as a lower molecular weight isoform that could have repercussions on the cell cycle of infected cells274. As such, future mutagenesis studies that address different portions of the envelope protein could be important in understanding the mechanism behind early and late infection stages of
JSRV and ENTV in sheep.
In this thesis, I was also able to exploit the tissue specificity of the JSRV envelope to pseudotype LV for gene therapy. We employed an AAV vector to generate murine lungs expressing human Hyal2 which allowed us to test tropism by delivering Jenv and Eenv pseudotyped LV to hHyal2-expressing murine lung tissue slices. I also identified differences in transduction efficiencies between ovine and murine tissues depending on
LV pseudotype. For example, VSVg-pseudotyped LV performed well in sheep, but not in mice, in contrast to EBOV-pseudotyped LV. Moreover, I observed that LV pseudotyped with two CT mutants of Jenv were capable of efficiently transducing ovine and hHyal2-expressing murine lung tissue slices. I found that modification of the
33, 67, 101, 275 Y590XXM593 motif was important for transduction efficiency , perhaps because these modifications resulted in overexpression of the envelope. In the past, intra- tracheal fetal stage delivery of Jenv LV has not resulted in off-target expression113. This
136
coupled with our data showing the potential to make significantly high titers of Jenv LV using CT point mutants make the use of Jenv pseudotyped LV for lung gene therapy a promising venture.
In this thesis, I set out to address tissue tropism of JSRV and ENTV and discovered in the process a potential new aspect of betaretroviral pathogenesis that involves ovine nasal turbinate chondrocyte. We also constructed a novel tool for measuring Jenv and
Eenv tropism by creating a hHyal2 stably expressing murine lung model. Finally, I exploited the tissue specificity of Jenv by mutating its C-terminus, leading to high titers of LV pseudotyped by Jenv mutants, similar to titers obtained with VSVg LV, the standard in the field. Our work sheds light on the pathogenesis of JSRV and ENTV, but also raises many questions regarding the replication cycle of these viruses, and particularly of ENTV. Future experiments such as delivery of JSRV-ENTV chimeras to neonatal lambs should address the role of Hyal2 within chondrocytes and ENTV infection in nasal turbinate tissue, and specifically cartilage.
137
References
1. Leroux, C, Girard, N, Cottin, V, Greenland, T, Mornex, JF, and Archer, F (2007). Jaagsiekte Sheep Retrovirus (JSRV): from virus to lung cancer in sheep. Vet Res 38: 211-228. 2. Dunlap, KA, Palmarini, M, Adelson, DL, and Spencer, TE (2005). Sheep endogenous betaretroviruses (enJSRVs) and the hyaluronidase 2 (HYAL2) receptor in the ovine uterus and conceptus. Biology of reproduction 73: 271-279. 3. Haig, D (2012). Retroviruses and the placenta. Curr Biol 22: R609-613. 4. Javier, RT, and Butel, JS (2008). The history of tumor virology. Cancer Res 68: 7693-7706. 5. Steele, VE, and Lubet, RA (2010). The Use of Animal Models for Cancer Chemoprevention Drug Development. Semin Oncol 37: 327-338. 6. Jolicoeur, P, Bouchard, L, Guimond, A, Ste-Marie, M, Hanna, Z, and Dievart, A (1998). Use of mouse mammary tumour virus (MMTV)/neu transgenic mice to identify genes collaborating with the c-erbB-2 oncogene in mammary tumour development. Biochem Soc Symp 63: 159-165. 7. Ye, N, Wang, B, Quan, ZF, Pan, HB, Zhang, ML, and Yan, QG (2014). The research progress of the interactions between miRNA and Wnt/beta-catenin signaling pathway in breast cancer of human and mice. Asian Pac J Cancer Prev 15: 1075-1079. 8. Taneja, P, Frazier, DP, Kendig, RD, Maglic, D, Sugiyama, T, Kai, F, et al. (2009). MMTV mouse models and the diagnostic values of MMTV-like sequences in human breast cancer. Expert Rev Mol Diagn 9: 423-440. 9. Hajj, HE, Nasr, R, Kfoury, Y, Dassouki, Z, Nasser, R, Kchour, G, et al. (2012). Animal models on HTLV-1 and related viruses: what did we learn? Front Microbiol 3: 333. 10. Sharp, JM, Angus, KW, Gray, EW, and Scott, FM (1983). Rapid transmission of sheep pulmonary adenomatosis (jaagsiekte) in young lambs. Brief report. Arch Virol 78: 89-95. 11. Palmarini, M, Sharp, JM, Lee, C, and Fan, H (1999). In Vitro Infection of Ovine Cell Lines by Jaagsiekte Sheep Retrovirus. J Virol 73: 10070-10078. 12. Palmarini, M, Sharp, JM, de las Heras, M, and Fan, H (1999). Jaagsiekte sheep retrovirus is necessary and sufficient to induce a contagious lung cancer in sheep. J Virol 73: 6964-6972. 13. Palmarini, M, Fan, H, and Sharp, JM (1997). Sheep pulmonary adenomatosis: a unique model of retrovirus-associated lung cancer. Trends Microbiol 5: 478-483.
138
14. Walsh, SR, Linnerth-Petrik, NM, Yu, DL, Foster, RA, Menzies, PI, Diaz-Méndez, A, et al. (2013). Experimental transmission of enzootic nasal adenocarcinoma in sheep. Veterinary Research 44: 66. 15. Sharp, JM, and DeMartini, JC (2003). Natural history of JSRV in sheep. Curr Top Microbiol Immunol 275: 55-79. 16. Walsh, SR, Linnerth-Petrik, NM, Laporte, AN, Menzies, PI, Foster, RA, and Wootton, SK (2010). Full-length genome sequence analysis of enzootic nasal tumor virus reveals an unusually high degree of genetic stability. Virus Res 151: 74-87. 17. Gardiner, N, Jogai, S, and Wallis, A (2014). The revised lung adenocarcinoma classification—an imaging guide. J Thorac Dis 6: S537-546. 18. Mornex, JF, Thivolet, F, De las Heras, M, and Leroux, C (2003). Pathology of human bronchioloalveolar carcinoma and its relationship to the ovine disease. Curr Top Microbiol Immunol 275: 225-248. 19. Griffiths, DJ, Martineau, HM, and Cousens, C (2010). Pathology and pathogenesis of ovine pulmonary adenocarcinoma. Journal of comparative pathology 142: 260-283. 20. Maeda, N, and Fan, H (2008). Signal transduction pathways utilized by enzootic nasal tumor virus (ENTV-1) envelope protein in transformation of rat epithelial cells resemble those used by jaagsiekte sheep retrovirus. Virus Genes 36: 147-155. 21. Ruscetti, SK (2011). Deregulation of Signal Transduction Pathways by Oncogenic Retroviruses. In: Dudley, J (ed). Retroviruses and Insight into Cancer. Springer. 22. King, AMQ, Lefkowitz, EJ, Mushegian, AR, Adams, MJ, Dutilh, BE, Gorbalenya, AE, et al. (2018). Changes to taxonomy and the International Code of Virus Classification and Nomenclature ratified by the International Committee on Taxonomy of Viruses (2018). Arch Virol. 23. Ross, SR (2010). Mouse Mammary Tumor Virus Molecular Biology and Oncogenesis. Viruses 2: 2000-2012. 24. Rein, A (2011). Murine leukemia viruses: objects and organisms. Advances in virology 2011: 403419. 25. Palmarini, M, and Fan, H (2003). Jaagsiekte Sheep Retrovirus and Lung Cancer, Springer Science and Business Media, 84pp. 26. Buchschacher, GL (2001). Introduction to retroviruses and retroviral vectors. Somat Cell Mol Genet 26: 1-11. 27. Coffin, J (1997). Retroviruses, Cold Spring Harbor Laboratory Press, Cold Spring Harbor (NY). 28. Bai, J, Zhu, R-Y, Stedman, K, Cousens, C, Carlson, J, Sharp, JM, et al. (1996). Unique Long Terminal Repeat U3 Sequences Distinguish Exogenous Jaagsiekte Sheep
139
Retrovirus Assoiated with Ovine Pulmonary Carcinoma from Endogenous Loci in the Sheep Genome. Journal of Virology 70: 3159-3168. 29. Palmarini, M, Gray, CA, Carpenter, K, Fan, H, Bazer, FW, and Spencer, TE (2001). Expression of Endogenous Betaretroviruses in the Ovine Uterus: Effects of Neonatal Age, Estrous Cycle, Pregnancy, and Progesterone. J Virol 75: 11319-11327. 30. Sonigo, P, Barker, C, Hunter, E, and Wain-Hobson, S (1986). Nucleotide sequence of Mason-Pfizer monkey virus: an immunosuppressive D-type retrovirus. Cell 45: 375-385. 31. Gifford, R, and Tristem, M (2003). The Evolution, Distribution and Diversity of Endogenous Retroviruses. Virus Genes 26: 291-315. 32. Rosati, S, Pittau, M, Alberti, A, Pozzi, S, York, DF, Sharp, JM, et al. (2000). An accessory open reading frame (orf-x) of jaagsiekte sheep retrovirus is conserved between different virus isolates. Virus Research 66: 109-116. 33. Maeda, N, Palmarini, M, Murgia, C, and Fan, H (2001). Direct transformation of rodent fibroblasts by jaagsiekte sheep retrovirus DNA. Proc Natl Acad Sci U S A 98: 4449-4454. 34. Nitta, T, Hofacre, A, Hull, S, and Fan, H (2009). Identification and mutational analysis of a Rej response element in Jaagsiekte sheep retrovirus RNA. J Virol 83: 12499-12511. 35. Hofacre, A, Nitta, T, and Fan, H (2009). Jaagsiekte sheep retrovirus encodes a regulatory factor, Rej, required for synthesis of Gag protein. J Virol 83: 12483-12498. 36. Hofacre, A, and Fan, H (2010). Jaagsiekte sheep retrovirus biology and oncogenesis. Viruses 2: 2618-2648. 37. Dannull, J, Surovoy, A, Jung, G, and Moelling, K (1994). Specific binding of HIV- 1 nucleocapsid protein to PSI RNA in vitro requires N-terminal zinc finger and flanking basic amino acid residues. EMBO J 13: 1525-1533. 38. Berkowitz, R, J, L, and Goff, S (1993). Specific Binding of Human Immunodeficiency Virus Type 1 gag Polyprotein and Nucleocapsid Protein to Viral RNAs Detected by RNA Mobility Shift Assays. Journal of Virology 67: 7190-7200. 39. Prats, AC, Housset, V, de Billy, G, Cornille, F, Prats, H, Roques, B, et al. (1991). Viral RNA annealing activities of the nucleocapsid protein of Moloney murine leukemia virus are zinc independent. Nucleic Acids Res 19: 3533-3541. 40. Lee, EG, and Linial, ML (2004). Basic Residues of the Retroviral Nucleocapsid Play Different Roles in Gag-Gag and Gag-Ψ RNA Interactions. J Virol 78: 8486-8495. 41. D'Souza, V, Melamed, J, Habib, D, Pullen, K, Wallace, K, and Summers, MF (2001). Identification of a high affinity nucleocapsid protein binding element within the Moloney murine leukemia virus Psi-RNA packaging signal: implications for genome recognition. J Mol Biol 314: 217-232.
140
42. Dey, A, York, D, Smalls-Mantey, A, and Summers, MF (2005). Composition and sequence-dependent binding of RNA to the nucleocapsid protein of Moloney murine leukemia virus. Biochemistry 44: 3735-3744. 43. Yuan, B, Li, X, and Goff, SP (1999). Mutations altering the moloney murine leukemia virus p12 Gag protein affect virion production and early events of the virus life cycle. EMBO J 18: 4700-4710. 44. Murcia, PR, Arnaud, F, and Palmarini, M (2007). The Transdominant Endogenous Retrovirus enJS56A1 Associates with and Blocks Intracellular Trafficking of Jaagsiekte Sheep Retrovirus Gag▿. J Virol 81: 1762-1772. 45. Hong, S, Choi, G, Park, S, Chung, A-S, Hunter, E, and Rhee, SS (2001). Type D Retrovirus Gag Polyprotein Interacts with the Cytosolic Chaperonin TRiC. Journal of Virology 75: 2526-2534. 46. Armezzani, AV, Mariana. Spencer, Thomas E. Palmarini, Massimo. Arnaud, Frederick. (2014). Menage a Trois: The Evolutionary Interplay between enJSRVs and Domestic Sheep. Viruses 6: 4926-4945. 47. Weil, AF, Ghosh, D, Zhou, Y, Seiple, L, McMahon, MA, Spivak, AM, et al. (2013). Uracil DNA glycosylase initiates degradation of HIV-1 cDNA containing misincorporated dUTP and prevents viral integration. Proc Natl Acad Sci U S A 110: E448-457. 48. Liu, SL, Duh, FM, Lerman, MI, and Miller, AD (2003). Role of virus receptor Hyal2 in oncogenic transformation of rodent fibroblasts by sheep betaretrovirus env proteins. J Virol 77: 2850-2858. 49. Rai, SK, Duh, FM, Vigdorovich, V, Danilkovitch-Miagkova, A, Lerman, MI, and Miller, AD (2001). Candidate tumor suppressor HYAL2 is a glycosylphosphatidylinositol (GPI)-anchored cell-surface receptor for jaagsiekte sheep retrovirus, the envelope protein of which mediates oncogenic transformation. Proc Natl Acad Sci U S A 98: 4443-4448. 50. Tedbury, PR, and Freed, EO (2015). The cytoplasmic tail of retroviral envelope glycoproteins. Prog Mol Biol Transl Sci 129: 253-284. 51. Aguilar, HC, Anderson, WF, and Cannon, PM (2003). Cytoplasmic Tail of Moloney Murine Leukemia Virus Envelope Protein Influences the Conformation of the Extracellular Domain: Implications for Mechanism of Action of the R Peptide. J Virol 77: 1281-1291. 52. Cote, M, Zheng, YM, Albritton, LM, and Liu, SL (2008). Fusogenicity of Jaagsiekte sheep retrovirus envelope protein is dependent on low pH and is enhanced by cytoplasmic tail truncations. J Virol, vol. 82: United States. pp 2543-2554. 53. Walsh, SR, de Jong, JG, van Vloten, JP, Gerpe, MC, Santry, LA, and Wootton, SK (2017). Truncation of the enzootic nasal tumor virus envelope protein cytoplasmic tail increases Env-mediated fusion and infectivity. J Gen Virol 98: 108-120.
141
54. Alberti, A, Murgia, C, Liu, SL, Mura, M, Cousens, C, Sharp, M, et al. (2002). Envelope-induced cell transformation by ovine betaretroviruses. J Virol 76: 5387-5394. 55. Lindemann, D, Steffen, I, and Pöhlmann, S (2013). Cellular entry of retroviruses. Adv Exp Med Biol 790: 128-149. 56. Aizawa, S, Suda, Y, Furuta, Y, Yagi, T, Takeda, N, Watanabe, N, et al. (1990). Env-derived gp55 gene of Friend spleen focus-forming virus specifically induces neoplastic proliferation of erythroid progenitor cells. EMBO J 9: 2107-2116. 57. Ross, SR, Schofield, JJ, Farr, CJ, and Bucan, M (2002). Mouse transferrin receptor 1 is the cell entry receptor for mouse mammary tumor virus. Proc Natl Acad Sci U S A 99: 12386-12390. 58. Van Hoeven, NS, and Miller, AD (2005). Improved enzootic nasal tumor virus pseudotype packaging cell lines reveal virus entry requirements in addition to the primary receptor Hyal2. J Virol 79: 87-94. 59. Csoka, AB, Frost, GI, and Stern, R (2001). The six hyaluronidase-like genes in the human and mouse genomes. Matrix biology : journal of the International Society for Matrix Biology 20: 499-508. 60. Dirks, C, Duh, FM, Rai, SK, Lerman, MI, and Miller, AD (2002). Mechanism of cell entry and transformation by enzootic nasal tumor virus. J Virol 76: 2141-2149. 61. Goff, SP (2007). Host factors exploited by retroviruses. Nat Rev Microbiol 5: 253- 263. 62. Hu, WS, and Hughes, SH (2012). HIV-1 Reverse Transcription. Cold Spring Harb Perspect Med 2. 63. Rosales Gerpe, M (2014). The Role of APOBEC3 in Controlling Retroviral Spread and Zoonoses. In: Ottawa, Uo (ed). Thesis: University of Ottawa. p 147. 64. Marquet, R, Isel, C, Ehresmann, C, and Ehresmann, B (1995). tRNAs as primer of reverse transcriptases. Biochimie 77: 113-124. 65. Andersen, JL, Le Rouzic, E, and Planelles, V (2008). HIV-1 Vpr: mechanisms of G2 arrest and apoptosis. Experimental and molecular pathology 85: 2-10. 66. Elis, E, Ehrlich, M, Prizan-Ravid, A, Laham-Karam, N, and Bacharach, E (2012). p12 tethers the murine leukemia virus pre-integration complex to mitotic chromosomes. PLoS Pathog 8: e1003103. 67. Hull, S, Lim, J, Hamil, A, Nitta, T, and Fan, H (2012). Analysis of jaagsiekte sheep retrovirus (JSRV) envelope protein domains in transformation. Virus Genes 45: 508-517. 68. Zabransky, A, Sakalian, M, and Pichova, I (2005). Localization of self-interacting domains within betaretrovirus Gag polyproteins. Virology 332: 659-666. 69. Arnaud, F, Varela, M, Spencer, TE, and Palmarini, M (2008). Coevolution of endogenous Betaretroviruses of sheep and their host. Cell Mol Life Sci 65: 3422-3432.
142
70. Demirov, DG, Ono, A, Orenstein, JM, and Freed, EO (2002). Overexpression of the N-terminal domain of TSG101 inhibits HIV-1 budding by blocking late domain function. Proc Natl Acad Sci U S A 99: 955-960. 71. Chudak, C, Beimforde, N, George, M, Zimmerman, A, Lausch, V, Hanke, K, et al. (2013). Identification of late asembly domains of the human endogenous retrovirus-K (HML-2). Retrovirology 10: 140. 72. Martin-Serrano, J, Yaravoy, A, Perez-Caballero, D, and Bieniasz, PD (2003). Divergent retroviral late-budding domains recruit vacuolar protein sorting factors by using alternative adaptor proteins. Proc Natl Acad Sci U S A 100: 12414-12419. 73. Slagsvold, T, Pattni, K, Malerod, L, and Stenmark, H (2006). Endosomal and non-endosomal functions of ESCRT proteins. Trends in Cell Biology 16: 317-326. 74. Palmarini, M, Mura, M, and Spencer, TE (2004). Endogenous betaretroviruses of sheep: teaching new lessons in retroviral interference and adaptation. J Gen Virol 85: 1- 13. 75. Spencer, TE, Mura, M, Gray, CA, Griebel, PJ, and Palmarini, M (2003). Receptor usage and fetal expression of ovine endogenous betaretroviruses: implications for coevolution of endogenous and exogenous retroviruses. J Virol 77: 749-753. 76. Lerche, NW (2010). Simian retroviruses: infection and disease--implications for immunotoxicology research in primates. J Immunotoxicol 7: 93-101. 77. Czarneski, J, Rassa, JC, and Ross, SR (2003). Mouse mammary tumor virus and the immune system. Immunologic research 27: 469-480. 78. Montiel, NA (2010). An updated review of simian betaretrovirus (SRV) in macaque hosts. J Med Primatol 39: 303-314. 79. Youssef, G, Wallace, WA, Dagleish, MP, Cousens, C, and Griffiths, DJ (2015). Ovine pulmonary adenocarcinoma: a large animal model for human lung cancer. ILAR J 56: 99-115. 80. Svara, T, Gombac, M, Vrecl, M, Juntes, P, Kostanjsek, R, Pogacnik, A, et al. (2006). Enzootic nasal adenocarcinoma of sheep in Slovenia. Journal of veterinary medicine A, Physiology, pathology, clinical medicine 53: 26-29. 81. De las Heras, M, Ortín, A, Cousens, C, Minguijón, E, and Sharp, JM (2003). Enzootic nasal adenocarcinoma of sheep and goats. Curr Top Microbiol Immunol 275: 201-223. 82. De las Heras, M, Garcia de Jalon, JA, and Sharp, JM (1991). Pathology of enzootic intranasal tumor in thirty-eight goats. Vet Pathol 28: 474-481. 83. Rosen, CA, Haseltine, WA, Lenz, J, Ruprecht, R, and Cloyd, MW (1985). Tissue selectivity of murine leukemia virus infection is determined by long terminal repeat sequences. J Virol 55: 862-866.
143
84. Hull, S, and Boris-Lawrie, K (2002). RU5 of Mason-Pfizer monkey virus 5' long terminal repeat enhances cytoplasmic expression of human immunodeficiency virus type 1 gag-pol and nonviral reporter RNA. J Virol 76: 10211-10218. 85. McGee-Estrada, K, Palmarini, M, Hallwirth, C, and Fan, H (2005). A Moloney murine leukemia virus driven by the Jaagsiekte sheep retrovirus enhancers shows enhanced specificity for infectivity in lung epithelial cells. Virus Genes 31: 257-263. 86. McGee-Estrada, K, Palmarini, M, and Fan, H (2002). HNF-3beta is a critical factor for the expression of the Jaagsiekte sheep retrovirus long terminal repeat in type II pneumocytes but not in Clara cells. Virology 292: 87-97. 87. McGee-Estrada, K, and Fan, H (2007). Comparison of LTR enhancer elements in sheep beta retroviruses: insights into the basis for tissue-specific expression. Virus Genes 35: 303-312. 88. McGee-Estrada, K, and Fan, H (2006). In vivo and in vitro analysis of factor binding sites in Jaagsiekte sheep retrovirus long terminal repeat enhancer sequences: roles of HNF-3, NF-I, and C/EBP for activity in lung epithelial cells. J Virol 80: 332-341. 89. Eckstrand, CD, Castillo, D, McDonnel, SJ, Hillman, CN, Vapniarsky, N, Shanthalingam, S, et al. (2013). Genetic variability and in vitro transcriptional permissibility of primary ovine beta-retrovirus promoter isolates. Am J Vet Res 74: 1421- 1427. 90. Palmarini, M, Datta, S, Omid, R, Murgia, C, and Fan, H (2000). The Long Terminal Repeat of Jaagsiekte Sheep Retrovirus Is Preferentially Active in Differentiated Epithelial Cells of the Lungs. J Virol, vol. 74. pp 5776-5787. 91. Müller, R, and Müller, D (1984). Co-transfection of normal NIH/3T3 DNA and retroval LTR sequences: a novel strategy for the detection of potential c-onc genes. EMBO J 3: 1121-1127. 92. Yu, DL, Linnerth-Petrik, NM, Halbert, CL, Walsh, SR, Miller, AD, and Wootton, SK (2011). Jaagsiekte sheep retrovirus and enzootic nasal tumor virus promoters drive gene expression in all airway epithelial cells of mice but only induce tumors in the alveolar region of the lungs. J Virol 85: 7535-7545. 93. Wootton, SK, Halbert, CL, and Miller, AD (2005). Sheep retrovirus structural protein induces lung tumours. Nature 434: 904-907. 94. Caporale, M, Cousens, C, Centorame, P, Pinoni, C, De las Heras, M, and Palmarini, M (2006). Expression of the jaagsiekte sheep retrovirus envelope glycoprotein is sufficient to induce lung tumors in sheep. J Virol 80: 8030-8037. 95. Gui, GP, Puddefoot, JR, Vinson, GP, Wells, CA, and Carpenter, R (1997). Altered cell-matrix contact: a prerequisite for breast cancer metastasis? Br J Cancer 75: 623-633. 96. Nykopp, TK, Rilla, K, Tammi, MI, Tammi, RH, Sironen, R, Hämäläinen, K, et al. (2010). Hyaluronan synthases (HAS1-3) and hyaluronidases (HYAL1-2) in the
144
accumulation of hyaluronan in endometrioid endometrial carcinoma. BMC Cancer 10: 512. 97. Lerman, MI, and Minna, JD (2000). The 630-kb lung cancer homozygous deletion region on human chromosome 3p21.3: identification and evaluation of the resident candidate tumor suppressor genes. The International Lung Cancer Chromosome 3p21.3 Tumor Suppressor Gene Consortium. Cancer Res 60: 6116-6133. 98. Miller, AD (2008). Hyaluronidase 2 and its intriguing role as a cell-entry receptor for oncogenic sheep retroviruses. Semin Cancer Biol 18: 296-301. 99. Palmarini, M, Maeda, N, Murgia, C, De-Fraja, C, Hofacre, A, and Fan, H (2001). A phosphatidylinositol 3-kinase docking site in the cytoplasmic tail of the Jaagsiekte sheep retrovirus transmembrane protein is essential for envelope-induced transformation of NIH 3T3 cells. J Virol 75: 11002-11009. 100. Zavala, G, Pretto, C, Chow, YH, Jones, L, Alberti, A, Grego, E, et al. (2003). Relevance of Akt phosphorylation in cell transformation induced by Jaagsiekte sheep retrovirus. Virology 312: 95-105. 101. Cousens, C, Maeda, N, Murgia, C, Dagleish, MP, Palmarini, M, and Fan, H (2007). In vivo tumorigenesis by Jaagsiekte sheep retrovirus (JSRV) requires Y590 in Env TM, but not full-length orfX open reading frame. Virology 367: 413-421. 102. Liu, S-L, and Miller, AD (2006). Oncogenic transformation by the jaagsiekte sheep retrovirus envelope protein. Oncogene 26: 789-801. 103. Palmarini, M, Hallwirth, C, York, D, Murgia, C, de Oliveira, T, Spencer, T, et al. (2000). Molecular Cloning and Functional Analysis of Three Type D Endogenous Retroviruses of Sheep Reveal a Different Cell Tropism from That of the Highly Related Exogenous Jaagsiekte Sheep Retrovirus. Journal of Virology 74: 8065-8076. 104. Hull, S, and Fan, H (2006). Mutational analysis of the cytoplasmic tail of jaagsiekte sheep retrovirus envelope protein. J Virol 80: 8069-8080. 105. Miller, AD, Vigdorovich, V, Strong, RK, Fernandes, RJ, and Lerman, MI (2006). Letter to the Editor: Hyal2, where are you? Osteoarthritis and Cartilage 14: 1315-1317. 106. de la Motte, C, Nigro, J, Vasanji, A, Rho, H, Kessler, S, Bandyopadhyay, S, et al. (2009). Platelet-Derived Hyaluronidase 2 Cleaves Hyaluronan into Fragments that Trigger Monocyte-Mediated Production of Proinflammatory Cytokines. Am J Pathol 174: 2254-2264. 107. Andre, B, Duterme, C, Van Moer, K, Mertens-Strijthagen, J, Jadot, M, and Flamion, B (2011). Hyal2 is a glycosylphosphatidylinositol-anchored, lipid raft- associated hyaluronidase. Biochemical and biophysical research communications 411: 175-179. 108. Albeiroti, S, Ayasoufi, K, Hill, DR, Shen, B, and de la Motte, CA (2015). Platelet hyaluronidase-2: an enzyme that translocates to the surface upon activation to function in extracellular matrix degradation. Blood 125: 1460-1469.
145
109. Modelski, MJ, Menlah, G, Wang, Y, Dash, S, Wu, K, Galileo, DS, et al. (2014). Hyaluronidase 2: a novel germ cell hyaluronidase with epididymal expression and functional roles in mammalian sperm. Biology of reproduction 91: 109. 110. Chow, G, Knudson, CB, and Knudson, W (2006). Human hyaluronidase-2 is localized intracellularly in articular chondrocytes and other cultured cell lines. Osteoarthritis Cartilage 14: 1312-1314. 111. Cote, M, Kucharski, TJ, and Liu, SL (2008). Enzootic nasal tumor virus envelope requires a very acidic pH for fusion activation and infection. J Virol 82: 9023-9034. 112. Cousens, C, Alleaume, C, Bijsmans, E, Martineau, HM, Finlayson, J, Dagleish, MP, et al. (2015). Jaagsiekte sheep retrovirus infection of lung slice cultures. Retrovirology 12: 31. 113. Davey, MG, Zoltick, PW, Todorow, CA, Limberis, MP, Ruchelli, ED, Hedrick, HL, et al. (2012). Jaagsiekte sheep retrovirus pseudotyped lentiviral vector-mediated gene transfer to fetal ovine lung. Gene Ther 19: 201-209. 114. Coil, DA, Strickler, JH, Rai, SK, and Miller, AD (2001). Jaagsiekte sheep retrovirus Env protein stabilizes retrovirus vectors against inactivation by lung surfactant, centrifugation, and freeze-thaw cycling. J Virol 75: 8864-8867. 115. Qi, J-w, Wu, X-l, Liu, S-y, and Cao, G-f (2012). Expression of Endogenous Beta Retroviruses and Hyal-2 mRNA in Immune Organs of Fetuses and Lambs. 116. Walsh, SR, Gerpe, MC, and Wootton, SK (2016). Construction of a molecular clone of ovine enzootic nasal tumor virus. Virol J 13: 209. 117. de Graaf, IA, Olinga, P, de Jager, MH, Merema, MT, de Kanter, R, van de Kerkhof, EG, et al. (2010). Preparation and incubation of precision-cut liver and intestinal slices for application in drug metabolism and toxicity studies. Nat Protoc 5: 1540-1551. 118. Lyons-Cohen, MR, Thomas, SY, Cook, DN, and Nakano, H (2017). Precision-cut Mouse Lung Slices to Visualize Live Pulmonary Dendritic Cells. J Vis Exp. 119. Maund, SL, Nolley, R, and Peehl, DM (2014). Optimization and comprehensive characterization of a faithful tissue culture model of the benign and malignant human prostate. Lab Invest 94: 208-221. 120. van Geer, MA, Kuhlmann, KF, Bakker, CT, ten Kate, FJ, Oude Elferink, RP, and Bosma, PJ (2009). Ex-vivo evaluation of gene therapy vectors in human pancreatic (cancer) tissue slices. World J Gastroenterol 15: 1359-1366. 121. Bai, Y, Krishnamoorthy, N, Patel, KR, Rosas, I, Sanderson, MJ, and Ai, X (2016). Cryopreserved Human Precision-Cut Lung Slices as a Bioassay for Live Tissue Banking. A Viability Study of Bronchodilation with Bitter-Taste Receptor Agonists. Am J Respir Cell Mol Biol 54: 656-663.
146
122. Kim, HJ, Kim, Y, Park, SJ, Bae, B, Kang, HR, Cho, SH, et al. (2015). Airway Smooth Muscle Sensitivity to Methacholine in Precision-Cut Lung Slices (PCLS) from Ovalbumin-induced Asthmatic Mice. Korean J Physiol Pharmacol 19: 65-71. 123. Rosner, SR, Ram-Mohan, S, Paez-Cortez, JR, Lavoie, TL, Dowell, ML, Yuan, L, et al. (2014). Airway contractility in the precision-cut lung slice after cryopreservation. Am J Respir Cell Mol Biol 50: 876-881. 124. Freese, A, Kaplitt, MG, O'Connor, WM, Abbey, M, Langer, D, Leone, P, et al. (1997). Direct gene transfer into human epileptogenic hippocampal tissue with an adeno-associated virus vector: implications for a gene therapy approach to epilepsy. Epilepsia 38: 759-766. 125. Sanderson, MJ (2011). Exploring lung physiology in health and disease with lung slices. Pulm Pharmacol Ther 24: 452-465. 126. Held, HD, Martin, C, and Uhlig, S (1999). Characterization of airway and vascular responses in murine lungs. Br J Pharmacol 126: 1191-1199. 127. Liberati, TA, Randle, MR, and Toth, LA (2010). In vitro lung slices: a powerful approach for assessment of lung pathophysiology. Expert Rev Mol Diagn 10: 501-508. 128. Liu, R, An, L, Liu, G, Li, X, Tang, W, and Chen, X (2015). Mouse lung slices: An ex vivo model for the evaluation of antiviral and anti-inflammatory agents against influenza viruses. Antiviral Res 120: 101-111. 129. Wu, W, Booth, JL, Duggan, ES, Wu, S, Patel, KB, Coggeshall, KM, et al. (2010). Innate immune response to H3N2 and H1N1 influenza virus infection in a human lung organ culture model. Virology 396: 178-188. 130. Wu, W, Zhang, W, Booth, JL, and Metcalf, JP (2012). Influenza A(H1N1)pdm09 virus suppresses RIG-I initiated innate antiviral responses in the human lung. PLoS One 7: e49856. 131. Löndt, BZ, Brookes, SM, Nash, BJ, Núñez, A, Stagg, DA, and Brown, IH (2013). The infectivity of pandemic 2009 H1N1 and avian influenza viruses for pigs: an assessment by ex vivo respiratory tract organ culture*. Influenza and Other Respiratory Viruses 7: 393-402. 132. Van Poucke, SG, Nicholls, JM, Nauwynck, HJ, and Van Reeth, K (2010). Replication of avian, human and swine influenza viruses in porcine respiratory explants and association with sialic acid distribution. Virol J 7: 38. 133. Medina, MF, Kobinger, GP, Rux, J, Gasmi, M, Looney, DJ, Bates, P, et al. (2003). Lentiviral vectors pseudotyped with minimal filovirus envelopes increased gene transfer in murine lung. Molecular therapy : the journal of the American Society of Gene Therapy 8: 777-789. 134. Kobinger, GP, Weiner, DJ, Yu, QC, and Wilson, JM (2001). Filovirus- pseudotyped lentiviral vector can efficiently and stably transduce airway epithelia in vivo. Nat Biotechnol 19: 225-230.
147
135. Ulasov, IV, Rivera, AA, Nettelbeck, DM, Rivera, LB, Mathis, JM, Sonabend, AM, et al. (2007). An oncolytic adenoviral vector carrying the tyrosinase promoter for glioma gene therapy. International journal of oncology 31: 1177-1185. 136. Zhu, ZB, Rivera, AA, Makhija, SK, Lu, B, Wang, M, Izumi, M, et al. (2007). Targeting lung cancer using an infectivity enhanced CXCR4-CRAd. Lung Cancer 55: 145-156. 137. Diallo, JS, Roy, D, Abdelbary, H, De Silva, N, and Bell, JC (2011). Ex vivo infection of live tissue with oncolytic viruses. J Vis Exp. 138. Zurzolo, C, Polistina, C, Saini, M, Gentile, R, Aloj, L, Migliaccio, G, et al. (1992). Opposite polarity of virus budding and of viral envelope glycoprotein distribution in epithelial cells derived from different tissues. J Cell Biol 117: 551-564. 139. Lamp, B, Dietzel, E, Kolesnikova, L, Sauerhering, L, Erbar, S, Weingartl, H, et al. (2013). Nipah virus entry and egress from polarized epithelial cells. J Virol 87: 3143- 3154. 140. Javier, RT (2008). Cell polarity proteins: common targets for tumorigenic human viruses. Oncogene 27: 7031-7046. 141. Wang, TG, Ye, J, Lairmore, MD, and Green, PL (2000). In vitro cellular tropism of human T cell leukemia virus type 2. AIDS Res Hum Retroviruses 16: 1661-1668. 142. Buckley, SM, Howe, SJ, Sheard, V, Ward, NJ, Coutelle, C, Thrasher, AJ, et al. (2008). Lentiviral transduction of the murine lung provides efficient pseudotype and developmental stage-dependent cell-specific transgene expression. Gene Ther 15: 1167-1175. 143. Pickles, RJ, Fahrner, JA, Petrella, JM, Boucher, RC, and Bergelson, JM (2000). Retargeting the coxsackievirus and adenovirus receptor to the apical surface of polarized epithelial cells reveals the glycocalyx as a barrier to adenovirus-mediated gene transfer. J Virol 74: 6050-6057. 144. Caporale, M, Martineau, H, De las Heras, M, Murgia, C, Huang, R, Centorame, P, et al. (2013). Host species barriers to Jaagsiekte sheep retrovirus replication and carcinogenesis. J Virol 87: 10752-10762. 145. Johnson, C, Sanders, K, and Fan, H (2010). Jaagsiekte Sheep Retrovirus Transformation in Madin-Darby Canine Kidney Epithelial Cell Three-Dimensional Culture▿. J Virol 84: 5379-5390. 146. Linnerth-Petrik, NM, Santry, LA, Yu, DL, and Wootton, SK (2012). Adeno- associated virus vector mediated expression of an oncogenic retroviral envelope protein induces lung adenocarcinomas in immunocompetent mice. PLoS One 7: e51400. 147. Liu, SL, and Miller, AD (2005). Transformation of Madin-Darby Canine Kidney Epithelial Cells by Sheep Retrovirus Envelope Proteins. J Virol 79: 927-933. 148. Modlich, U, Navarro, S, Zychlinski, D, Maetzig, T, Knoess, S, Brugman, MH, et al. (2009). Insertional transformation of hematopoietic cells by self-inactivating lentiviral
148
and gammaretroviral vectors. Molecular therapy : the journal of the American Society of Gene Therapy 17: 1919-1928. 149. Garcia, V, Krishnan, R, Davis, C, Batenchuk, C, Le Boeuf, F, Abdelbary, H, et al. (2014). High-throughput titration of luciferase-expressing recombinant viruses. J Vis Exp: 51890. 150. Tiscornia, G, Singer, O, and Verma, IM (2006). Production and purification of lentiviral vectors. Nat Protoc 1: 241-245. 151. Verhoeyen, E, and Cosset, FL (2009). Hematopoietic stem cell targeting with surface-engineered lentiviral vectors. Cold Spring Harb Protoc 2009: pdb.prot5276. 152. Diallo, JS, Vähä-Koskela, M, Le Boeuf, F, and Bell, J (2012). Propagation, purification, and in vivo testing of oncolytic vesicular stomatitis virus strains. Methods Mol Biol 797: 127-140. 153. Lane, C, Leitch, J, Tan, X, Hadjati, J, Bramson, JL, and Wan, Y (2004). Vaccination-induced autoimmune vitiligo is a consequence of secondary trauma to the skin. Cancer Res 64: 1509-1514. 154. Smallwood, SE, Rahman, MM, Smith, DW, and McFadden, G (2010). Myxoma virus: propagation, purification, quantification, and storage. Curr Protoc Microbiol Chapter 14: Unit 14A.11. 155. Mould, RC, AuYeung, AWK, van Vloten, JP, Susta, L, Mutsaers, AJ, Petrik, JJ, et al. (2017). Enhancing Immune Responses to Cancer Vaccines Using Multi-Site Injections. Sci Rep 7: 8322. 156. Santry, LA, McAusland, TM, Susta, L, Wood, GA, Major, PP, Petrik, JJ, et al. (2017). Production and Purification of High Titer Newcastle Disease Virus for use in Preclinical Mouse Models of Cancer. Molecular Therapy - Methods & Clinical Development. 157. Hoffman, RM (2015). Application of GFP imaging in cancer. Lab Invest 95: 432- 452. 158. Van de Lest, CH, Versteeg, EM, Veerkamp, JH, and Van Kuppevelt, TH (1995). Elimination of autofluorescence in immunofluorescence microscopy with digital image processing. J Histochem Cytochem 43: 727-730. 159. Riss, TLM, Richard A; Niles, Andrew L; Duellman, Sarah; Benink, Helene A; Worzella, Tracey J; Minor, Lisa (2013). Cell Viability Assays. In: Cousens, NB, K; et al. (ed). Assay Guidance Manual. Eli Lilly & Company and the National Center for Advancing Translational Science: Bethesda, MD. 160. Le Boeuf, F, Selman, M, Son, HH, Bergeron, A, Chen, A, Tsang, J, et al. (2017). Oncolytic Maraba Virus MG1 as a Treatment for Sarcoma. Int J Cancer 141: 1257- 1264. 161. Brun, J, McManus, D, Lefebvre, C, Hu, K, Falls, T, Atkins, H, et al. (2010). Identification of genetically modified Maraba virus as an oncolytic rhabdovirus.
149
Molecular therapy : the journal of the American Society of Gene Therapy 18: 1440- 1449. 162. Pol, JG, Zhang, L, Bridle, BW, Stephenson, KB, Rességuier, J, Hanson, S, et al. (2014). Maraba virus as a potent oncolytic vaccine vector. Molecular therapy : the journal of the American Society of Gene Therapy 22: 420-429. 163. Hummel, J, Bienzle, D, Morrison, A, Cieplak, M, Stephenson, K, DeLay, J, et al. (2017). Maraba virus-vectored cancer vaccines represent a safe and novel therapeutic option for cats. Sci Rep 7: 15738. 164. Parato, KA, Breitbach, CJ, Le Boeuf, F, Wang, J, Storbeck, C, Ilkow, C, et al. (2012). The oncolytic poxvirus JX-594 selectively replicates in and destroys cancer cells driven by genetic pathways commonly activated in cancers. Molecular therapy : the journal of the American Society of Gene Therapy 20: 749-758. 165. Yaacov, B, Eliahoo, E, Elihaoo, E, Lazar, I, Ben-Shlomo, M, Greenbaum, I, et al. (2008). Selective oncolytic effect of an attenuated Newcastle disease virus (NDV-HUJ) in lung tumors. Cancer Gene Ther 15: 795-807. 166. Chow, YH, Alberti, A, Mura, M, Pretto, C, Murcia, P, Albritton, LM, et al. (2003). Transformation of rodent fibroblasts by the jaagsiekte sheep retrovirus envelope is receptor independent and does not require the surface domain. J Virol 77: 6341-6350. 167. Danilkovitch-Miagkova, A, Duh, FM, Kuzmin, I, Angeloni, D, Liu, SL, Miller, AD, et al. (2003). Hyaluronidase 2 negatively regulates RON receptor tyrosine kinase and mediates transformation of epithelial cells by jaagsiekte sheep retrovirus. Proc Natl Acad Sci U S A 100: 4580-4585. 168. Patel, S, Turner, PR, Stubberfield, C, Barry, E, Rohlff, CR, Stamps, A, et al. (2002). Hyaluronidase gene profiling and role of hyal-1 overexpression in an orthotopic model of prostate cancer. Int J Cancer 97: 416-424. 169. Sugimoto, J, and Schust, DJ (2009). Review: Human Endogenous Retroviruses and the Placenta. Reproductive Sciences 16: 1023-1033. 170. Steckbeck, JD, Kuhlmann, AS, and Montelaro, RC (2014). Structural and Functional Comparisons of Retroviral Envelope Protein C-Terminal Domains: Still Much to Learn. Viruses 6: 284-300. 171. Barker, CS, Pickel, J, Tainsky, M, and Hunter, E (1986). Molecular cloning of the Mason-Pfizer monkey virus genome: biological characterization of genome length clones and molecular comparisons to other retroviruses. Virology 153: 201-214. 172. Rassa, JC, Meyers, JL, Zhang, Y, Kudaravalli, R, and Ross, SR (2002). Murine retroviruses activate B cells via interaction with toll-like receptor 4. Proc Natl Acad Sci U S A 99: 2281-2286. 173. Wang, E, Obeng-Adjei, N, Ying, Q, Meertens, L, Dragic, T, Davey, RA, et al. (2008). Mouse mammary tumor virus uses mouse but not human transferrin receptor 1 to reach a low pH compartment and infect cells. Virology 381: 230-240.
150
174. Aydin, H, Cook, JD, and Lee, JE (2014). Crystal structures of beta- and gammaretrovirus fusion proteins reveal a role for electrostatic stapling in viral entry. J Virol 88: 143-153. 175. Martin, JL, Cao, S, Maldonado, JO, Zhang, W, and Mansky, LM (2016). Distinct Particle Morphologies Revealed through Comparative Parallel Analyses of Retrovirus- Like Particles. 176. Copreni, E, Palmieri, L, Castellani, S, and Conese, M (2010). A VSV-G Pseudotyped Last Generation Lentiviral Vector Mediates High Level and Persistent Gene Transfer in Models of Airway Epithelium In Vitro and In Vivo. Viruses 2: 1577- 1588. 177. Hastie, E, Cataldi, M, Marriott, I, and Grdzelishvili, VZ (2013). Understanding and altering cell tropism of vesicular stomatitis virus. Virus Res 176. 178. Watson, DJ, Kobinger, GP, Passini, MA, Wilson, JM, and Wolfe, JH (2002). Targeted transduction patterns in the mouse brain by lentivirus vectors pseudotyped with VSV, Ebola, Mokola, LCMV, or MuLV envelope proteins. Molecular therapy : the journal of the American Society of Gene Therapy 5: 528-537. 179. Kim, S, and Lim, K (2017). Stability of Retroviral Vectors Against Ultracentrifugation Is Determined by the Viral Internal Core and Envelope Proteins Used for Pseudotyping. Mol Cells 40: 339-345. 180. Yu, ZY, McKay, K, van Asperen, P, Zheng, M, Fleming, J, Ginn, SL, et al. (2007). Lentivirus vector-mediated gene transfer to the developing bronchiolar airway epithelium in the fetal lamb. J Gene Med 9: 429-439. 181. Finkelshtein, D, Werman, A, Novick, D, Barak, S, and Rubinstein, M (2013). LDL receptor and its family members serve as the cellular receptors for vesicular stomatitis virus. Proc Natl Acad Sci U S A 110: 7306-7311. 182. Lee, H, Song, JJ, Kim, E, Yun, CO, Choi, J, Lee, B, et al. (2001). Efficient gene transfer of VSV-G pseudotyped retroviral vector to human brain tumor. Gene Ther 8: 268-273. 183. Papanikolaou, E, Kontostathi, G, Drakopoulou, E, Georgomanoli, M, Stamateris, E, Vougas, K, et al. (2013). Characterization and comparative performance of lentiviral vector preparations concentrated by either one-step ultrafiltration or ultracentrifugation. Virus Res 175: 1-11. 184. Cronin, J, Zhang, XY, and Reiser, J (2005). Altering the Tropism of Lentiviral Vectors through Pseudotyping. Curr Gene Ther 5: 387-398. 185. Sinn, PL, Arias, AC, Brogden, KA, and McCray, PB (2008). Lentivirus Vector Can Be Readministered to Nasal Epithelia without Blocking Immune Responses▿��. J Virol 82: 10684-10692. 186. Babiuk, S, Parkyn, G, Copps, J, Larence, JE, Sabara, MI, Bowden, TR, et al. (2007). Evaluation of an ovine testis cell line (OA3.Ts) for propagation of capripoxvirus
151
isolates and development of an immunostaining technique for viral plaque visualization. Journal of veterinary diagnostic investigation : official publication of the American Association of Veterinary Laboratory Diagnosticians, Inc 19: 486-491. 187. He, Y, Zhang, Q, Wang, J, Zhou, M, Fu, M, and Xu, X (2017). Full-length genome sequence analysis of enzootic nasal tumor virus isolated from goats in China. Virol J 14. 188. Hida, D, Danielson, BT, Knudson, CB, and Knudson, W (2015). CD44 knock- down in bovine and human chondrocytes results in release of bound HYAL2. Matrix biology : journal of the International Society for Matrix Biology 48: 42-54. 189. Antonioli, E, Piccinato, CA, Nader, HB, Cohen, M, Goldberg, AC, and Ferretti, M (2015). Modulation of Hyaluronan Synthesis by the Interaction between Mesenchymal Stem Cells and Osteoarthritic Chondrocytes. Stem Cells Int 2015: 640218. 190. Bergeron, D, Barbeau, B, Léger, C, and Rassart, E (1995). Experimental bias in the evaluation of the cellular transient expression in DNA co-transfection experiments. Cell Mol Biol Res 41: 155-159. 191. Howcroft, TK, Kirshner, SL, and Singer, DS (1997). Measure of transient transfection efficiency using beta-galactosidase protein. Anal Biochem 244: 22-27. 192. Huszár, T, Mucsi, I, Terebessy, T, Masszi, A, Adamkó, S, Jeney, C, et al. (2001). The use of a second reporter plasmid as an internal standard to normalize luciferase activity in transient transfection experiments may lead to a systematic error. J Biotechnol 88: 251-258. 193. VandenDriessche, T, Thorrez, L, Naldini, L, Follenzi, A, Moons, L, Berneman, Z, et al. (2002). Lentiviral vectors containing the human immunodeficiency virus type-1 central polypurine tract can efficiently transduce nondividing hepatocytes and antigen- presenting cells in vivo. Blood 100: 813-822. 194. Zufferey, R, Donello, JE, Trono, D, and Hope, TJ (1999). Woodchuck hepatitis virus posttranscriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J Virol 73: 2886-2892. 195. Salmon, P, Kindler, V, Ducrey, O, Chapuis, B, Zubler, RH, and Trono, D (2000). High-level transgene expression in human hematopoietic progenitors and differentiated blood lineages after transduction with improved lentiviral vectors. Blood 96: 3392-3398. 196. Henke, W, Herdel, K, Jung, K, Schnorr, D, and Loening, SA (1997). Betaine improves the PCR amplification of GC-rich DNA sequences. Nucleic Acids Res 25: 3957-3958. 197. Jensen, MA, Fukushima, M, and Davis, RW (2010). DMSO and betaine greatly improve amplification of GC-rich constructs in de novo synthesis. PLoS One 5: e11024. 198. Bryksin, AV, and Matsumura, I (2010). Overlap extension PCR cloning: a simple and reliable way to create recombinant plasmids. BioTechniques 48: 463-465.
152
199. Stahl, E, Martin, TG, Praetorius, F, and Dietz, H (2014). Facile and scalable preparation of pure and dense DNA origami solutions. Angew Chem Int Ed Engl 53: 12735-12740. 200. Kutner, RH, Zhang, XY, and Reiser, J (2009). Production, concentration and titration of pseudotyped HIV-1-based lentiviral vectors. Nat Protoc 4: 495-505. 201. Cooper, AR, Patel, S, Senadheera, S, Plath, K, Kohn, DB, and Hollis, RP (2011). Highly Efficient Large-Scale Lentiviral Vector Concentration by Tandem Tangential Flow Filtration. J Virol Methods 177: 1-9. 202. Linnerth-Petrik, NM, Santry, LA, Petrik, JJ, and Wootton, SK (2014). Opposing functions of Akt isoforms in lung tumor initiation and progression. PLoS One 9: e94595. 203. Edelstein, M (2017). Vectors Used in Gene Therapy Trials. John Wiley and Sons Ltd: The Journal of Gene Medicine. 204. Escors, D, and Breckpot, K (2010). Lentiviral vectors in gene therapy: their current status and future potential. Archivum immunologiae et therapiae experimentalis 58: 107-119. 205. Connolly, JB (2002). Lentiviruses in gene therapy clinical research. Gene Ther 9: 1730-1734. 206. Castellani, S, and Conese, M (2010). Lentiviral vectors and cystic fibrosis gene therapy. Viruses 2: 395-412. 207. Morgan, RA, Dudley, ME, Wunderlich, JR, Hughes, MS, Yang, JC, Sherry, RM, et al. (2006). Cancer regression in patients after transfer of genetically engineered lymphocytes. Science 314: 126-129. 208. Anderson, WF, Blaese, RM, and Culver, K (1990). The ADA human gene therapy clinical protocol: Points to Consider response with clinical protocol, July 6, 1990. Human gene therapy 1: 331-362. 209. Blaese, RM, Culver, KW, Chang, L, Anderson, WF, Mullen, C, Nienhuis, A, et al. (1993). Treatment of severe combined immunodeficiency disease (SCID) due to adenosine deaminase deficiency with CD34+ selected autologous peripheral blood cells transduced with a human ADA gene. Amendment to clinical research project, Project 90-C-195, January 10, 1992. Human gene therapy 4: 521-527. 210. Cavazzana-Calvo, M, Hacein-Bey, S, de Saint Basile, G, Gross, F, Yvon, E, Nusbaum, P, et al. (2000). Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science 288: 669-672. 211. Gaspar, HB, Parsley, KL, Howe, S, King, D, Gilmour, KC, Sinclair, J, et al. (2004). Gene therapy of X-linked severe combined immunodeficiency by use of a pseudotyped gammaretroviral vector. Lancet 364: 2181-2187. 212. Centers for Disease, C (1981). Pneumocystis pneumonia--Los Angeles. MMWR Morbidity and mortality weekly report 30: 250-252.
153
213. Friedman, LF (1991). Gene Therapy Techniques. Curr Opin Biotechnol 2: 840- 844. 214. Kang, EM, and Malech, HL (2012). Gene therapy for chronic granulomatous disease. Methods in enzymology 507: 125-154. 215. Ryser, MF, Roesler, J, Gentsch, M, and Brenner, S (2007). Gene therapy for chronic granulomatous disease. http://dxdoiorg/101517/147125987121799. 216. Lewis, PF, and Emerman, M (1994). Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 68: 510-516. 217. Wilson, JM (2012). Adeno-associated Virus and Lentivirus Pseudotypes for Lung-directed Gene Therapy. http://dxdoiorg/101513/pats200409-041MS. 218. Hogan, BLM, Barkauskas, CE, Chapman, HA, Epstein, JA, Jain, R, Hsia, CCW, et al. (2014). Repair and regeneration of the respiratory system: complexity, plasticity, and mechanisms of lung stem cell function. Cell Stem Cell 15: 123-138. 219. Rawlins, EL, Okubo, T, Xue, Y, Brass, DM, Auten, RL, Hasegawa, H, et al. (2009). The role of Scgb1a1+ Clara cells in the long-term maintenance and repair of lung airway, but not alveolar, epithelium. Cell Stem Cell 4: 525-534. 220. Schambach, A, Zychlinski, D, Ehrnstroem, B, and Baum, C (2013). Biosafety Features of Lentiviral Vectors. Human gene therapy 24: 132-142. 221. Wanisch, K, and Yáñez-Muñoz, RJ (2009). Integration-deficient Lentiviral Vectors: A Slow Coming of Age. Molecular therapy : the journal of the American Society of Gene Therapy 17: 1316-1332. 222. Papayannakos, C, and Daniel, R (2013). Understanding lentiviral vector chromatin targeting: working to reduce insertional mutagenic potential for gene therapy. Gene Ther 20: 581-588. 223. Bukrinsky, MI, Haggerty, S, Dempsey, MP, Sharova, N, Adzhubel, A, Spitz, L, et al. (1993). A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 365: 666-669. 224. Naldini, L, Blomer, U, Gallay, P, Ory, D, Mulligan, R, Gage, FH, et al. (1996). In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272: 263-267. 225. Hematti, P, Hong, BK, Ferguson, C, Adler, R, Hanawa, H, Sellers, S, et al. (2004). Distinct Genomic Integration of MLV and SIV Vectors in Primate Hematopoietic Stem and Progenitor Cells. PLoS Biol 2. 226. Naldini, L, Blömer, U, Gage, FH, Trono, D, and Verma, IM (1996). Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci U S A 93: 11382-11388. 227. Heinzinger, NK, Bukrinsky, MI, Haggerty, SA, Ragland, AM, Kewalramani, V, Lee, MA, et al. (1994). The Vpr protein of human immunodeficiency virus type 1
154
influences nuclear localization of viral nucleic acids in nondividing host cells. Proc Natl Acad Sci U S A 91: 7311-7315. 228. Griesenbach, U, Geddes, DM, and Alton, EW (2004). Gene therapy for cystic fibrosis: an example for lung gene therapy. Gene Ther 11 Suppl 1: S43-50. 229. Zufferey, R, Dull, T, Mandel, RJ, Bukovsky, A, Quiroz, D, Naldini, L, et al. (1998). Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 72: 9873-9880. 230. al Yacoub, N, Romanowska, M, Haritonova, N, and Foerster, J (2007). Optimized production and concentration of lentiviral vectors containing large inserts. J Gene Med 9: 579-584. 231. Fassler, M, Weissberg, I, Levy, N, Diaz-Griffero, F, Monsonego, A, Friedman, A, et al. (2013). Preferential lentiviral targeting of astrocytes in the central nervous system. PLoS One 8: e76092. 232. Schlegel, R, Tralka, TS, Willingham, MC, and Pastan, I (1983). Inhibition of VSV binding and infectivity by phosphatidylserine: is phosphatidylserine a VSV-binding site? Cell 32: 639-646. 233. Fuller, S, von Bonsdorff, CH, and Simons, K (1984). Vesicular stomatitis virus infects and matures only through the basolateral surface of the polarized epithelial cell line, MDCK. Cell 38: 65-77. 234. Wang, G, Slepushkin, V, Zabner, J, Keshavjee, S, Johnston, JC, Sauter, SL, et al. (1999). Feline immunodeficiency virus vectors persistently transduce nondividing airway epithelia and correct the cystic fibrosis defect. J Clin Invest 104: R55-62. 235. Kang, Y, Stein, CS, Heth, JA, Sinn, PL, Penisten, AK, Staber, PD, et al. (2002). In Vivo Gene Transfer Using a Nonprimate Lentiviral Vector Pseudotyped with Ross River Virus Glycoproteins. J Virol 76: 9378-9388. 236. Sinn, PL, Hickey, MA, Staber, PD, Dylla, DE, Jeffers, SA, Davidson, BL, et al. (2003). Lentivirus Vectors Pseudotyped with Filoviral Envelope Glycoproteins Transduce Airway Epithelia from the Apical Surface Independently of Folate Receptor Alpha. 237. Amirache, F, Levy, C, Costa, C, Mangeot, PE, Torbett, BE, Wang, CX, et al. (2014). Mystery solved: VSV-G-LVs do not allow efficient gene transfer into unstimulated T cells, B cells, and HSCs because they lack the LDL receptor. Blood 123: 1422-1424. 238. Farrow, N, Miller, D, Cmielewski, P, Donnelley, M, Bright, R, and Parsons, DW (2013). Airway gene transfer in a non-human primate: Lentiviral gene expression in marmoset lungs. Sci Rep 3. 239. Johnson, LG, Olsen, JC, Naldini, L, and Boucher, RC (2000). Pseudotyped human lentiviral vector-mediated gene transfer to airway epithelia in vivo. Gene Ther 7: 568-574.
155
240. Sinn, PL, Shah, AJ, Donovan, MD, and McCray, PB, Jr. (2005). Viscoelastic gel formulations enhance airway epithelial gene transfer with viral vectors. Am J Respir Cell Mol Biol 32: 404-410. 241. Rejman, J, Di Gioia, S, Bragonzi, A, and Conese, M (2007). Pseudomonas aeruginosa infection destroys the barrier function of lung epithelium and enhances polyplex-mediated transfection. Human gene therapy 18: 642-652. 242. Courtney, JM, Ennis, M, and Elborn, JS (2004). Cytokines and inflammatory mediators in cystic fibrosis. J Cyst Fibros 3: 223-231. 243. Alton, E, Beekman, JM, Boyd, AC, Brand, J, Carlon, MS, Connolly, MM, et al. (2017). Preparation for a first-in-man lentivirus trial in patients with cystic fibrosis. Thorax 72: 137-147. 244. Mitomo, K, Griesenbach, U, Inoue, M, Somerton, L, Meng, C, Akiba, E, et al. (2010). Toward Gene Therapy for Cystic Fibrosis Using a Lentivirus Pseudotyped With Sendai Virus Envelopes. Molecular therapy : the journal of the American Society of Gene Therapy 18: 1173-1182. 245. Sinn, PL, Department of Pediatrics, PiGT, Carver College of Medicine, The University of Iowa, Iowa City, IA 52242, USA, Burnight, ER, Department of Pediatrics, PiGT, Carver College of Medicine, The University of Iowa, Iowa City, IA 52242, USA, Shen, H, Department of Pediatrics, PiGT, Carver College of Medicine, The University of Iowa, Iowa City, IA 52242, USA, et al. (2005). Inclusion of jaagsiekte sheep retrovirus proviral elements markedly increases lentivirus vector pseudotyping efficiency. Molecular Therapy 11: 460-469. 246. Liu, SL, Lerman, MI, and Miller, AD (2003). Putative Phosphatidylinositol 3- Kinase (PI3K) Binding Motifs in Ovine Betaretrovirus Env Proteins Are Not Essential for Rodent Fibroblast Transformation and PI3K/Akt Activation. Journal of Virology 77: 7924- 7935. 247. Geraerts, M, Michiels, M, Baekelandt, V, Debyser, Z, and Gijsbers, R (2005). Upscaling of lentiviral vector production by tangential flow filtration. J Gene Med 7: 1299-1310. 248. Liu, SL, Halbert, CL, and Miller, AD (2004). Jaagsiekte Sheep Retrovirus Envelope Efficiently Pseudotypes Human Immunodeficiency Virus Type 1-Based Lentiviral Vectors. J Virol 78: 2642-2647. 249. Mundle, ST, Giel-Moloney, M, Kleanthous, H, Pugachev, KV, and Anderson, SF (2015). Preparation of pure, high titer, pseudoinfectious Flavivirus particles by hollow fiber tangential flow filtration and anion exchange chromatography. Vaccine 33: 4255- 4260. 250. Liu, C, Wong, E, Miller, D, Smith, G, Anson, D, and Parsons, D (2010). Lentiviral airway gene transfer in lungs of mice and sheep: successes and challenges. J Gene Med 12: 647-658.
156
251. van Lieshout, LP, Domm, JM, Rindler, TN, Frost, KL, Sorensen, DL, Medina, SJ, et al. (2018). A Novel Triple-Mutant AAV6 Capsid Induces Rapid and Potent Transgene Expression in the Muscle and Respiratory Tract of Mice. Molecular Therapy - Methods & Clinical Development 9: 323-329. 252. Cribbs, AP, Kennedy, A, Gregory, B, and Brennan, FM (2013). Simplified production and concentration of lentiviral vectors to achieve high transduction in primary human T cells. BMC Biotechnol 13: 98. 253. Ichim, CV, and Wells, RA (2011). Generation of high-titer viral preparations by concentration using successive rounds of ultracentrifugation. Journal of translational medicine 9: 137. 254. Sullivan, HA, and Wickersham, IR (2015). Concentration and purification of rabies viral and lentiviral vectors. Cold Spring Harb Protoc 2015: 386-391. 255. Svardal, AM, and Pryme, IF (1980). The effect of "shearing" forces during ultracentrifugation on the membrane-bound polysomes associated with light rough microsomes of MPC-11 cells. Molecular biology reports 6: 105-110. 256. Jiang, W, Hua, R, Wei, M, Li, C, Qiu, Z, Yang, X, et al. (2015). An optimized method for high-titer lentivirus preparations without ultracentrifugation. Sci Rep 5. 257. Merten, OW, Hebben, M, and Bovolenta, C (2016). Production of lentiviral vectors. Mol Ther Methods Clin Dev 3: 16017-. 258. Sutton, TC, and Subbarao, K (2015). Development of animal models against emerging coronaviruses: From SARS to MERS coronavirus. Virology 479-480: 247-258. 259. Li, W, Moore, MJ, Vasilieva, N, Sui, J, Wong, SK, Berne, MA, et al. (2003). Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature 426: 450-454. 260. Milone, MC, and O'Doherty, U (2018). Clinical use of lentiviral vectors. Leukemia. 261. Santry, LA, Ingrao, JC, Yu, DL, de Jong, JG, van Lieshout, LP, Wood, GA, et al. (2017). AAV vector distribution in the mouse respiratory tract following four different methods of administration. BMC Biotechnol 17: 43. 262. Chubet, RG, and Brizzard, BL (1996). Vectors for expression and secretion of FLAG epitope-tagged proteins in mammalian cells. BioTechniques 20: 136-141. 263. Halbert, CL, Standaert, TA, Aitken, ML, Alexander, IE, Russell, DW, and Miller, AD (1997). Transduction by adeno-associated virus vectors in the rabbit airway: efficiency, persistence, and readministration. J Virol 71: 5932-5941. 264. Halbert, CL, Rutledge, EA, Allen, JM, Russell, DW, and Miller, AD (2000). Repeat transduction in the mouse lung by using adeno-associated virus vectors with different serotypes. J Virol 74: 1524-1532.
157
265. Clary-Meinesz, C, Mouroux, J, Cosson, J, Huitorel, P, and Blaive, B (1998). Influence of external pH on ciliary beat frequency in human bronchi and bronchioles. Eur Respir J 11: 330-333. 266. Tømmeraas, K, and Melander, C (2008). Kinetics of hyaluronan hydrolysis in acidic solution at various pH values. Biomacromolecules 9: 1535-1540. 267. Duterme, C, Mertens-Strijthagen, J, Tammi, M, and Flamion, B (2009). Two Novel Functions of Hyaluronidase-2 (Hyal2) Are Formation of the Glycocalyx and Control of CD44-ERM Interactions*. J Biol Chem 284: 33495-33508. 268. Bourguignon, LYW, Singleton, PA, Diedrich, F, Stern, R, and Gilad, E (2004). CD44 Interaction with Na+-H+ Exchanger (NHE1) Creates Acidic Microenvironments Leading to Hyaluronidase-2 and Cathepsin B Activation and Breast Tumor Cell Invasion. 269. England, RJ, Homer, JJ, Knight, LC, and Ell, SR (1999). Nasal pH measurement: a reliable and repeatable parameter. Clin Otolaryngol Allied Sci 24: 67-68. 270. Fischer, H, and Widdicombe, JH (2006). Mechanisms of Acid and Base Secretion by the Airway Epithelium. J Membr Biol 211: 139-150. 271. Washington, N, Steele, RJ, Jackson, SJ, Bush, D, Mason, J, Gill, DA, et al. (2000). Determination of baseline human nasal pH and the effect of intranasally administered buffers. International journal of pharmaceutics 198: 139-146. 272. Goldring Kirsty, MB, Culley, L., Wondimu, Elisabeth, Otero, Miguel (2017). Chapter 3 - Cartilage and Chondrocytes. In: Firestein, GS (ed). Kelley and Firestein's Textbook of Rheumatology 10 edn., vol. 1. pp 34-59.e33. 273. SR, W, KJ, S, PI, M, and SK, W (2014). Development of an ante-mortem diagnostic test for enzootic nasal tumor virus and detection of neutralizing antibodies in host serum. Journal of General Virology 95: 1843-1854. 274. Hsu, T, Phung, A, Choe, K, Kim, JW, and Fan, H (2015). Role for a Zinc Finger Protein (Zfp111) in Transformation of 208F Rat Fibroblasts by Jaagsiekte Sheep Retrovirus Envelope Protein. J Virol 89: 10453-10466. 275. Maeda, N, Inoshima, Y, Fruman, DA, Brachmann, SM, and Fan, H (2003). Transformation of Mouse Fibroblasts by Jaagsiekte Sheep Retrovirus Envelope Does Not Require Phosphatidylinositol 3-Kinase. J Virol 77: 9951-9959.
158