Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Adenoviral-mediated gene transfer of the human lipoprotein lipase gene Ashbourne Excoffon, Katherine J. D. 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2000-564983.pdf [ 16.07MB ]
JSON: 831-1.0089746.json
JSON-LD: 831-1.0089746-ld.json
RDF/XML (Pretty): 831-1.0089746-rdf.xml
RDF/JSON: 831-1.0089746-rdf.json
Turtle: 831-1.0089746-turtle.txt
N-Triples: 831-1.0089746-rdf-ntriples.txt
Original Record: 831-1.0089746-source.json
Full Text

Full Text

ADENOVIRAL-MEDIATED GENE TRANSFER OF THE H U M A N LIPOPROTEIN LIPASE GENE By KATHERINE J.D. ASHBOURNE EXCOFFON B.Sc, The University of Guelph, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES Genetics Programme We accept this thesis as conforming to the required standard T H E U N I V E P V & T M ( O T > B R I T I S H C O L U M B I A August 2000 © Katherine J.D. Ashbourne Excoffon, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of fYlfdjcxd 6fcJt/dfCz/(^^ ^K^rcuyn The University of British Columbia Vancouver, Canada Date AuCj'3/OQ DE-6 (2/88) Abstract Lipoprotein lipase is a pivotal enzyme involved in the metabolism o f circulating lipoproteins. Through the hydrolysis of the triglycerides (TG) in TG-r ich particles, free fatty acids and glycerol are available for uptake by muscle, for energy, or adipose for storage. Patients with complete and partial L P L deficiency exhibit marked hypertriglyceridemia and altered lipoprotein metabolism. Gene therapy to deliver functional L P L to the circulation has been proposed in order to reduce the clinical morbidity and atherogenic risk due to this metabolic disorder. In vitro analysis of first generation recombinant adenoviral (Ad) vectors expressing the human L P L gene from an R S V promoter or a C M V promoter ( A d - L P L ) revealed efficient infection and subsequent L P L expression in several cell lines. Delivery of A d - L P L ( R S V or C M V driven) to the liver of either normal mice or mice heterozygous for L P L deficiency results in an improved lipoprotein profile for at least 42 days and the correction of impaired fat load tolerance. Similarly, an A d incorporating a common polymorphism that deletes the last two amino acids o f L P L (Ser447Ter) results in an improved profile. Despite minimal hepatoxicity in these animals, the augmentation of L P L expression from natural sites, including muscle and adipose, was investigated using the C M V - d r i v e n A d - L P L vector. After intra-adipose or intra-muscular A d - L P L treatment via direct injection, expression within the adipose and muscle tissue compartments was significantly elevated until D14. Regardless of this over expression, no change in plasma L P L activity or T G levels was evident. Due to the neo-natal lethality associated with complete L P L deficiency in mice, a larger feline model of complete deficiency was explored. Complete correction in a cohort of homozygous L P L deficient cats was observed for 14 days before expression was extinguished, likely due to a vigorous immune response to both the adenovirus and the human L P L gene. ' In summary, the outcome effected by transferring the L P L gene to animal models, as well as somatic cells in tissue culture, have been defined for the first time. Evaluation of the systemic and tissue-specific efficiency, stability, potential toxicity and immunogenicty of adenoviral-mediated L P L gene expression has been delineated over the duration of these i i studies providing proof-of-principle towards L P L gene therapy. Although adenovirus is an efficient vector for gene delivery, future studies w i l l evaluate alternative vector systems that display longer term expression and less immunogenicity, such as the adeno-associated virus. i i i Table of Contents Page Abstract i i Table of Contents iv List o f Tables v i i i List o f Figures ix Lis t o f Abbreviations x i i i Acknowledgements xv Chapter 1: Introduction 1 1.1 Overview o f L P L biology/biochemistry 2 1.2 L P L and human disease 4 1.3 Animal models for L P L 6 1.3.1 Models of L P L Deficiency 6 Gene targeted knockout mouse 6 Cat 7 M i n k 9 The significance of L P L deficient animal models 10 1.3.2 L P L over-expression in mice 10 Global expression 10 Tissue specificover-expression 12 Muscle over-expression 13 Heart over-expression 15 Liver over-expression 16 1.3.3 L P L expression in other murine l ipid disease models 17 1.4 "Gene Therapy" 18 1.5 Gene transfer for L P L deficiency 18 1.5.1 Tissue specificity and regulation 19 Muscle 20 Adipose 21 Liver 22 Other tissues 23 1.5.2 Vector specificity 24 Adenovirus 24 Alternative vector systems 27 1.5.3 Promoter specificity 29 1.6 Objectives 30 Chapter 2: Methods and Materials 32 2.1 L P L c D N A s 33 2.2 Adenovirus construction 33 2.2.1 RSV-driven adenovirus 33 2.2.2 C M V - d r i v e n adenovirus 35 2.2.3 P C R analysis for replication competent adenovirus 36 iv 2.3 Tissue culture 36 2.3.1 Primary cell culture 36 2.3.2 Transformed cell culture 38 2.3.3 P-Galactosidase (LacZ) expression in cell culture 38 2.3.4 Anti-adenovirus neutralizing antibody assay 39 2.3.5 V L D L isolation and HepG2 cell V L D L - T G hydrolysis 39 2.4 Tissue/cell homogenates 39 2.5 Animal handling and injection techniques 40 2.5.1 Mouse handling and injection techniques 40 Murine viral intravenous infection 40 Murine viral intramuscular infection 40 Murine viral intra-adipose infection 41 2.5.2 Feline viral intravenous infection and sampling 41 2.5.3 Intravenous fat load test 43 Murine 43 Feline 43 2.6 Measurement of L P L activity and mass 43 2.7 Measurement o f lipids 45 2.8 Fast protein liquid chromatography ( F P L C ) 45 2.9 Measurement o f alkaline phosphatase activity 46 2.10 Sacrifice and histology 46 2.11 R N A Analysis 46 2.11.1 Northern analysis 47 2.11.2 R T - P C R 47 2.11.3 In situ hybridization (ISH) 48 2.12 Radiolabelled L ip id Analysis 48 2.12.1 Dissolution of oleate and V L D L isolation 48 2.12.2 V L D L isolation/injection/clearance/distribution 49 2.12.3 L i p i d isolation, chromatography, and measurement 49 Chapter 3: In vitro assessment of Adenovirus-LPL efficacy 51 3.1 Introduction 52 3.2 In vitro evaluation of A d - L P L efficacy 53 3.2.1 Pilot study on cell type versus L P L activity 53 3.2.2 Hepatocyte target 53 Efficiency of adenovirus-mediated gene transfer in 57 HepG2 cells Overexpression of ectopic L P L in HepG2 cells 57 Characterization of expressed human L P L activity 63 V L D L - T G hydrolysis by LPL-expressing HepG2 cells 63 3.2.3 Adipocyte target 65 In vitro 65 E x vivo 67 3.2.4 Myocyte and other cell targets 71 3.3 Discussion 73 v Chapter 4: In vivo assessment of Adenovirus-LPL efficacy after 82 intravenous, liver targeted delivery 4.1 Introduction 83 4.2 R S V promoter 84 4.2.1 A d - R S V - L P L gene transfer to heterozygous (+/-) L P L 85 deficient mice Expression levels 85 Dose-response analysis of A d - R S V - L P L 85 expression in C57B1/6 mice Hepatic expression of human L P L in vivo 87 Detection of human-specific L P L in mouse 90 post-heparin plasma L i p i d changes 90 Correction of hypertriglyceridemia in L P L 90 +/- mice Correction of impaired fat tolerance by hepatic 93 expression of h u L P L 4.2.2 Comparison of gene transfer to normal C D 1 mice versus 96 L P L +/- mice 4.2.3 A d - R S V - L P L gene transfer to human LPL-transgenic mice 96 L P L expression and plasma lipid levels 97 4.3 C M V promoter 100 4.3.1 A d - C M V - L P L gene transfer to heterozygous L P L deficient mice 100 4.3.2 A d - C M V - L P L gene transfer to LPL-transgenic mice 103 4.4 F F A clearance and V L D L clearance/distribution 105 4.4.1 F F A clearance: +/+ mice vs. +/- mice 109 4.4.2 V L D L clearance/distribution: A d - C M V - L P L vs. A P treated mice 111 4.5 Discussion 111 Chapter 5: In vivo assessment of Adeno-LPL efficacy via alternative 118 routes of delivery 5.1 Introduction 119 5.2 Adipose 119 5.2.1 L P L expression: plasma vs. tissue activity 120 5.2.2 A P plasma and tissue activity 128 5.3 Muscle 128 5.3.1 L P L expression: plasma vs. tissue activity 131 5.3.2 A P plasma and tissue activity 133 5.4 Discussion 133 Chapter 6: In vitro and in vivo assessment of the LPL Ser447Ter 138 alteration by Adeno-mediated gene transfer 6.1 Introduction 139 6.2 A d - C M V - 4 4 7 In vitro 141 6.3 A d - C M V - 4 4 7 In vivo 141 6.3.1 Dose response and time course 141 v i 6.3.2 Ad-CMV-447 at a reduced dose 146 6.3.3 Human associations 149 6.4 Discussion 153 Chapter 7: Adenovirus-mediated correction of feline LPL deficiency 158 7.1 Introduction 159 7.2 RSV promoter 159 7.2.1 Ad-LacZ pilot study in normal cats 160 7.2.2 Ad-RSV-LPL pilot in LPL +/- cats 160 Expression levels 163 Lipid changes 163 Toxicity 169 7.2.3 Pilot in LPL-/-cats 169 7.3 C M V promoter 169 7.3.1 Pilot in LPL+/+Cats 170 Expression levels and lipid changes 170 7.3.2 LPL-/-cats 170 LPL expression levels 170 Lipid changes 173 Toxicity 177 7.4 Discussion 184 Chapter 8: Conclusions and recommendations for further work 188 8.1 Summary of results 189 8.1.1 In vitro analysis 189 8.1.2 Hepatic target in vivo 190 8.1.3 Adipocyte and myocyte targets in vivo 190 8.1.4 LPL mutant analysis 191 8.1.5 Feline correction of complete LPL deficiency 191 8.2 Further investigations 192 References 195 vii List of Tables Page Table 1.1 Transgenic human L P L mouse models 11 Table 3.1 List o f cell culture lines and origins 56 Table 3.2 L P L activity and mass tissue based alterations after either 70 A d - R S V / C M V - L P L or LacZ infection of isolated murine fat pads Table 3.3 Screening of A d - R S V - L P L in vitro and in vivo 76 Table 4.1 Lipase activity analysis in liver tissue homogenates o f 89 L P L +/- mice after A d - R S V - L P L / L a c Z gene transfer Table 5.1 Endogenous L P L activity in muscle and adipose tissue 121 Table 5.2 Plasma L P L activity and triglyceride levels after intraVenous, 122 intra-adipose or intra-muscular A d - C M V - L P L injection Table 5.3 Histological evaluation of tissue after intravenous, 127 intra-adipose or intramuscular delivery of A d - C M V - L P L or A P Table 6.1 Dose response of A d - C M V - 4 4 7 in L P L + / - m i c e 143 Table 6.2 Pre- and post-heparin L P L activity, mass and l ipid analysis 144 seven days after A d - C M V - 4 4 7 administration Table 6.3 L i p i d changes three days after low dose A d - C M V - 4 4 7 150 administration to L P L +/- mice Table 6.4 L P L activity and mass in a human cohort of 447 carriers 152 Table 7.1 Dose response o f A d - R S V - L a c Z in cats 161 Table 7.2 Observed toxicity in cats treated with A d - R S V - L P L or L a c Z 165 Table 7.3 Plasma L P L activity and mass in normal cats treated with 171 A d - C M V - L P L or A P Table 7.4 Plasma L P L activity, mass, triglycerides and tissue L P L activity 172 in L P L -/- cats Table 7.5 Toxicity in L P L deficient cats 7 days after A d - C M V - L P L or 182 A P gene transfer v i i i List of Figures Page Figure 1.1 Genomic structure of the human L P L gene 3 Figure 1.2 Adenoviral genome 25 Figure 2.1 P C R screening for wi ld type adenovirus 37 Figure 2.2 Intra-adipose (IA) injection of adenovirus 42 Figure 3.1 Schematic diagram o f A d - R S V - L P L / L a c Z 54 Figure 3.2 Schematic diagram of A d - C M V - L P L / 4 4 7 / 1 9 4 / A P / L a c Z 55 Figure 3.3 A d - R S V - L a c Z infection of HepG2 cells 58 Figure 3.4 Northern analysis of A d - R S V - L P L infected HepG2 cells 59 Figure 3.5 M O I versus L P L activity and immunoreactive protein in HepG2 61 cells A ) 24 hours post infection or B) 48 hours post infection Figure 3.6 Time course of A d - R S V - L P L expression from HepG2 cells 62 Figure 3.7 Characterization of human L P L activity expressed from 64 HepG2 cells Figure 3.8 A ) V L D L - T G hydrolysis over time by LPL-expressing 66 or control HepG2 cells B) F P L C analysis of T G after a 4 hour incubation with LPL-expressing or control HepG2 cells Figure 3.9 A d - R S V - L a c Z infection of 3T3-L1 cells 68 Figure 3.10 L P L activity alterations after either A d - R S V - or C M V - L P L 69 infection of 3 T3-L1 cells Figure 3.11 A d - R S V - L a c Z infection of C 2 C i 2 cells 72 Figure 3.12 Northern analysis of A d - R S V - L P L infected C 2 C i 2 cells 74 Figure 3.13 Ad-RSV-LPL-media ted L P L activity across several cell lines 75 Figure 4.1 Dose-response analysis of A d - R S V - L P L expression in 86 C57B1/6 mice Figure 4.2 L P L in situ hybridization after A d - R S V - L P L or L a c Z gene 88 transfer in L P L +/- mice A ) A d - R S V - L a c Z infected mouse liver at day 7 B) A d - R S V - L P L infected mouse liver at day 7 C ) A d - R S V - L P L infected mouse liver at day 42 D) A d - R S V - L P L infected mouse liver at day 60 Figure 4.3 A ) Alteration in L P L activity after A d - R S V - L P L or L a c Z gene 91 transfer in L P L +/- mice B) Human-specific L P L activity and mass in A d - R S V - L P L treated L P L +/- mice Figure 4.4 L i p i d alterations after A d - R S V - L P L or LacZ gene transfer to 92 L P L +/- mice A ) Plasma triglyceride B) HDL-Cholesterol C) Total cholesterol ix Figure 4.5 FPLC analysis of lipoprotein composition after Ad-RSV-LPL 94 or LacZ gene transfer A) Untreated wild type mice B) Untreated LPL +/- mice C) Ad-RSV-LacZ treated +/- mice at day 7 D) Ad-RSV-LPL treated +/- mice at day 7 E) Ad-RSV-LacZ treated +/- mice at day 60 F) Ad-RSV-LPL treated +/- mice at day 60 Figure 4.6 Intravenous fat-load tolerance in wild type and LPL +/- mice 95 receiving either Ad-RSV-LPL or LacZ at day 7 Figure 4.7 LPL activity and mass changes in LPL transgenic mice 98 receiving Ad-RSV-LPL or LacZ A) Activity B) Human specific immmunoreactive mass Figure 4.8 Lipid alterations after Ad- RSV-LPL or LacZ gene transfer to 99 LPL transgenic mice A) Plasma triglyceride B) HDL-Cholesterol C) Total cholesterol Figure 4.9 LPL activity and mass changes in LPL+/-mice 101 receiving Ad-CMV-LPL or AP A) Activity B) Human specific immmunoreactive mass Figure 4.10 Plasma triglycerides in LPL +/- mice receiving Ad-CMV-LPL 102 or Ad-CMV-AP Figure 4.11 Plasma alkaline phosphotase activity in LPL +/- mice 104 receiving Ad-CMV-LPL or AP Figure 4.12 LPL activity (A) and mass (B) changes in LPL transgenic mice 106 receiving Ad-CMV-LPL or LacZ Figure 4.13 Plasma triglycerides in LPL transgenic mice receiving 107 Ad-CMV-LPL or AP Figure 4.14 Plasma alkaline phosphotase activity in LPL transgenic mice 108 receiving Ad-CMV-LPL or AP Figure 4.15 Free fatty acid clearance in normal and LPL+/- mice 110 Figure 4.16 VLDL clearance in LPL+/-mice receiving Ad-CMV-LPL or AP 112 A) Plasma 3 H counts B) Plasma 3 H counts as a percentage of radioactivity at 2 minutes Figure 5.1 Plasma human specific LPL immunoreactive mass after 123 IV, IA or IM delivery of Ad-CMV-LPL or AP Figure 5.2 RT-PCR on adipose, muscle and liver tissue after intravenous, 124 intra-adipose or intra-muscular delivery of Ad-CMV-LPL or AP Figure 5.3 LPL activity and mass in adipose tissue homogenates after 126 intra-adipose delivery of Ad-CMV-LPL or AP A) Human specific LPL immunoreactive mass B) LPL activity x Figure 5.4 Plasma alkaline phosphatase activity after intravenous, 129 intra-adipose or intramuscular delivery of Ad-CMV-AP A) Intra-adipose treatment B) Intra-muscular treatment C) Intravenous treatment Figure 5.5 Tissue alkaline phosphatase activity after intra-adipose 130 or intramuscular delivery of Ad-CMV-AP A) Intra-adipose treatment B) Intra-muscular treatment Figure 5.6 LPL activity (A) and mass (B) in muscle tissue homogenates 132 after intra-muscular delivery of Ad-CMV-LPL Figure 6.1 Ad-CMV-447 mediated LPL activity and mass in HepG2 cells 142 Figure 6.2 Time course of LPL activity (A) and mass (B) after 145 Ad-CMV-447 intravenous delivery to LPL +/- mice Figure 6.3 Time course of lipid changes after Ad-CMV-447 treatment 147 A) Triglyceride B) HDL-Cholesterol C) Total Cholesterol Figure 6.4 LPL mass and activity 5 days after low dose administration of 148 Ad-CMV-447 Figure 6.5 Time course of lipid changes after low dose administration of 151 Ad-CMV-447 Figure 6.6 Points of regulation for LPL 156 Figure 7.1 In situ hybridization on the livers of cats treated with „ 162 Ad-RSV-LPL or LacZ Figure 7.2 Plasma LPL activity in cats treated with Ad-RSV-LPL or LacZ 164 Figure 7.3 Neutralizing antibody reaction in LPL heterozygous cats after 166 Ad-RSV-LPL delivery Figure 7.4 Plasma triglyceride changes in cats treated with Ad-RSV-LPL 167 or LacZ Figure 7.5 Intravenous fat tolerance at baseline and in heterozygous 168 LPL deficient cats treated with Ad-RSV-LPL or LacZ Figure 7.6 RT-PCR analysis of various tissues at day 5 in cats treated with 174 Ad-CMV-LPL or AP Figure 7.7 Plasma LPL activity and mass changes in LPL deficient cats 175 treated with Ad-CMV-LPL or AP A) LPL activity B) LPL immunoreactive mass Figure 7.8 Plasma triglyceride changes in LPL deficient cats treated with 176 Ad-CMV-LPL or AP xi Figure 7.9 FPLC analysis of lipoprotein composition in LPL deficient 178 cats treated with Ad-CMV-LPL or AP A) LPL deficient cats pre-gene transfer B) LPL deficient cats 7 days after Ad-CMV-LPL gene transfer C) LPL deficient cats 7 days after Ad-CMV-LacZ gene transfer D) LPL deficient cats 28 days after Ad-CMV-LPL gene transfer Figure 7.10 Intravenous fat tolerance in LPL deficient cats 9 days after 179 Ad-CMV-LPL or AP gene transfer Figure 7.11 Neutralizing anti-LPL antibody response in LPL deficient cats 181 after Ad-CMV-LPL or AP gene transfer Figure 7.12 Oil red O staining in the liver, spleen and lungs of LPL deficient 183 cats 5 days after Ad-CMV-LPL or AP gene transfer A) Liver, Ad-CMV-AP, day 5 B) Spleen, Ad-CMV-AP, day 5 C) Lung, Ad-CMV-AP, day 5 D) Liver, Ad-CMV-LPL, day 5 E) Spleen, Ad-CMV-LPL, day 5 F) Lung, Ad-CMV-LPL, day 5 xii List of Abbreviations aa - amino acid AAV - adeno-associated virus Ab - antibody Ad - adenovirus AP - alkaline phosphatase Apo - apolipoprotein bLPL - bovine lipoprotein lipase CAD - coronary artery disease CAR - Coxsakievirus-adenovirus receptor CETP - cholesterol ester transfer protein eld - combined lipase deficiency CM - chylomicron CMV - cytomegalovirus ctl - control ELISA - enzyme-linked immunosorbant assay FFA - free fatty acid FH - familial hypercholesterolemia FPLC - fast protein liquid chromatography HDL-C - high density lipoprotein cholesterol HL - hepatic lipase hLPL - human lipoprotein lipase HSL - hormone sensitive lipase HSPG - heparan sulphate proteoglycan IA - intra-adipose IDL - intermediate density lipoprotein IM - intra-muscular IP - intra-peritoneal ISH - in situ hybridization IV - intravenous kb - kilobase LacZ - P-galactosidase LDL - low density lipoprotein LDL-R - low density lipoprotein receptor LPL - lipoprotein lipase LPL mass - lipoprotein lipase human specific immunoreactive mass LRP - low density lipoprotein receptor-related protein mAb - monoclonal antibody mCK - mouse creatine kinase MOI - multiplicity of infection mu - map units PBS - phosphate buffered saline PCR - polymerase chain reaction pfu - plaque forming unit PHP - post-heparin plasma xiii RSV - rous sarcoma virus RT-PCR - reverse transcription polymerase chain reaction TC - total cholesterol TG - triglyceride VLDL - very low density lipoprotein wk - week wt - wild type yr - year 447 - Ser447Ter LPL mutation +/+ - wild type/normal +/- - heterozygous -/- - homozygous deficient xiv Acknowledgements It was a privilege to work with Dr. Michael Hayden and I am grateful for the guidance, encouragement and support that he provided. I would also like to thank the members of my supervisory committee, Dr. Carolyn Brown, Dr. Kei th Humphries for all o f their advice and encouragement during my training. In particular I would like to thank my co-supervisor Dr. Suzanne Lewis for always being positive, encouraging and helping me focus on the important aspects of every project. There are many friends and colleagues that contributed greatly to this thesis and I would like to thank them for their support, patience and friendship. In particular I would like to thank Dr. Guoqing L i u for all of his advice and collaborative efforts, without which many of these exciting results may not have been possible. I would especially like to thank Suzanne Clee for being my best and most reliable friend and colleague in the lab. Special thanks to Nagat Bissada, L i and Fudan Miao who have been integral to the completion of this thesis both technically and emotionally. Thank you to Odell Loubser, Elizabeth Almqvist , Hope Kubryn, Andrew Martin, Roshni Singaraja, Kathy Kalvinou and Margaret Rasheed for taking time for me. A n d thank you to other friends in the Hayden lab, C M M T and West Point Grey Prebyterian Church for making the past 5 years enjoyable. Enormous thanks to my parents Howard and Mary Louise and my brother Robert, as well as all of my other brothers and sisters for their love and constant support. I would like to dedicate this thesis to my husband Simon Excoffon and to the glory o f God. Thank you Simon from the bottom of my heart for your love, support, technical assistance and patience throughout this whole experience. This thesis was supported in part by scholarships from N S E R C and U B C . xv Chapter 1: Introduction 1 1.1 Overview of lipoprotein lipase (LPL) biology/biochemistry LPL plays a pivotal role in the regulation of lipoprotein and lipid metabolism. The noncovalently-linked 55 kDA glycosylated homodimer is transported to the vascular endothelium by an unknown mechanism, where it binds heparan sulphate proteoglycans at the luminal surface {1}. Subsequent catabolism of triglycerides from both chylomicrons (CM) and very low density lipoproteins (VLDL) allows the uptake and utilization of the free fatty acids and glycerol for energy and storage in muscle and adipose tissue respectively {2}. Chylomicron and VLDL remnants are either shuttled into high density lipoprotein (HDL) {3} or low density lipoprotein (LDL) particle formation {4} respectively, or are taken up by the liver and repackaged into new VLDL particles. LPL has an obligatory requirement for its activator apolipoprotein (apo) CII, a small protein of 79 amino acids that is enriched on CM and VLDL particles {2}. Inhibitors of LPL include fatty acids, apo Oil, and possibly apo E. Another inhibitor is high concentrations of salt (IM NaCl). This latter inhibitor is very useful in biochemically distinguishing between LPL and hepatic lipase (HL), a close relative of LPL that is produced by the liver and is a hydrolase for phospholipids in HDL and intermediate density lipoproteins (IDL). Although the cellular origin of LPL in the circulation is unclear, and may represent an accumulation from several tissue sources, its site of action is at the luminal surface of the vascular endothelium. Due to its non-covalent interaction with heparan sulphate proteoglycans, LPL may be displaced into the plasma by an intravenous bolus injection of heparin. Thus, LPL activity and protein levels can be simply assessed by taking a small sample of post-heparin plasma (PHP). Aliquots of this PHP can then be used in either a synthetic, radiolabelled triglyceride (TG) assay for lipolytic activity or be measured by LPL specific antibodies for protein levels. Lipid measures are best performed in pre-heparin samples since the release of LPL will cause rapid lipolysis of the TG in the sample. Despite the ability to study the biochemical basis of this protein for many years, the protein sequence of human LPL remained elusive until 1987 when it was deduced from its cDNA nucleotide sequence {5}. Analysis of the genomic structure {6} revealed 10 exons spanning approximately 30kb, chromosomally localized to 8p22 {7}, producing a protein 475 amino acids long which becomes a mature protein of 448 residues after cleavage of a signal peptide (Figure 1.1). The first exon encodes a 5' untranslated region, the signal 2 o © OS 00 *o -i ^ 2 ITS C N 0* a * tS a a a a s J S ^ a o cu a « « i a W> fl c o X u <u D u, CX . CD t/3 CD CU o £> .3 o 5 Vi oj <D o "S ~ -s 00 O 60 >-. CU S a ? CD a c T3 00 CD <u c .5 cU ~ 00 U-i J ° OH g 11 3 »-> Xi o CD "cT 5 s O c 8 .2 S •a "S § s 00 oo o 2a t/J CD C C — o -a 3 peptide, and the first 2 amino acids of the mature protein while the tenth exon specifies a long 3' untranslated region. Exons 4, 5 and 6 form the catalytic site and are the locations of the majority of inactivating mutations potentially causing LPL deficiency {8}. 1.2 LPL and human disease Although complete LPL deficiency is not common, occurring in approximately 1 in 106 persons, the frequency is much higher in the French Canadian population where it occurs in 1 in 5000 {9}. The clinical manifestations of complete LPL deficiency in humans stem from infancy with a failure to thrive, colicky abdominal pain, hepatosplenomegaly, chylomicronemia characterized by lactescent plasma, eruptive xanthomata, lipemia retinalis and life threatening pancreatitis {2}. Lipid lowering drugs are ineffective and even rigid dietary restrictions are often poorly tolerated. The development of new therapies for LPL deficiency would represent a major advance for persons suffering from this disorder. Genetic LPL deficiency is generally classified into three categories depending on the LPL protein characteristics {10}. Type I hypertriglyceridemic patients have very low to absent LPL protein mass. Type II patients produce little pre-heparin LPL protein but the level increases following heparin treatment. In class Type III, there are large amounts of circulating pre-heparin protein with little change in levels after heparin challenge. Although the utility of this classification system is limited, with some compound heterozygote patients spanning two of these classes, clarification of the presence or absence of plasma LPL protein may be important for deciding which patients are most likely to tolerate gene transfer without an immune reaction to LPL itself. Recently, patients with mutations in the LPL gene which result in partial defects in LPL catalytic function have been identified and, in fact, are very common in the general population. Collectively, known mutations resulting in partial catalytic defects in LPL are now estimated to occur with a frequency of between 5-7% in the general population {11}. The clinical presentation may be quiescent, evident only by marginally elevated triglyceride levels in the non-stressed state, with profound hypertriglyceridemia triggered by factors such as normal pregnancy, obesity or diabetes {12}. Postprandial metabolic studies have been performed on individuals heterozygous for mutations in the LPL gene {13,14}, 4 demonstrating an unmasking of the lipolytic defect after a fat challenge, resulting in prolonged postprandial lipemia and significant disturbances in lipoprotein levels and composition. There is also some evidence that specific mutations that alter, but do not abolish, LPL activity, such as Asn291Ser, Asp9Asn or Ser447Ter, exist commonly in the general population {15}. The significance of this is not yet fully understood although they are implicated in atherosclerosis susceptibility. The Ser447Ter mutation is of particular interest since in many studies it has been associated with decreased TG and increased HDL - a very desirable lipoprotein profile that is associated with decreased atherosclerosis {15-21}. Correspondingly, in most studies this mutation seems to confer protection against coronary artery disease (CAD) {21}. Although the mechanism behind these effects has been mainly speculative, when considering genetic reconstitution for partial or complete LPL deficiency, it is interesting to consider using a more beneficial form of the LPL gene. Clearly, plasma lipoproteins represent one important risk factor for the development of CAD. The most obvious mechanism involving impaired LPL action is its impact upon lipoprotein concentrations and particle composition, accompanied by increased TG and reduced levels of high density lipoprotein cholesterol (HDL-C) in plasma. Lipoprotein profiles in humans homozygous {22} and heterozygous {23} for mutations in the LPL gene are consistent with an enhanced susceptibility toward atherosclerosis. In view of the high frequency of mutations in the LPL gene in the general population, it is suggested that LPL deficiency may engender a significant risk factor for hyperlipidemia and atherogenesis. Conversely, increases in LPL activity by several mechanisms, including studies of transgenic mice which over express LPL primarily in adipose, heart and skeletal muscle tissues, demonstrate a significantly improved lipid profile including a reduction in plasma TG levels and decreased total cholesterol (TC):HDL ratio {24-27}. Moreover, suppression of diet-induced atherosclerosis has been reported in both low-density lipoprotein receptor (LDL-R) and apolipoprotein E (apo E) knockout mice over-expressing LPL {28}. Similarly, administration of drugs (NO-1886 or fenofibrate) {29-32} have been found to promote the action of LPL, resulting in lowered plasma TG, improved capacity to handle lipid loads and increased HDL levels. The action of these drugs is generally to increase LPL at the transcriptional level {32}. Although benefits may be observed in heterozygotes, they offer 5 little relief to a person completely deficient for LPL activity. Additionally, poor absorption and liver toxicity of these drugs have deferred their general use in humans. Currently there is no alternative to a severely restricted dietary regime for patients deficient for LPL. Such a diet is poorly tolerated resulting in several hospital admissions yearly and potentially life threatening complications due to recurrent or chronic chylomicronemia. We hypothesized that the LPL gene successfully delivered and expressed by an adenoviral vector, may be useful in correcting the lipolytic defects due to LPL deficiency. 1.3 Animal models for LPL One of the most useful tools to support studies of somatic cell genetic reconstitution is an animal model mimicking the human disorder. Investigators interested in, studying LPL are fortunate to have three models of LPL deficiency. These include the heterozygous gene targeted knock-out murine model {33,34}, a homozygous deficient naturally occurring feline model {35-37} and a homozygous deficient naturally occurring mink model {38-40}. Animal models that indicate the clinical ramifications of gene over-expression are also of significance. One of the important questions that must be applied to any type of gene transfer approach is where to target the gene. Over-expression in one specific location may lead to unexpected advantages or perhaps toxicity. If the goal is to attain prolonged expression with no toxicity, transgenic models allow the evaluation of systemic versus tissue specific expression for the lifetime of an organism. There are currently at least 8 different types of LPL transgenic mouse models that have been bred onto various background strains. Some of these alternate backgrounds include other genetic alterations such as mutations in the LDL-R and apo E genes. 1.3.1 Models of LPL deficiency Gene targeted knockout mouse Two different, independently created, gene-targeted LPL knockout murine models were created in 1995. Weinstock et al {34} used the neomycin drug resistance gene to knockout LPL in exon 1 whereas Coleman et al {33} used the same technique to knockout LPL in exon 8. Unfortunately, complete LPL deficiency is lethal in both murine models. This is perhaps not surprising since another mouse model with combined lipase deficiency 6 (eld), manifesting a deficiency in both LPL and HL due to an error in post-translational processing, also is lethal {41}. LPL deficient mice survive only hours postnatally before suffering death presumably due to congestion of the peripheral and pulmonary circulation by large TG-rich lipoproteins. This hypothesis has recently been challenged by the creation of a transgenic mouse line expressing LPL within the liver {42}. When this line is used to rescue the LPL knockout line, the TG levels remain extremely high suggesting that the mortality may be related to a nutritional insufficiency rather than the concentration of TG-rich particles. Several transgenic lines expressing LPL in specific tissue compartments have been used to rescue the completely LPL deficient mice from death, generally maintaining a mild elevation of TG levels reminiscent of the heterozygous state {42-44}. Without a suitable rescue procedure limited to neonates, such as regulatable expression or transient gene transfer, the use of the murine model for studies of complete LPL deficiency is negated and emphasizes the importance of other larger models. However, mice heterozygous for LPL deficiency mimic the human condition and have an approximately two to three-fold elevation in TG levels associated with decreased post-heparin plasma (PHP) LPL activity levels and a lower tolerance to fat loading in comparison to their wild type (WT) litter mates. We adopted this model of heterozygous LPL deficiency for most of the preliminary experiments for evaluation the effect of adenoviral-mediated LPL gene transfer on hypertriglyceridemia, lipoprotein profile, and tolerance to intravenous fat loading, in vivo. Cat The feline model for LPL deficiency was first characterized by Jones et al in 1983 {45}. The cats originated from an inbred litter in a small town in New Zealand and initially presented with peripheral neuropathy, some hind limb paralysis and numerous subcutaneous nodules over the bony prominences. Upon further examination, fasting lipemia and lipemia retinalis were identified and a diagnosis of primary hyperlipoproteinaemia was made. Upon analysis of the lipoproteins in affected animals, the majority of cholesterol and TG were in the chylomicrons while in normal animals HDL and VLDL, respectively, are the major carriers. An intravenous injection of heparin resulted in the release of only approximately 7 0.5% of normal lipolytic activity, perhaps representing hepatic lipase activity. Further biochemical characterization confirmed that these cats have a defect in the catalytic activity of post-heparin plasma LPL {46}. The peripheral neuropathy appears to be caused by lipid granulomata causing lesions and compression in peripheral nerves {47}. Pathologically, hepatocytes, epithelial cells of the proximal convoluted tubule of the kidney and macrophages in the liver, spleen and lymph node contain large numbers of lipid vacuoles {48}. Additionally, several other changes were noted in the kidneys, especially in older animals, including chylomicron emboli within the glomerular capillaries and other changes similar to that seen in diabetes mellitus. It was suggested that these changes represent insults caused by the emboli. Molecular characterization revealed a missense mutation in exon 8 of the cat LPL gene at amino acid residue 412 (Gly->Arg, residue 409 in humans) {35}. This residue is highly conserved in all species in which LPL has been cloned to date and the paralogous residue is also conserved in the human HL gene and the canine pancreatic lipase gene. This residue is postulated to be involved in lipid binding and binding to the LDL-receptor related protein {35,49}. Overall, the nucleotide and predicted amino acid (aa) similarity of 90 and 94.5%, respectively, between human and cat is higher than that of other species in which the LPL gene has been sequenced {35}. One difference is a three amino acid insertion at the beginning of the predicted mature peptide resulting in a protein length of 451 aa versus 448 normally seen in man. The question of whether circulating levels of protein are present in this model is still debated. This is an important issue since host immune responses toward the LPL transgene will significantly hinder expression without immune modulation (i.e. tolerization). LPL transcription within the appropriate tissue compartments has been confirmed {35,37} and coupled with the mutation being a missense mutation, this suggests that translation of at least a portion of the cat LPL protein is likely and thus some form of tolerance to the endogenous protein should be evident. An early study done by Peritz et al showed equivalent amounts of immunoreactive protein mass could be detected in both pre- and post-heparin plasma of LPL deficient cats suggesting a failure to bind to the endothelium or Type III hyperlipidemia {46}. Studies in our lab have not been able to duplicate these results using a similar panel of LPL specific antibodies {35}. Additionally, gene transfer of the 8 human protein to both partial and complete deficient animals results in a strong antibody response which breaks tolerance and is reactive to the cat LPL as well as the human LPL (see Chapter 7). Transfer of the normal cat LPL gene may ameliorate this response but has not yet been examined. Thus the classification of the deficiency of these cats remains uncertain. This cat model serves as an excellent model for human deficiency in several respects. LPL deficiency in the cat mimics the human condition manifesting with primary chylomicronemia, peripheral subcutaneous xanthomata, lipemia retinalis, reduced body mass, reduced growth rates and an impaired tolerance to fat load {35,37}. Differences include the occasional peripheral neuropathy and a lack of obvious pancreatitis. Although there are significant differences in the lipid transport system, including a lack of cholesterol ester transfer protein (CETP) and the major cholesterol carrying particle is HDL rather than LDL (human), the cat is more similar to human than mouse to human since the HDL fraction includes distinct HDL2 and HDL3 subclasses {50}. A few other benefits of this model include large litters, breeding several times a year, large tissue and blood sample sizes and relative ease of handling. Mink In 1997, Christopherson et al first described an LPL deficient mink model {38}. These animals were initially discovered on a mink farm in Norway due to their chylomicronemia. Although this model is similar to the feline model it differs by suffering from pancreatitis in a manner apparently similar to the human syndrome. Additionally, it has trace amounts of LPL activity and normal amounts of LPL protein in their tissues. In 1998 the genetic basis for the deficiency was discovered as a missense mutation, Pro214Leu {39}. This amino acid is well conserved in all animals and likely is involved in the catalytic activity of the enzyme. One outstanding question is how homozygous LPL deficient patients clear the severe chylomicronemia without the aid of LPL. Although clearance is extremely retarded when compared to the normal or heterozygous condition, clearance of the large lipoproteins can happen within hours {51}. Biochemical investigations in the LPL deficient mink reveal that both the core constituents and the TG in large chylomicron particles are taken up by the liver 9 {40}. This contrasts with the normal situation where only a small fraction (<10%) of the TG follows the core proteins in the chylomicrons into the liver. Despite perhaps a greater similarity to the human disease, there are a few draw backs to working with the mink model including breeding capabilities which remain at one seasonal litter per year and a fierce temperament requiring trapping and tranquilization. In contrast, the feline model is able to be bred 2-3 times per year and may have similar litter sizes to the mink. Additionally, if handled correctly, the feline model is very tame and will cooperate in a multitude of experiments. The significance of LPL deficient animal models These models are of key importance to our further understanding the molecular processes of LPL regulation and activity at the molecular, cellular and biochemical level. Using a replication deficient adenovirus containing the LPL cDNA, the feasibility of gene transfer for the alleviation of complete or partial LPL deficiency was investigated. In addition, further elucidation of the role of complete or partial LPL deficiency in lipoprotein metabolism was also addressed. 1.3.2 LPL over-expression in mice The ability to add genes into the genome of rodents has provided many insights into many fields of research. Several transgenic mouse models have been generated to study the effects of enhanced LPL expression (Table 1.1). These models confirm the potential therapeutic TG-lowering benefit of increased LPL expression as well as indicating appropriate target sites and levels of expression. What they do not tell us are the consequences of LPL gene delivery and subsequent over-expression in an adult animal, especially if completely deficient for LPL activity prior to gene transfer. Global expression In the early 1990's, three different groups developed LPL transgenic mice using the cytomegalovirus (CMV) promoter {26}, the CMV immediate early enhancer/chicken p-actin promoter {27}or the mouse metallotheionein I promoter {52}. 10 'x <L> ."2 S 4> * I so 3 <L> > CJ —1 J Q-•J C (3 CJ < 60 c CL, X w I bH > O cn a> 3 cj cj C (L> CJ <L> e o Z T 3 oo LM <4H * ' —> TJ-CO E 00 eo "a. x in I O t. co Ji -c oo ca E 1 oo o a. T3 "§ O E o §1 a x W u. CJ > o "co O a > u 4) c o Z <N C N t. 5 -u o o ^ E E £ <0 -C •@ * "2 C — CO • * oo ,cj oo s» cj £ 2 S3 ON x x x s CN CN CN| £ —< CN CN ) C "i S 2 « E c O cd O -2. co <L> 3 E o S-< 0 -2 1 i •§ I o a. > .5 u < c co cu o on 3 E.S » 2 S -° o. c CN CN c > CO c/D a> G 5 N O , X w l -ej > o SH B a> o vi O „ OH •a .S E o ct E M ? © N C (Ll 0 o ss < -a D- L L . X "C U IJ u 1 O tn S3 oo » > 3 ^3 ° E o 2 < - Z VO OS <n m J 3 <— X CN —" Ji od u - oo -Vt 3 ( N in E — x C N CN i/T CO c S ® .5 w TR fa 3 o.SP 1 O m CN OS _ ON (L) C ti a OO OO g S S 52 13 3 « E "<3 C 00 U 3 X ^ X — n o <L> 3 E co oo NO as (Ll c •co e u 00 3 E is + 'lo e 3 S , CO ID <U OO OO 3 CO c o o + 6 a. 60 c co -G u O Z •a c 3 a 60 cj CO o o I c o o IO CJ 13 00 3 s ON in cj • r - u w CO .s CO <L> 3 E —i OH CJ C o z o in i o CO O L . CJ CO X) o OS I o 00 c*-i 00 ,C8 co CO 00 oi ^ ^ « « s s o. 0, _c .e <-<-<-<-X X X X — 00 o . — < SI 8' pq LH CJ > o Si CJ c o CJ m CJ L -.a s L . O a E o 00 -J >> "a o L H CJ > VO CO CJ I i E .§53 « §. C/3 CO H 11 Despite slightly differing expression patterns (see Table 1.1), all three groups revealed a moderate decrease in VLDL-TG levels to approximately 50% or less of wild type levels. In post-heparin plasma, LPL activity levels were increased by approximately 1.25 to 1.7 times the wild type level. Variable changes in LPL activity were observed in the tissues of the mice, largely reflecting the expression patterns of the transgene in the respective groups of mice. When these mice were challenged with a high sucrose or carbohydrate diet, which normally causes an increase in VLDL-TG, or a high cholesterol diet, the hypertriglyceridemia or hypercholesterolemia was suppressed {27,52}. Further studies on some of these mice show that over-expression through the CMV/actin promoter increases the activity of hormone-sensitive lipase (HSL) in the adipose tissue of mice {53}. HSL is responsible for the hydrolysis of the TG within the lipid droplets in adipocytes into glycerol and FFAs for release into the plasma. The importance of this finding was not described and may reflect the delicate balance of LPL between the tissues for normal lipoprotein metabolism. In another study, diabetes was induced using streptozotocin in both transgenic and non-transgenic mice {54}. The amount of hypertriglyceridemia and hypercholesterolemia was significantly suppressed in transgenic mice. This over-expression, however, did not alter the weight loss induced by diabetes demonstrating that although LPL is pivotal in lipid metabolism, there are many factors involved in diabetes induced weight alteration. Regardless of the promoter system employed and the additional stress induced (e.g. diet, diabetes), global over-expression of LPL in mice was able to reduce the TG levels with little to no toxicity. This indicates that a somatic gene transfer system capable of over-expressing LPL may be beneficial in hypertriglyceridemic individuals irrespective of the cause of hypertriglyceridemia. Tissue specific over-expression A few tissue specific LPL transgenic mouse models have been developed. These include muscle and/or heart specific transgenic mice and liver transgenic mice. In direct contrast to the minimal toxicity observed upon global over-expression of LPL, tissue 12 specific over-expression in mice, with the exception of the heart, is associated with toxicity to the tissue expressing the transgene. Muscle over-expression Three different groups have developed transgenic mice that over-express LPL specifically in the muscle tissue compartment using the mouse muscle creatine kinase (mCK) promoter {25,43,55-58}. Two of these groups used the normal wild type LPL gene while one group used an inactive version of LPL. In the first model, published by Levak-Frank et al (see Table 1.1), the mCK promoter allowed expression in both skeletal and cardiac muscle {25}. The effect of wild type LPL expression from these tissues is a reduction in plasma TG levels in proportion to the level of LPL over-expression, elevated FFA uptake into the muscle, weight loss and premature death. Additionally, one major feature of muscle based transgenesis is a development of severe myopathy in proportion to the level of LPL over-expression. This includes muscle fiber degeneration, glycogen storage, fiber atrophy and proliferation of mitochondria and peroxisomes likely due to the increase in FFA uptake. Although this induced myopathy could be compared to several human myopathies {56}, it may be most similar to the myopathy induced by clofibrate, a hypolipidemic drug which acts in part by upregulating LPL expression. Thus it was suggested that by altering the partitioning of FFA through the action of LPL within specific tissue compartments, various diseases may be modeled. Interestingly, the mice described by Jensen et al {57} over-expressed LPL in the skeletal muscle and not the cardiac muscle. They only described their low expressors that had a 2-14 fold increase in LPL activity in the muscle compartment. This level of over-expression was associated with a small decrease in plasma TG, elevated glucose levels and a decrease in carcass lipid content in males only. Plasma TG and FFA levels were also decreased after either a high fat or high carbohydrate diet while insulin and glucose were both increased. Additionally, muscle based LPL over-expression protected against lipid accumulation normally induced by a high fat diet. Interestingly, no histology or toxicity was described in this paper. When crossed to heterozygous LPL knockout mice, the muscle transgenic mice first described by Levak-Frank et al {43} were able to rescue the lethality of complete LPL 13 deficiency. For unexplained reasons, the mice developed by Jensen et al were not able to rescue complete LPL deficiency (personal communication). This may have been due to factors such as the site of integration (i.e. closely linked to the LPL gene) or due to the distribution of expression (i.e. over-expression only in the skeletal muscle and not the heart). When on the LPL knockout background, LPL expression in the muscle was elevated but non-existent in the adipose compartment. Surprisingly, post-heparin plasma LPL activities were equivalent to levels found in wild type mice however plasma TG levels were decreased. HDL was also reduced. These reductions were likely due to an increased fractional catabolic rate. Studies on the adipose tissue compartment revealed that although these mice had normal growth curves and normal amounts of adipose, the composition of the adipose was significantly altered due to up-regulation of de novo fatty acid synthesis {59}. The third muscle transgenic mouse model was created using a mutant version of the human LPL gene under the control of the mCK promoter. The mutation Aspl56Asn, an amino acid in the LPL catalytic site triad, produces enzymatically inactive hLPL that is secreted normally {8}. This model was created to further test the bridging effect of LPL in vivo. Expression was only detected in the skeletal muscle and to a lesser extent in the heart. Mutant hLPL expression led to an increase of approximately 3,100-3,500 ng/ml of homodimeric LPL in the post-heparin plasma of these mice with no associated catalytic activity. Only 5ng/ml of this hLPL was found in the pre-heparin plasma suggesting that binding of the mutant hLPL to endogenous proteoglycans was not affected. These mice were crossed to heterozygous LPL knockout mice and could not rescue homozygous deficiency. However, on the heterozygous background, muscle expression of the catalytically inactive mutant was able to reduce TG by 20-30% in the VLDL and LDL fractions. Cholesterol was also significantly reduced in the VLDL and HDL fractions. This level of muscle based over-expression was again associated with an increase in FFA uptake into the muscle and toxicity (mitochondriopathy, glycogen accumulation). These results suggest that the LPL protein level, irrespective of activity, can have a significant effect on lipoprotein metabolism and the resulting profile. 14 The muscle appears to be a successful site for increasing LPL expression and decreasing TG levels but the associated toxicity may preclude exclusively targeting this site for somatic gene transfer. Heart over-expression Using 8kb of the 5' flanking region of the normal mouse LPL gene, fused to a human LPL minigene, hLPL transgenic mice were developed {44}. Surprisingly, these mice expressed hLPL predominantly in the cardiac muscle (CM) with no observable hLPL mRNA or enzyme activity in either the smooth muscle or adipose tissues of these mice. LPL activity in the CM was responsive to fasting (8hr) and feeding with increased LPL activities of 130% and 20% respectively in comparison to wild type (wt) mice while activities in the smooth muscle and adipose were similar to wt mice. Post-heparin plasma activity was increased by 10% (fasted) and 180% (fed). Correspondingly, TG levels were decreased 50-60%. Crossing these mice to heterozygous LPL knockout mice was sufficient to rescue the neonatal lethality due to complete deficiency. This also produced a model to investigate the partitioning of energy without endogenous expression in the adipose and muscle tissue compartments. In comparison to wt mice, LPL activity in the CM was decreased 34% (fasted) and 41% (fed). No activity was detected in the smooth muscle or adipose tissue of these rescued mice. Post-heparin plasma LPL activity was decreased 67% (fasted) or 18% (fed). Although 30hrs after birth, there was a 10-fold increase in plasma TG levels, surprisingly, after weaning TG levels were decreased 18-28%. Additionally, the growth curve and body-mass index of these rescued knockout animals was identical to wt mice. Morphological and histological examination did not reveal any significant changes in the cardiac, smooth muscle or adipose tissues. Upon fatty acid composition analysis, it was found that there was a decreased polyunsaturated fatty acid (PUFA) concentrations in smooth muscle and adipose. This shift was balanced by increases in saturated fatty acids that can be endogenously synthesized indicating that the import of fatty acids from the plasma may be severely hampered. 15 The lack of toxicity in these mice as well as the potent TG lowering effect of the over-expression of LPL inNcardiac muscle suggests that this may be an excellent tissue target for somatic gene transfer. Liver over-expression Recently transgenic mice that express human LPL in the liver were developed using the portion of the human apo AI gene promoter that had previously been shown to confer liver specific expression {42}. Normally LPL is only found in the liver during the perinatal period after which expression only recurs in response to toxicity or regeneration (e.g. during cancer). On a normal background, liver over-expression was able to reduce plasma TG by 8-20% and no gross or histological differences between the livers of normal and transgenic animals were apparent. When crossed to the heterozygous LPL knockout mice, expression in the liver was able to rescue neonatal LPL deficiency. Surprisingly these mice were able to survive to weaning despite severe cachexia during the suckling period, characterized by massive hyperlipidemia and reduced weight gain. At 18hrs after birth there were increased plasma ketones, glucose and excessive intracellular lipid droplets within the liver. Based on these results it was suggested that perhaps the mortality associated with complete deficiency was due to a nutritional deficiency rather than the hypertriglyceridemia which originally was proposed to induce pulmonary capillary obstruction and microinfarcts in other organs leading to death in these animals {33,34}. Mice that survived to weaning were able to catch up in weight on a high carbohydrate diet and survive for greater than 4 months. Expression exclusively in the liver resulted in a slower VLDL turnover but greater VLDL mass clearance, increased VLDL-TG production and 3-4 fold elevation of plasma ketones. This suggests that VLDL in these mice enter a futile cycle of uptake and production by the liver along with increased ketone production. During cachexia this may prove to spare glucose which may help sustain brain function during periods of metabolic stress. In light of these results, liver based LPL expression may cause toxicity to a completely deficient patient with the majority of the lipid burden being trapped in the liver. However, in people partially deficient for LPL experiencing a crisis situation or predisposed 16 to heart disease due to elevated TG levels, hepatically directed gene transfer may be a suitable approach toward LPL over-expression. 1.3.3 LPL expression in other murine lipid disease models To further investigate the role of LPL in lipid metabolism and atherosclerosis, LPL transgenic mice have been crossed onto either the apo E or the LDL-R null backgrounds {28,60}. Both of these models have well documented atherosclerosis profiles in mice and in both models, the over-expression of LPL was able to significantly suppress the hypertriglyceridemia. In the LDL-R knock-out mice it was also able to reduce the hypercholesterolemia. The result was significantly reduced atherosclerosis revealing that the protective effect of LPL is mainly through an altered lipoprotein profile. One additional study, done in our lab, has revealed that the site of LPL expression is important in predicting the atherogenic effect of LPL {61}. Mice on the apoE null background were crossed to either LPL transgenic mice or LPL knock-out mice. In comparison to apoE null mice with wild type levels of LPL activity, atherosclerosis was decreased in both heterozygous and transgenic LPL mice. Interestingly, no difference in lesions was observed if mice were on wild type or LDL-R knock-out background {61,62}. This may be due to the variation in lesion size when mice are fed an atherogenic diet versus the spontaneous lesions observed in apoE null mice. In consideration of the apoE model results and as has been observed in other studies {63}, it strongly suggested that expression in macrophages versus plasma (derived from parenchymal tissues) might be a decisive factor towards the pro- or anti-atherogenicity of LPL expression. Although the mouse is an excellent experimental model for studies such as these, differences between human and mouse lipid metabolism (e.g. no CETP) preclude direct comparison. However, despite these differences, when considering the over-expression of LPL for either partial or complete LPL deficient patients, these studies indicate that the site of expression may have significant ramifications. 17 1.4 "Gene Therapy" The goal of gene therapy is to use genetic material to correct any disease stemming from a genetic deficiency or alteration. The disease may be inborn (e.g. LPL deficiency) or acquired (e.g. cancer) and may be caused by a defect or mutation in a single gene or several genes. The major issues surrounding implementation of gene therapy differ slightly depending on the type of disease being treated. In the case of a recessive genetic disease, such as LPL deficiency, the goal is to have long term, stable, regulated expression. Obstacles include the immune response to both the vector and the gene product itself, the stability of the transferred genetic material, the tissue targeting methodology (including vectorology), and the regulation of the genetic material once it is where you want it to be. Despite these challenges, this may be the best approach towards a long term treatment regimen for LPL deficiency since enzyme replacement strategies are not viable due to the rapid degradation of LPL after intravenous injection {64}. For example, in our feline model of complete LPL deficiency the half-life of a bolus injection of bovine LPL is approximately 10 minutes {37}. Although this may reduce the TG levels significantly, the transient presence of the protein would not sustain a decrease beyond the next period of feeding. Many different vectors, both viral and non-viral based, are being studied for the purpose of gene therapy. Advantages and disadvantages have been well documented for each mode of transfer with respect to transfer efficiency, episomal uptake versus genomic integration, vector mediated immunogenicity or toxicity and stability of expression {65,66}. The adenovirus possesses many positive features for gene transfer that are intrinsic to its natural biology including a wide host range, infectivity of quiescent cells, high titers, large insert capacity and ease of manipulation {67}. These features will be expanded upon in a subsequent section. 1.5 Gene transfer for LPL deficiency LPL deficiency is generally considered a recessive disease although partial deficiency is also associated with a phenotype. For either partial or complete deficiency, an overall increase in LPL gene expression is desired. The amount of expression required for the correction of the clinical features of complete deficiency is still debated but may require as little as 10-20% of normal (~200 mU/ml) {42,68}. Optimally, achieving levels greater 18 than those observed in normal subjects may have the additional advantage of preventing repercussions such as coronary artery disease (CAD). The major caveat to attaining this level with little to no toxicity is to choose appropriate expression site(s). If a high level of expression is desired, wide spread expression may be preferable to local expression. The optimal situation would likely be naturally regulated expression from at least the major naturally expressing tissues (cardiac and skeletal muscle, adipose) with lesser amounts from other natural sites of expression. However, the probability of an immune response against the LPL protein itself is an additional concern that cannot be overlooked when contemplating the treatment of complete deficiency. 1.5.1 Tissue specificity and regulation LPL is pivotal in lipoprotein metabolism and the partitioning of energy within the body. It is thus understandable that this enzyme would be under strict regulation {69}. At the fetal and neonatal stage, LPL mRNA can be detected in most tissues including adipose, muscle, lung, brain, kidney, spleen and liver. However, shortly after birth these levels change substantially. Expression increases dramatically in the adipose (brown and white) and, muscle (skeletal and cardiac), while it remains similar in the lung, kidney and spleen and extinguishes in the liver. Expression is also seen in the ovary, macrophages and at extremely high levels in the lactating mammary tissue. Alterations and endogenous levels of LPL expression in the plasma and in some of these tissues remains a matter of debate and may be a matter of species or even strain specific differences in regulation. For example, macrophage LPL secretion and mRNA is 2-3 fold higher in the atherosclerosis susceptible C57BL/6J mice versus the resistant C3H/HeN and A/J mice {63}. Indeed, this difference in expression level has been suggested to be a major cause of the susceptibility of the C57BL/6J mice. After maturation, LPL remains under strict regulation and tissue compartment levels can change rapidly. For example, LPL activity can increase or decrease dramatically in various tissues depending on the energy requirements of the individual. During periods of feeding, LPL activity surges in the adipose tissue and is reduced in the muscle. This is to allow the partitioning of TG into the adipose tissue for storage. During periods of fasting the inverse occurs with LPL activity increasing within the muscle compartments and 19 decreasing in the adipose. LPL also changes with many other physiological factors such as exercise, obesity and lactation {2,69}. Generally these changes are under the control of hormones such as insulin {69-72}. Insulin increases LPL activity by promoting both synthesis and secretion as well as suppressing fat mobilization from adipose tissue. LPL is also thought to be transcriptionally and post-translationally regulated by FFA which likely acts indirectly through the peroxisomal proliferation response element (PPRE) {73-76}. One very important cytokine involved in LPL regulation is the tumor necrosis factor (TNF) {77,78}. LPL and TNF levels are inversely correlated and may play a major role in cancer induced cachexia when TNF levels increase and LPL levels diminish or obesity when TNF in the adipose is low and LPL is high {77}. As a potential "gatekeeper" for the distribution of FFA between tissues such as adipose and muscle, it has long been thought that this enzyme is responsible, at least in part, for the body type of an individual. If a particular set point in a tissue is proportionately higher or lower than another tissue, this may cause an obese or lean phenotype. When considering targets for LPL gene transfer, it is important to consider the ramifications of altering this set point and redistributing lipid partitioning within the body potentially leading to obesity, weight loss or even cachexia. Muscle Throughout development and adulthood, both skeletal and, to a lesser extent, cardiac muscle always appears to have high levels of expression. These levels are highly regulated but not as tightly as seen in the adipose tissue. As described above, there have been a few successful muscle transgenic models created resulting in elevated tissue LPL levels and a reduction in plasma TG. Originally it was hypothesized that over-expression in this compartment may create a leaner phenotype since muscle is a primary site for P-oxidation, the break down of fat for energy. What was observed was a significant myopathy in proportion to the level of over-expression {25}. One of the difficulties with this tissue, from the perspective of gene transfer, is the efficiency of infection with Ad based vectors. Most studies, including work done in this thesis, have shown that somewhere between 2-15% of the myofibers can be infected with a single injection of adenovirus {79-81}. In mice, this infectability goes up if the mouse is 20 between the ages of 3-5 wks or if there is damage to the muscle prior to Ad application {80}. The levels of expression still do not approach that seen in other tissues such as the liver. Using alternative vector systems such as the adeno-associated virus (AAV), which is capable of infecting muscle at a very high rate {82,83}, the true utility of muscle based adult somatic gene transfer may be evaluated. One consideration to pay close attention to is the potential myopathy, especially if only part of the muscle is to be targeted. Adipose The adipose compartment is an interesting target from the perspective of gene transfer studies. LPL within this compartment appears to be more highly regulated than any other tissue and thus has been the source of many studies both in vitro and in vivo {74,84-87}. This is likely due to the fact that adipose is the storage depot for any excess fat stores and the overall nutrition of the body is dependent on the appropriate regulation of these stores. One of the early signs of adipocyte differentiation and maturation is the expression of LPL, which apparently remains high but tightly (post-translationally) regulated for the life of the cell {84,86}. During periods of feeding intracellular stores are rapidly secreted while during fasting, by unknown mechanisms, these stores are not released. Thus LPL is pivotal in the etiology of body composition, obesity and weight loss {77}. Since LPL plays such a large part in the life and growth of an adipocyte, one may hypothesize that although over-expression of LPL within the cell can be achieved if using a constitutive promoter, secretion of this newly synthesized LPL may not be achieved. This may cause toxicity to the cell due to excessive levels of a protein that must be tightly controlled. In this situation LPL may act as a toxin and cause the destruction of the adipocyte thus making the patient to lose adipose mass. However, if over-expression leads to disregulated secretion, since this tissue is able to accommodate large influxes of FFA, this may cause an obese phenotype due to the high local levels of LPL. In vitro, adipocytes (both immature and mature) have proven difficult to transfect. The adenovirus has shown little exception to this rule. This lack of transfection ability has been demonstrated in vivo as well. Thus the adipose tissue compartment has not been 21 extremely well studied for gene transfer with the majority of studies targeting breast cancer and revealing an approximate 5% transfection rate of cells in the direct vicinity of injection {88}. Although this may be a sufficient number of modified cells for growth factors, alternative vector systems capable of transfecting this tissue compartment at a higher rate may provide a more conclusive answer to this targeting question for LPL. Liver Once past the neonatal stage, LPL expression in the liver is only seen in situations associated with illness or regeneration (i.e. elevated cytokines, cachexia {42,89,90}). This is likely mediated by specific regulatory features of the LPL promoter rather than cellular differences in the processing of the LPL gene, transcript and/or protein. Inherently, this presents one of the significant caveats to targeting the liver. The fact that the liver is the natural site of LPL degradation as well as VLDL production in the adult may present additional problems. LPL must be exposed to the plasma compartment where it can encounter its lipoprotein substrates for triglyceride hydrolysis to ensue. If expressed primarily in a tissue that normally breaks it down, expression levels may need to be higher than in other tissues. This may be of greater significance than transcriptional regulation because LPL, as is shown herein, can be successfully expressed at very high levels within the liver resulting in a significant change in TG-rich lipoprotein levels in the plasma. Hypothetically, LPL at the hepatic endothelium could rapidly hydrolyze the TG in any VLDL and chylomicron particles passing through the liver, locally raising the concentration and uptake of FFA into the liver. The presence of excess LPL may augment bridging and uptake of the resulting remnant particles, increasing the lipid content of the liver even further. After the remnants are broken down and resynthesized into new VLDL particles, they would be secreted back into the plasma where the local concentration of LPL may be high enough to trap the new VLDL and its TG within the liver creating a futile cycle of VLDL production and uptake. This could potentially result in toxicity due to the disrupted nutrition of the liver and malnutrition of other tissues especially in completely deficient individuals as shown by Merkel et al in the LPL liver-expressing transgenic mice {42}. 22 However, it is now known that LPL expressed from this site is active and able to participate in lipoprotein metabolism. If expressed at low levels, this may in fact be an excellent target site. Other tissues Since LPL expression has been observed in several other tissues, the issue of targeting some of these sites should be considered. Many gene transfer studies to the lung have been performed utilizing the adenovirus as well as other vectors {91}. The major problems associated with this site are the difficulty to infect these cells, the regenerative capacity of these cells causing a high rate of transgene loss, and rapid inflammation. However, from the perspective of LPL expression this in fact may be a good site if it could be achieved at a high rate since the secreted LPL would likely readily have access to the lipoproteins within the blood. Again a major potential concern would be fatty infiltration within this tissue leading to toxicity. The kidney has not been well studied for gene transfer {92} and might be considered too small an organ for a high level of gene expression, again leading to concerns of fatty infiltration. In fact, this is often a major site of organ failure in the oldest of our LPL deficient cats due to fatty infiltration (unpublished observations). Additionally, people with renal complications often have abnormal LPL expression in the kidney {93}. Thus it is not likely a desirable site for increasing the lipid burden. In light of transgenic studies in mice, solely targeting the heart appears to be a viable site for LPL over-expression {44}. Despite its small size, this site would have excellent access to lipoproteins and potentially a large effect on lipoprotein metabolism. However, expression from this site would have to be carefully directed and regulated since targeting the vasculature in addition to the heart, or the vasculature alone, could lead to increased atherosclerosis - definitely an undesirable side effect to LPL over-expression. The above sites, as well as several others, have proven to be relatively difficult to target and confine expression within. The major drawback to over-expression in any tissue is the threat of toxicity due to FFA.influx and lipid accumulation. 23 1.5.2 Vector specificity Considering the tight and specific regulation of LPL, we initially hypothesized that the type of vector system employed for gene transfer should have a natural affinity to tissues other than the liver. Early experiments, involving a retroviral vector, revealed that another major concern was the expression level. At the commencement of this thesis, the adenovirus was becoming an increasingly utilized gene transfer tool due to its many beneficial characteristics. The caveat to using the adenovirus was that its best target site is the liver. Despite this issue, the high infection and expression levels within both cycling and non-cycling cells persuaded us to investigate the feasibility of adeno-mediated gene transfer. Adenovirus Adenoviral gene transfer has been studied in several disorders of lipoprotein metabolism including familial hypercholesterolemia (FH) {94-96}, and hepatic lipase deficiency {97}, as well as achieving physiological, or even greater, levels of apolipoproteins, such as apoAI {98} and apoE {99}, in the normal or deficient state respectively, in an attempt to modulate atherogenic risk (reviews {100-102}. The adenovirus is a non-enveloped, intermediate sized icosahedron, about 70nm in diameter, composed solely of protein and DNA. Serotypes from many species ranging from avian to human, have been isolated and include more than 50 different human serotypes. Human serotypes 2, 5, and 12 have been most extensively characterized and have served as valuable tools in the study of the molecular biology of DNA replication, transcription, RNA processing, and protein synthesis in mammalian cells {67}. The adenoviral genome consists of approximately 36kb of double stranded DNA that allows stable insertion and maintenance of foreign genes through successive rounds of viral replication. Deletions which allow the insertion of foreign genes and a genome size permissive for subsequent packaging have been made primarily in the El and E3 early gene regions (see Figure 1.2). The El genes are required for normal viral replication and are transcomplemented by growing the virus in human kidney 293 cells. The E3 region primarily functions in host immune resistance and is not essential. The maximal genomic allowance for successful encapsidation into virions is approximately 105% of the wild type 24 so o 00 (1 o o c o o 3 ' a u SO fl « u H es -J • • e CN O CN •t S © S ^ T3 ^° 5 ^ . H ll S3 00 O 0 2 CJ S o g g in co "—' 1 s o 8 c 5 • cj ro O cj CO CO o 53 CO 3 ro O CN -a CJ H oo 00 fl CJ b oo 2 5 genome size, allowing around 2kb of extra DNA. Deletions of 3kb in the El region and 1.9kb from the E3 region facilitate insertions of close to 7kb. The LPL cDNA, which is only 1.6kb, is easily accommodated along with regulatory elements. Due to the availability of this first generation adenovector, this thesis has focused on biologically relevant studies aimed to lay a foundation for future studies moving towards gene therapy for LPL deficiency. Recently a new class of adenovectors called helper dependent or gutted adenoviruses have been created {103,104}. These viruses are currently the best hope for adenovirus clinically since all of the endogenous viral DNA, except for the packaging sequences, is deleted and they only contain the desired patient specific DNA. This reduces the immune response to the virus allowing longer-term expression (> lyr) in the absence of an immune response to the transgene(s) included. Of particular concern for LPL is the intrinsic hepatotrophism of the adenovirus when intravenously administered {105}. Although LPL transcription is not natural to the adult liver {90,106}, we hypothesized that LPL should be able to express in this site when under the control of another promoter. However, since the liver is the major site of LPL degradation in vivo, it was unknown whether hepatically derived LPL would be able to escape the liver and move to the peripheral endothelium where its action is most desired. This question was a major focus of our initial studies and the liver proved to be a relatively good site for producing metabolically active LPL. Other than those performed by our group, two other groups have published studies on Ad-LPL delivery {107,108}. One study looked at the alteration in lipoprotein profiles in apoE and LDL receptor (LDL-R) deficient mouse models {108}. Consistent with our work, this study, employing a CMV-based promoter, showed significantly elevated hepatic and post-heparin plasma LPL activity. Both total plasma cholesterol and TG were reduced with significant alterations mainly in the VLDL/chylomicron remnant fractions as demonstrated by fast protein liquid chromatography (FPLC). The other study utilized adenovectors as a powerful approach to further understand the biology of LPL. Kobayashi et al undertook the analysis of structure and function of LPL and its close relative hepatic lipase (HL) by mixing the structural domains of the two lipases and measuring alterations in plasma cholesterol and phospholipid in HL deficient mice after Ad-mediated delivery {107}. The adenovectors 26 contained a luciferase marker gene, the wild type LPL or HL gene, or lipase mutants in which the lid covering the catalytic site of either enzyme was exchanged. Plasma lipase activity was increased with the administration of any of the four lipase containing Ads. A significant difference in phospholipase activity was observed with the lipases containing the HL lid revealing preferential phospholipase activity. It was concluded that the lid to the catalytic site is a major structural motif that bestows differential phospholipase activities between HL and LPL in vivo and is one of the factors conferring the distinct physiological roles of these two similar lipases. This second study in particular demonstrates the powerful use of adenovectors as tools for the dissection of complex biological questions in vivo. Alternative vector systems Although there are many systems available for delivering genes into target cells, few have been examined for their potential for LPL gene delivery. As with Ad, all delivery systems have limitations. In particular the limitations become obvious when the desired gene product requires expression at relatively high levels. A system that provides sufficient expression for hormones such as growth hormone and erythropoetin is often insufficient for an enzyme such as LPL. The first vector system investigated for LPL was the retrovirus {109}. This is an RNA virus that is capable of inserting into the genomic DNA of a cell after reverse transcription upon infection. The major advantage of this system is that after insertion into the cell's DNA, the gene is transmitted to the progeny of that cell. Some of the disadvantages include the fact that the integration site of the viral DNA can not be controlled and the cell must be replicating for reverse transcription and integration to occur. Thus this system has been mainly applied in ex vivo applications where cells (i.e. hematopoietic, myoblasts or hepatocytes) are removed from the body, grown and infected with the virus in culture and then returned to the body. This allows the person requiring the cellular modification to be their own donor and reduce most of the immune response to both the virus and foreign cells. This does not reduce the potential immune response to the gene product if it is foreign to the body. In vivo studies with direct injections of this virus have been mainly limited to pre-clinical protocols involving liver resection, muscle injury or 2 7 growth factor pre-treatment in order to trigger cell cycling before administration of the retrovirus {110}. These experiments tested a series of recombinant myeloproliferative sarcoma virus (MPSV)-based retroviruses with LPL expression under the control of the constitutive MPSV long terminal repeat (LTR) {109}. Elevated LPL mRNA, protein mass and activity could be readily observed in several cell lines in vitro, including several hematopoietic cell lines, myoblasts and fibroblasts. Differential expression levels were seen to be discordant with RNA levels in a few cell lines suggesting post-transcriptional mechanisms may affect the efficiency of RNA turn-over, translation or protein turn-over. Primary human fibroblasts yielded the most significant increase in activity and protein levels and may represent a viable target for retroviral mediated-gene transfer. Unfortunately this system proved to be technically difficult and applications in vivo were unsuccessful (personal communication). Another system successfully employed in vivo was the direct injection of naked DNA {81,111}. This system has been applied to several genes, especially hormones such as the vascular endothelial growth factor {112,113}.A few of the advantages of this system are that DNA is fairly easily prepared and can potentially be of any size accommodating large genes and regulatory sequences {81}. DNA is also generally non-immunogenic allowing readministration as many times as is necessary. The major limitation is the amount of DNA that actually is able to be expressed within the target cells. Although the gene may be detectable by PCR, little to no expression is often seen. Plasmid DNA containing the human LPL gene under the control of the muscle specific promoter of the mouse creatine kinase gene (mCK) was delivered to mice either intramuscularly (IM) or intraperitoneally (IP) on a regular or high fat diet, respectively {111}. IM treatment resulted in a significant increase in LPL mRNA as detected by RT-PCR within the muscle of both the treated quadriceps as well as the opposite side only in mice given the hLPL plasmid. Repeated IM delivery resulted in a significant increase in mRNA. Both IM and IP delivery resulted in a decrease in plasma TG (26% and 38% respectively). Similar experiments were attempted within this lab yielding increases in tissue specific LPL activity but no changes in plasma TG levels (data not shown). The reasons for this experimental difference may have been due to the promoter since our experiments used the strong viral CMV promoter rather than a tissue specific promoter. 28 The retrovirus and naked DNA delivery systems were initially deemed appropriate for LPL gene replacement due to the natural affinity of these systems to cells which naturally produce LPL, namely the hematopoietic (macrophage) and muscular skeletal tissue compartments. Another potential system that has an affinity for myoblasts is the adeno-associated virus (AAV) {114}. This is a small DNA virus that was initially found contaminating laboratory adenovirus stocks. It is hypothesized that its small size allows movement of the virus between the cell junctions in the muscle compartment basically bathing each muscle fiber and myoblast in virus particles thus bringing the virus in close proximity of the target and facilitating infection {114-116}. Although this virus is small, allowing only approximately 4.5kb of foreign DNA, it is large enough to easily accommodate our 1.6kb hLPL construct and regulatory regions. Currently the development of this vector system is underway in our lab and the goal is to apply this vector to the muscle and potentially the liver and adipose compartments (either coordinately or individually) in our animal models. Other groups are also investigating AAV as a gene transfer tool and recently Marshall et al have shown a role for LPL in insulin secretion from pancreatic islet cells {117}. Additionally, recent in vivo studies indicate that this vector system may allow integration and prolonged expression for greater than a year {82,83}. Initial phase 1 clinical studies are currently underway using this vector which contains the human blood clotting factor IX gene for the treatment of hemophilia B {118}. 1.5.3 Promoter Specificity As described above, several transgenic and gene transfer experiments have been performed utilizing various promoter sequences to drive the expression of the human LPL gene. Among these are the ubiquitous CMV or RSV promoters {26,27,119,120}, the metallothionein promoter which requires induction with ZnS04 {52}, the muscle creatine kinase promoter which confers muscle specific expression {25,57,58,111}, a fragment of the human apo AI gene promoter which controls liver specific expression {42} and even 8kb of the endogenous LPL promoter which surprisingly results solely in cardiac expression {44}. Although each promoter yielded interesting and significant results, each had limitations such as a limited expression range in tissues and/or levels. Deciding upon gene regulation is one of the most challenging elements to designing an expression vector. Many regulatory 29 elements have been described but most are not well characterized and may give varying results depending on the gene or project. For example, the mCK promoter produced LPL transgenic expression in hearts of the model produced by Levak-Frank et al {25} but not by Jensen et al {57}. Optimally, genes used for gene therapy purposes should either involve their own promoter and be transferred to regions of natural endogenous gene expression or replace the endogenous mutant gene leaving the control regions intact. In this manner the body would be able to regulate the gene in the appropriate manner for the life of the organism. Unfortunately, this is not feasible at this time. The next best option is likely to have a regulatable expression system. Several are currently being developed but limitations such as toxicity, tissue specificity and time to turn on and off remain a barrier to their application for gene therapy {83,121}. In light of these facts, preliminary studies using specific, well characterized promoter systems to experimentally determining tissue specificity, tolerance to expression before toxicity, acceptable levels of toxicity and dose response relationships to further characterize the particular gene and its respective phenotype are essential. This thesis explores this latter approach using two different viral promoter systems: the cytomegalovirus (CMV) promoter and the Rous sarcoma virus (RSV) promoter. Although classical characteristics of these two promoters differ, they both result in ubiquitous gene expression. The particular characteristics of these viruses are elaborated upon in Chapter 3 of this thesis. 1.6 Objectives The specific objective of this thesis was to investigate the feasibility of adenoviral-mediated gene transfer to model systems of LPL deficiency. This objective included the evaluation of the alterations in lipid metabolism, tissue specificity, gene regulation and toxicity associated with the delivery of this gene to various compartments within the host target system. Several hypotheses were investigated as studies progressed from in vitro to in vivo model systems including the validity of the vector system, the systemic effect of targeting the human LPL gene to tissues that naturally express LPL (adipose, muscle) and one that does not (liver), the xenogenic effects of human LPL in model organisms (mouse, 30 cat) in vivo, and the validity of using a common polymorphic form of the LPL gene (Ser447Ter). 31 Chapter 2: Methods and Materials 32 2.1 LPL cDNAs The wild type human LPL cDNA fragment was previously cloned by RT-PCR from human adipose tissue total RNA using the following 5' and 3' UTR primers respectively; 5'-ATA GAA TTC GGA TCC ATC GAT/GC TCC TCC AGA GGG ACG GCG CCC CG-3' (which introduces an EcoRI, BamHI and Clal site 5' of the LPL coding sequence) and 5'-TAT GTC GAC TAG ATA TC/GCC GTT CTT TGT TCT GTA GAT TCG CCC-3' (introducing Sail, Xbal and EcoRV sites 3' of the LPL coding sequence) {119}. The LPL cDNA obtained was confirmed by sequencing. The Ser447Ter mutant cDNA was previously derived from the wild type human 1.6kb LPL cDNA, altered by site directed mutagenesis and confirmed by sequencing {122}. 2.2 Adenovirus construction 2.2.1 RSV-driven Adenovirus A recombinant serotype 5 adenovirus (Ad5), containing a reporter p-galactosidase gene under RSV promoter control (Ad-RSV-LacZ) was obtained from Rhone-Poulenc Rorer for use in concurrent control in vitro and in vivo infections {123}. Generation of our recombinant Ad-RSV-LPL vector (Figure 3.1) utilized the same shuttle plasmid, pAd.RSVpGal, which contains the P-galactosidase gene from E. coli coupled to the nuclear localization signal of SV40 (nls lacZ gene) with a bovine growth hormone poly A site and SV40 small t intron from a pSV2 vector {124} under transcriptional control of the plasmid RSV long terminal repeat (LTR). The LPL cDNA was double digested with Clal and Sail and the resulting 1.6 kb fragment was introduced into the shuttle plasmid, pAd.RSVpGal, replacing the nls lacZ gene with the human LPL cDNA. The human LPL gene was inserted 1.3 map units (mu) downstream from the left end of the adenovirus genome and is followed by mu 9.4-17 of Ad5 allowing for homologous recombination. The shuttle vector and the right end of the C/a/-digested, E3-deleted fragment of the Ad5 genome were co-transfected in 293 cells by calcium-phosphate precipitation. At day 7 plaques were screened for homologous recombinants by PCR using primers: 5'-GTT CCG GGT CAA AGT TGG CG-3' in the adenovector sequence, 5' to the RSV promoter, and 3'-AGG TGC ACA CCA ATG TGG TG-5' in the 5' untranslated region of the LPL transgene. Positive clones were screened for in vitro LPL activity released to the 33 r medium after addition of heparin at 5 U/ml. Five clones with the highest LPL activity were chosen for amplification in 293 cell suspension culture. Cells were grown to 5-6 x 105 cells/ml, spun down briefly, resuspended in a small volume of media and infected with adenovirus (Ad) at a MOI of 1-5 for 1 hour. Media (S-MEM/5% FBS) was then added and cells were allowed to grow for another 36-48 hours. Upon observation of cytopathic effects, cells were spun down, media decanted and the cells resuspended in approximately 10 ml of 10 mM Tris, pH 7.9. After 3 rounds of freeze thawing, cell debris was spun down briefly. Purification of high titer recombinant virus (10npru/ml) was performed by double rounds of CsCl density gradient ultracentrifugation. Briefly, the first spin was a step gradient of 3 ml each of CsCl of density 1.2 and 1.45 (10 mM Tris, pH 7.9). The cell supernatant was then added on top and spun in a Beckman SW41 swing bucket rotor at 36,000 rpm for 4 hours. The upper bands were removed by pipette and the Ad band was put into a 15mL falcon tube and diluted to 2.5mL with 10 mM Tris. The second spin was a CsCl isopycnic gradient (a mix of an equal volume of CsCl d=1.2 (4mL) and d=1.45 (4mL)). The Ad was then put on top and spun overnight (at least 16 hours) @ 35,000 rpm, 18C. The Ad band was again isolated and the purified virus stocks were passed through a PD10 (Pharmacia LKB Biotechnology Inc., Piscataway, NJ) column and stored in 10% glycerol and 1% BSA at -80°C. Titers of the viral stocks were determined by plaque assay using 293 cells. Briefly, serial dilutions of viral stock from 10" to 10" was made in 2% FBS/DMEM infection medium and added to 293 monolayer cells in 6 well culture dishes, grown to approximately 80% confluency. After 1 hour incubation at 37°C, infection medium was removed and 2 ml of 0.8% agarose in DMEM/2%FBS was spread slowly on the monolayers. Three days later, 1 ml of DMEM/2%FBS in 0.8% agar was added and the plaques were read after at least 7 days. The titer was calculated as plaque forming units per milliliter (pfu/ml) and was approximately 1-2 x 1011 pfu/ml. Virus preparations were quantified by Lowry protein assay, using bovine serum albumin (BSA) as a standard and the relationship of lmg/ml protein = 3.4 x 10 adenovirus particles/ml {125}. Measurements within our lab consistently revealed 75-100 particles per pfu from both Ad-RSV-LacZ and Ad-RSV-LPL preparations. 34 2.2.2 CMV-driven Adenovirus Generation of our recombinant Ad-CMV-LPL vector (Figure 3.2) utilized the Adeno-Quest™ kit purchased from Quantum Biotechnologies Inc. (Montreal, QC). The same shuttle plasmid was used for the generation of the wild type and 447 LPL viruses as well as human secreted alkaline phosphatase (AP). Briefly, each of these cDNAs was cloned into the shuttle plasmid pQBI-AdCMV5, which contains the CMV5 promoter and enhancer and a globin poly A. The LPL cDNAs (wild type or 447) were double digested with Hindlll and Xbal and inserted into the BamHI site via blunt end ligation. The AP cDNA, which was a generous gift from Rhone-Poulenc Rorer, was digested with Hindlll and Xhol, and inserted into pQBI-AdCMV5 at the BamHI site via blunt end ligation. The human LPL gene was inserted 1.5 map units (mu) downstream from the left end of the adenovirus genome and is followed by mu 9.4-15.5 of Ad5 allowing for homologous recombination. The shuttle vector and the right end of the C7a/-digested, E3 -deleted fragment of the Ad5 genome were co-transfected in 293 cells by calcium-phosphate precipitation and overlaid with 0.8% agarose in DMEM/5% FBS. Ad-CMV-LacZ, which contains the (5-galactosidase gene from E. coli with a globin poly A site under transcriptional control of the plasmid cytomegalovirus promoter and enhancer was obtained in low titer from Quantum Biotechnologies. After plaque purification, as described below, a high expressing clone was selected and amplified. At day 9 and 14 plaques were screened in vitro for LPL or AP activity released to the medium after addition of heparin at 5 U/ml. Two clones with the highest LPL activity, mass or AP activity were chosen, plaque purified a second time and amplified in 293A cells on 15 cm plates. Purification of high titer recombinant virus (~3xl010pfu/ml) was performed as above by double rounds of CsCl density gradient ultracentrifugation. The purified virus stocks were dialysized against 4 changes of HEPES buffered saline (HBS, 20mM Hepes, 150 mM NaCl, pH 7.3) with 10% glycerol over 16-18 hours and stored at -80°C. Titers of the viral stocks were determined by plaque assay using 293 cells as described above. The titer was calculated as plaque forming units per milliliter (pfu/ml) and was 1-3 x 1010pfu/ml. Virus preparations were quantified by Lowry protein assay consistently revealed -50 particles per pfu from all Ad-CMV-LPL/447/AP and LacZ and preparations. 35 2.2.3 PCR Analysis for Replication Competent Adenovirus Control adenovirus DNA was obtained by adding approximately 200ul (lxlO9 pfu) of CsCl purified adenovirus dropwise to a solution of 1 mg/ml pronase in 10 mM EDTA and 1% SDS in lOmM Tris-HCL, pH 7.5 and incubating at 37°C for one hour. One phenol extraction was performed to remove the capsid proteins. DNA was precipitated with 1/10 volume of 30% sodium acetate and either 1 volume of isopropanol or 2 volumes of 95% ethanol. DNA was dissolved in 10 mM Tris, pH 7.5 and quantitated by measuring the absorbance at 260nm. DNA from 293 and HepG2 cells was isolated using a Qiagen tissue isolation DNA kit. Stocks were screened by PCR for wildtype adenovirus contamination. Polymerase chain reaction (PCR) amplification of 100 ng DNA, respectively sampled from the Ad-RSV-LPL/LacZ or Ad-CMV-LPL/AP constructs, 293 and HepG2 cells were performed according to Zhang et al. (1995){ 126}, with the following modifications. Reactions were performed in a total volume of 50 pi in the presence of 2 uM each of the four specific primers, 50mM KCI, 2mM MgCl 2, 10 mM Tris-HCl pH 8.4 at 70°C, 0.1% Triton®X-100, 0.2 mM of each dNTP, and 1.25 units Taq DNA polymerase (BRL; Gaithersburg, MD). The reactions were incubated for 2 min at 95°C, followed by 30 cycles at 95°C for 30 sec, 56 °C for 30 sec, and 72 °C for 60 sec. Four oligonucleotides that correspond to the Ad genome were used as previously described {126}; primer pElA-1 (5'-ATT ACC GAA GAA ATG GCC GC), primer pEl A-2 (5'-CCC ATT TAA CAC GCC ATG CA), primer pE2B-l (5'-TCG TTT CTC AGC AGC TGT TG) and primer pE2B-2 (5'-CAT CTG AAC TCA AAG CGT GG). The PCR products were analyzed in a 1% agarose gel. No wildtype adenovirus was obvious by this method. An example showing Ad-RSV-LPL and 293 cells are shown in Figure 2.1. 2.3 Tissue Culture 2.3.1 Primary Cell Culture Primary cat and mouse fibroblasts were obtained by taking skin biopsies at sacrifice. The skin was washed with 70% alcohol, taken with a sterile scalpel and placed in cold PBS for transportation to tissue culture facilities. The tissue was rinsed 3 times with sterile PBS, minced using a sterile scalpel and incubated in 200 U/ml collagenase (Sigma) in D-MEM over night. The media was then changed to standard media (D-MEM/10% FBS, 36 Figure 2.1 PCR screening for wild-type adenovirus. (A) Ad-RSV-LPL DNA, (B) 293 cellular DNA. No recombinant adenovirus containing the required El regulatory region was observed in any adenovector stocks by PCR. In contrast, the transcomplementing El region was obvious in the packaging cell line 293. 37 penicillin/streptomycin, glutamine) and cells were allowed to grow for approximately 3-4 days. The tissue was then removed and cells were released by trypsin (0.5%, Gibco BRL, Gaithersburg, MD) and seeded into plates. After 2 days, trypsinized cells were counted and seeded appropriately into 6 or 12 well dishes for infection experiments (5xl05 or lxl0 5 cells/well respectively). 2.3.2 Transformed Cell Culture The human HepG2, the kidney COS (monkey) and 293 (human embryonic), murine muscle C 2 C 1 2 or adipose 3T3L1 cell lines (American Type Culture Collection; ATCC) were grown in standard supplemented D-MEM/10% FBS. Cells were grown as a monolayer and 293 cells were also grown in 1 liter S-MEM/5% FBS suspension flasks for the purpose of large scale amplification of recombinant virus. All cells were kept at 37°C in a humidified atmosphere of 5% C0 2 . Viral infections were performed by co-cultivation of cells and virus in DMEM/2% FBS for 1 hr at various multiplicities of infection (MOI) as defined by the ratio of pfu to cell number, the latter counted routinely from 2-3 individual wells of a 6 or 12 well culture plate. Multiplicities of infection between 12.5-800 were initially tested, with least viral-mediated toxicity obtained at MOI < 100. Medium was collected at 24 hr, 48 hr post-infection or as specified in the text. Unless otherwise indicated, heparin was added to the medium to a final concentration of 5 U/ml for 10-15 min, at 37°C, prior to harvest. 2.3.3 P-Galactosidase (LacZ) Expression in Cell Culture After infection in vitro with the recombinant marker Ad-RSV or CMV-LacZ vector, the transduced cells grown on culture plates or chamber slides were fixed in 0.25% glutaraldehyde/phosphate-buffer saline (PBS) for 10 min at 4°C. Cells were then washed 2X with permeation buffer (0.1 % TritonX-100 and 0.25% Sodium Cholic acid in PBS) and incubated with a P-Gal substrate 5-bromo-4-chloro-3-indolyl fi-D-galactopyranoside (X-Gal) in 5mM K4Fe(CN)6/ 5mM K3Fe(CN)6/ 2mM MgCl. Endogenous p-Gal activity in cells was minimized by incubation with X-Gal substrate for only 30-60 min. Cells which stained blue were considered positive for LacZ gene expression. 38 2.3.4 Anti-Adenovirus Neutralizing Antibody assay Cat or mouse serum samples were diluted with 2% FBS/DMEM in 2-fold steps starting from 1:8. Each dilution of 50 pi was incubated with 6x104 pfu of an available adenoviral reporter vector (Ad-RSV-LacZ) carrying the bacterial P-galactosidase reporter gene (LacZ) for 1 hr at 37°C and applied to COS cells split the day before at 2xl04/well of a 96 well dish. After 1 hr incubation at 37°C, the mixture was removed and 10% FBS/DMEM was added. Cells were fixed and stained for P-galactosidase expression 24 hr later. The titer of neutralizing antibody was defined as the highest dilution with which <50% of cells stained positively for P-galactosidase expression. 2.3.5 VLDL Isolation and HepG2 Cell VLDL-TG Hydrolysis Fasted human plasma was layered with 0.15 M NaCl and spun for 18 hr at 30,000 rpm at 15°C in a Sorval 41 swing bucket rotor. The top layer was collected and re-spun once in similar conditions. VLDL was analyzed for TG and cholesterol by an enzymatic kit assay (Boelmnger-Mannheim, Mannheim, Germany). HepG2 cells were infected by either Ad-RSV-LacZ or Ad-RSV-LPL, each at MOI 100, and subsequently incubated in regular medium for 48 hours. After washing the cells with PBS, serum-free medium containing 1 mg/ml BSA and human VLDL was added, resulting in an approximate TG concentration of 350 pg/ml. TG levels in the media were monitored over a 4 hour period and VLDL in the medium at the end of the experiment was analyzed by FPLC as described below. 2.4 Tissue/cell homogenates Collected tissues or cells were homogenized in 1 ml of extraction buffer per 100 mg tissue (0.025M NH4CI pH 8.2, 5 mM EDTA, 0.4 mg/ml SDS, 8 mg/ml Triton X-100, 33 ug/ml heparin, broad range protease inhibitors (Boehringer Mannheim)). Samples were spun down (2 min, 14,000 rpm, 4°C) and supernatants from the tissue homogenates were stored at -80°C prior to analysis and processed for LPL mass and activity analysis as described below. 39 2.5 Animal handling and injection techniques All procedures involving experimental animals were performed in accordance with protocols from the Canadian Council on Animal Care and the University of British Columbia Animal Care Committee. 2.5.1 Mouse handling and injection techniques Mice were on a CD1, mixed C57BL/6 and 129 background, or back crossed at least 4-6 times to C57BL/6 background. Depending on the study, they were normal, heterozygous for a targeted disruption in the LPL gene {33}, wt litter mates, or transgenic for human LPL {26,57}. Mice were matched for age between 3 to 28 weeks and maintained on a normal chow diet. Prior to blood sampling, the mice were fasted for 12 hours (RSV experiments except for those done in transgenic animals) or 4 hours (CMV experiments) with free access to water. All blood was collected from either the retroorbital plexus or the saphenous vein using heparin coated capillary tubes (Scientific Products, McGaw Park, IL). To estimate LPL immunoreactive mass and activity, post-heparin plasma (PHP) was collected from mice 10 minutes following intravenous injection of 200 U/kg of sodium heparin in saline (Sigma Chemical Co., St. Louis, MO). Blood was spun briefly (5-6 min at 12,000 rpm, 4°C) and the plasma samples were stored at -80°C prior to analysis. Murine Viral Intravenous Infection A dose of 5xl09 pfu of either Ad-RSV-LPL or Ad-RSV-LacZ respectively was diluted to a 200 pi final volume in Dulbecco's Modified Eagle Medium (DMEM, Gibco BRL; Gaithersburg, MD) and infused intravenously through the tail vein to each mouse. For experiments involving Ad-CMV-LPL/AP or LacZ, unless otherwise described, 5x108 pfu was administered in a volume of 150-200ul diluted with either DMEM or isotonic saline solution. Murine Viral Intra-muscular Infection The legs (anterior tibialis and/or quadriceps) of mice between the ages of 3-5 weeks were shaved and wiped briefly with 70% alcohol for visualization of the muscle category to be injected. Depending on the experiment, a dose of 5 or 10x108 pfu of virus was diluted 40 into a volume of 50 or 200 pi of isotonic saline using an insulin syringe fitted with a small piece of tubing leaving approximately 2 mm of the needle exposed in order to control for injection depth. For bilateral treatment of the anterior tibialis muscles, a volume of 25 pi Q (2.5x10 pfu per side) was directly injected into the muscle. For multi-site treatment, 25pl Q (1.25x10° pfu per site) was administered to 4 sites on each side of the mice (total of 8 injections per mouse), including the tibialis anterior, and 3 separate site in the quadriceps. Murine Viral Intra-adipose Infection Mice at approximately 8-12 weeks old were used for these studies (Figure 2.2). Mice were anesthetized using halothane in a sterile environment. After being wiped with 70% ethanol, a small incision was made in the inguinal region of the abdomen allowing easy access to the epididymal fat pads. Each fat pad was withdrawn from the abdominal cavity and injected directly with 50ul of virus diluted with isotonic saline (2.5xl08 pfu per side, Q 5x10 pfu total dose per mouse) adjacent to the epididymus before being replaced into the abdominal cavity. Although it is unknown if leakage occurred once returned to the abdominal cavity, no leakage was apparent during injection. 2.5.2 Feline Viral Intravenous Infection and Sampling Wild-type, heterozygous and homozygous LPL-deficient cats (age 1 to 5 years, weighing 2.1 to 4.2 kg) were members of a colony of domestic cats originally from New Zealand, now maintained at the Animal Care Unit of University of British Columbia. They were fed a commercial cat food diet (Whiskas: Protein 57.5%, Fat 24.3% and Ash 18.2%). LPL genotyping of the cats was performed as previously described {35}. Jugular venous blood samples were obtained after overnight fast. In experiments involving intravenous fat challenge, frequent sampling of blood was done through a catheter inserted within the cephalic vein. After an overnight fast, cats were anesthetized with inhalation of isoflurone. A catheter was inserted into the cephalic vein and attached to a saline intravenous line. Adenoviral vector at 1.15x 10 or 8x 10 pfu/kg (RSV or CMV driven adenovectors respectively), or at the dose described in the text, was diluted to 10 ml with DMEM and infused through the injection port over 10 min. A portion (0.1 ml) of the same viral vector 41 42 preparation was injected iv to mice after infusion to the cats. Transgene expression (LPL and AP) was confirmed in this small cohort of LPL +/- gene targeted mice by measurement of the catalytic activity of LPL 2 days post-injection as described below. 2.5.3 Intravenous Fat Load Test Murine On Day 6 after Ad-RSV-LPL and Ad-RSV-LacZ infection, the mice were fasted for 12 hours. On Day 7, 3 groups of mice including wt uninfected controls (n = 9) or heterozygous Ad-RSV-LacZ injected controls (n = 9) and heterozygous Ad-RSV-LPL injected mice (n = 9), were given 250 pi of 20% Travamulsion (Clintec Nutrition Co. Mississauga, Ontario; containing 20g soybean oil and 1.2 g egg phosphatide per dl) intravenously via the tail vein (n = 4 per cohort). Approximately 50 pi blood was collected retroorbitally at the indicated times for a period of 24 hours. Plasma TG was measured enzymatically as described below. Feline After an overnight 16-hour fast, the cats were intravenously catheterized as described above. Travamulsion (Clintec Nutrition Co. Mississauga, Ontario) containing 20g soybean oil and 1.2 g egg phosphotide/dl was infused at 2 ml/kg via the injection port over 10 minutes. Approximately 1 ml of blood was collected into heparinized Eppendorf tubes over a total period of 6 hours. 2.6 Measurement of LPL Activity and Mass To estimate LPL immunoreactive mass and activity, PHP was collected from mice following intravenous injection of 200 U/kg of sodium heparin (Sigma Chemical Co; St. Louis, MO) via the tail vein. Cats were given an injection of 100 U/kg of sodium heparin via the jugular vein. Blood was subsequently collected 10 minutes post-heparin injection, placed immediately on ice and plasma was separated and collected after microfuge centrifugation at 4°C. PHP samples were immediately frozen at -80°C. Total LPL activity in post-heparin plasma was measured in duplicate using a [ H]-triolien emulsion substrate as previously described {68,127}. Briefly, 10-50 pi of sample 43 was mixed with 100 ul of a radiolabeled triolien substrate mixture. Human pre-heparin plasma was always included in the substrate mixture to ensure the presence of the LPL transactivating protein ApoCII. Duplicate samples were done with and without a 30 minute pre-incubation on ice in 1 M NaCl to differentiate between LPL and hepatic lipase. The samples were then vortexed and shaken at 37°C for 1 hr at which time belfrage extraction mixture (1.41 methanol: 1.25 chloroform: 1 heptane) and 0.05 M carbonate-borate buffer (pH 10.5) were added and vortexed again. The released fatty acids were extracted and counted in a beta-scintillation counter {128}. Human-specific LPL activity was obtained after a 2 hour, 4°C pre-incubation of plasma with the human-specific 5D2 monoclonal Ab, diluted . 1:8. This approach inhibits human LPL activity by at least 85% but does not alter endogenous mouse LPL activity {68,129}. One milli-unit (mU) of lipase activity is equivalent to 1 nmol free fatty acid released per minute at 37°C. A neutralizing anti-LPL antibody response was measured by pre-incubating purified bovine LPL with aliquots of plasma for 1 hour at 4°C. The substrate for the LPL activity assay (as above) was then added directly to these samples and the activity assay was completed as above. In the Ad-RS V studies, total (monomeric + dimeric) LPL protein mass (ng/105 cells/ml) was assayed by ELISA based on the two mAbs, 5F9 and 5D2, raised against purified bovine milk LPL {129}. These monoclonal antibodies against LPL were generous gifts from Dr. John Brunzell (University of Washington, Seattle, WA). The 5D2 mAb recognizes an epitope located at residue 400 of human LPL and does not recognize mouse LPL whereas 5F9 recognizes an unknown epitope. Since each mAb recognizes different epitopes of LPL they were used in a sandwich ELISA to assess LPL immunoreactive mass. 5F9 was routinely coated in 96-well plates to serve as the capture Ab while horseradish-peroxidase conjugated 5D2 served as the detection Ab. 200 ul of the 5F9 mAb in PBS was added to each microtiter plate well (Polysorp, Nunc Inc.), sealed and incubated at 37°C for 4 hours. The plates were then washed four times with PBS containing Tween-20 (0.05% v/v), sealed and stored at 4°C. Coated plates were routinely used within 2 weeks. Purified bovine LPL was stored in 50% glycerol at -20°C, and diluted in PBS containing glycerol (15% v/v), Tween-20 (0.1% v/v), BSA (50 ug/ml) and heparin (lmg/ml) (dilution buffer). Denatured medium containing 1.2 M guanidium hydrochloride (GuHCl) kept on ice for 1 h was diluted 44 10 times in dilution buffer and added to washed plates. After 16 h incubation at 4°C, the plates were thoroughly washed and horseradish-peroxidase conjugated 5D2 was added for a 4 h incubation at room temperature. The plates were again washed thoroughly and fresh substrate solution containing urea peroxide and 3,3',5,5'-tetramethyl benzidine was added and incubated in the dark for 5-10 min. 4M H2 SO4 was used to stop the reaction. The plates were read by a Bio-Rad EIA plate reader using a 450 nm filter. Standard curves were generated by the identical treatment of purified bovine LPL. Due to the unavailability of the 5F9 mAb, the Ad-CMV studies employed a polyclonal chicken antibody raised against bovine LPL for capture in an otherwise identical sandwich ELISA assay. 2.7. Measurement of Lipids To assess lipid levels, HDL-C was determined after LDL and VLDL-C precipitation with an equal volume of 10% polyethylene glycol 8000 solution. HDL-C, TC and TG were determined using commercial kits #C236691 and #450032 respectively, purchased from Boehringer Mannheim (Mannheim, Germany). 2.8 Fast Protein Liquid Chromatography (FPLC) In the experiments of VLDL hydrolysis, culture medium from Ad-RSV-LacZ and Ad-RSV-LPL infected cells conditioned with serum-free human VLDL for 4 hours, were analyzed by FPLC (Pharmacia LKB Biotechnology Inc., Piscataway, NJ). Two Superose™ 6 columns (Pharmacia LKB Biotechnology Inc., Piscataway, NJ) were connected in series and 400 pi of filtered (0.22 pm) medium was injected into the first column. The elution was started at a flow rate of 0.5 ml/min with a running buffer consisting of 0.15M NaCl, 1 mM EDTA and 0.02% NaN3, pH 8.2. Fractions of 0.5 ml were collected. TG and TC content were determined by an enzymatic kit assay (Boehringer Mannheim, Mannheim, Germany) in 96-well plates. For in vivo experiments, plasma was sampled and pooled from each mouse cohort at various time intervals. After 0.22 pm filtration, 200 pi of plasma was loaded on to the columns. The elution flow rate, running buffer and fractions were collected as above. 45 2.9 Measurement of Alkaline Phosphatase Activity Plasma and tissue culture samples were diluted 5x with PBS (20 ul sample, 80 pi PBS). The samples were heated to 65°C for 5 minutes and 45 ul was added in duplicate to 96 well ELISA plates (Nunc). An AP standard curve was made using 10 U/ml stock solution (Sigma Chemical Co., St. Louis, MO). Approximately 100 pi of 2x SEAP (2 M diethanolamine, 20 mM L-homoarginine, 1 mM MgCb, pH 9.8) was added to each well, followed by 50 pi of p-nitrophenyl phosphate substrate (1 tablet/1.25 ml water, Sigma product no. N-1891, St. Louis, MO) and allowed to incubate at room temperature for 30 min. The reaction was stopped by the addition of 3 N NaOH solution. Samples were read at 405 nM in an ELISA reader. 2.10 Sacrifice and Histology Necropsies on mice were performed immediately after sacrifice by CO2 administration. Necropsies on cats were performed immediately after sacrifice by administration of a lethal dose of sodium pentobarbital. Harvested tissues included liver, spleen, heart, lung, kidney, muscle and adipose. Each tissue was divided in to 3 pieces which were 1) fixed in phosphate buffered formalin for at least 24 hours and then processed in paraffin by Julie Chow at the University hospital Morphological Services Laboratory, 2) flash frozen or, 3) flash frozen in OCT™. All frozen samples were stored at -80°C. Following standard processing by Julie Chow, tissues were stained with hematoxylin and eosin for general morphometric analysis. Oil red O (ORO) was used to stain for neutral lipid. 2.11 RNA Analysis Total cellular RNA was extracted from samples according to the method of Chromczynski & Sacchi {130}, or a modification of that method using TRIzol (Gibco). Briefly, 50 mg tissue samples were homogenized in an acid guanidium thiocyanate solution, phenol extracted, separated by centrifugation and precipitated by 1 volume of isopropanol. RNA was resuspended in DEPC treated water with or without 0.5% SDS at 65°C for 10 minutes. 46 2.11.1 Northern Analysis RNA samples were run on a standard formaldehyde 1% agarose gel for 6 to 12 hours and transferred overnight to a nitrocellulose membrane. An a32P-labeled 1.4 kb human LPL cDNA probe was created by mixing 50-100 ng of cDNA with an oligo-labeling buffer (dNTPs, random hexamers, buffer), BSA, cc32P-dCTP, and Klenow polymerase. Membranes were hybridized with the labeled probe overnight at 60°C and washed in lx sodium citrate solution for approximately 30 min. Autoradiographs were obtained by exposure to fdm, in cassettes with intensifying screens, at -70°C for 12 hrs. 2.11.2 R T - P C R Reverse transcription (RT) was carried out on 1-2 ug of total RNA using oligo(dT) 12-18 primer in the presence and absence of reverse transcriptase (Gibco, Cat. No. 18089-011). Briefly, total RNA to a volume of lOul was mixed with lul oligo(dT). The samples were heated to 70°C for 10 minutes then put on ice for at least 1 minute. The other components for RT were then added (5 ul First strand buffer, 1 ul 0.1 M DTT, 1 ul 10 mM dNTP, 1 ul Rnase inhibitor, 1 ul Superscript II RTase) and mixed by pipetting. Samples were then heated to 42°C for 1 hr, 70°C for 15 min and stored at -20°C until used. PCR was carried out on 2 pi of a 1:100 dilution of RT product. The PCR for LPL was a modification of the method of Levesque et al {131}. Briefly, primers spanning exons 4, 5 and 6, which amplify the human, mouse and cat LPL genes were used (296bp product; LPLRT1: 5'GTAGGAAGTCTGACCAATAAG, LPLRT2: 5'GTGGACCAGCTAGTGAAGTG, 34 cycles of denaturing 30 sec. at 94°C, annealing 30 sec. at 64°C, and elongation 45 sec. at 72°C; terminal elongation 5 min.). Resolution of these PCR products was achieved by digestion with 2.5 units of BsaJI{A h at 60°C), which differentially cuts the cat, mouse and human PCR products (cat: 25, 271; mouse: 136, 160; human: 147, 149). The products were then separated using polyacrylamide gel electrophoresis, stained with ethidium bromide and photographed using ultraviolet light. The control PCR for feline studies was carried out using primers specific for the feline glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (fGAPDHl: 5 'TC ACC AGGGCTGCTTTT AAC; fGAPDH2: 5 'TGGG ATTTCC ATTG ATG AC A, 35 47 cycles of denaturing 30 sec. at 94°C, annealing 30 sec. at 58°C, and elongation 30 sec. at 72°C; terminal elongation 5 min.). The control PCR for murine studies involved primers specific for the murine p-actin gene purchased from Stratagene (Cat#302110-12). The products were then separated using agarose or acrylamide gel electrophoresis, stained with ethidium bromide, and photographed. 2.11.3 In Situ Hybridization (ISH) Livers from both the Ad-LPL and Ad-LacZ treated animals were harvested 7 and 42 days after the administration of each vector and at the end of the experiment on Day 60. ISH was performed on tissue sections as follows. Three mm paraffin-embedded tissue sections were cut onto Superfrost™ glass slides (Fisher, Nepeon, ON), baked, dewaxed and rehydrated in graded alcohols. The tissue was permeabilized with 0.2N HC1, 2 x SSC, and 1 mg/ml of proteinase K. Following dehydration in graded alcohols and drying, the hybridization solution (formamide, 10 mM Tris/EDTA, lOOug/ml tRNA, lOmM DTT, 0.1% SDS) containing a digoxigenin-labelled LPL strand-specific riboprobe (sense- or antisense-strand, labeling kit from Boehringer Mannheim, Mannheim, Germany), derived from a 1.6 kb human LPL cDNA sequence, was applied. Siliconized coverslips were mounted, sealed with rubber cement and incubated overnight at 42°C. Post-hybridization washing with 50% formamide overnight at 55°C was followed by 2 x SSC washes, blocking with 2% lamb serum and antibody application (1:500 sheep anti-digoxigenin polyclonal antibody, conjugated to alkaline phosphatase, Boehringer Mannheim, Mannheim, Germany). The phosphatase was then activated in a buffer of pH 9.5 and enzymatic development of substrate (NBT/BCIP, Sigma Chemical Co., St. Louis, MO) was performed for 48 hrs at 37°C. Slides were subsequently counterstained with carmalum and examined quantitatively for reaction product by light microscopy. 2.12 Radiolabelled Lipid Analysis 2.12.1 Oleate dissolving and VLDL isolation Approximately 500 uCi (100 pi) of 3H-oleate in toluene ([9,10(n)-3H] Oleic acid, Amersham, TRK140) was diluted with 1.4 ml hexanes and 0.7 ml 100% ethanol. Next 3 pi of 3 N NaOH was added and mixed. The solution was then dried under a stream of N 2 . The 48 dried lipid was then dissolved over night in 1.2 ml 0.15 M NaCl + 2 mg/ml fatty-acid free BSA (Sigma) at room temperature. For time course studies, mice were fasted 1 hour and were intravenously injected with 200 pi (100 pCi) via tail vein. Samples were taken from the retroorbital plexus at 2, 5, 15, 30 and 60 minutes. For isolation of VLDL, mice fasted 30 min were intravenously injected with 200 pi (100 pCi) via tail vein and blood was taken at 30 min by heart puncture. Mice were anesthetized by halothane for all blood sampling. Blood was spun in a table top microcentrifuge for 5 min full speed, the plasma pooled and left at 4°C overnight. The plasma then underwent mini-ultracentrifugation for 5 hrs at 4°C at 100,000xg (54,000 rpm). The top layer (d<1.006 g/ml) was isolated, pooled and diluted with saline. Samples were re-spun for 5 hrs at 4°C at 100,000xg. The top layer was isolated again and counted for 3 H radioactivity. Isolated VLDL was stored at 4°C until use and for no longer than 1 week. 2.12.2 VLDL isolation/injection/clearance/distribution Mice were either injected with Ad-CMV-LPL or AP (5xl08 pfu/mouse) or mock injected with saline solution. Day 2 pre-heparin plasma samples were taken to confirm the success of Ad-CMV-LPL via TG changes and Ad-CMV-AP activity directly in the plasma. On day 7, mice were fasted 2 hours and injected with ~3xl05 cpm of 3H-VLDL in 100 pi, diluted with saline. Blood samples were taken from the retroorbital plexus at 2, 5,15 and 30 minutes. Blood was taken at the time of sacrifice at 60 min post injection and was done by heart puncture. The mouse was then perfused with 10 ml PBS through the left ventricle and tissues were isolated (namely the liver, lung, kidney, spleen, heart, adipose, and skeletal muscle) and immediately frozen on dry ice. Blood was spun for 5 minutes at full speed and the plasma was isolated. All samples were stored at -80°C until analysis. 2.12.3 Lipid isolation, Chromatography and measurement Plasma samples were mixed and tissue samples were homogenized in a 2:1 chloroform-methanol (v/v) mixture in glass tubes and diluted to 20ml/g tissue. Tissue samples were then filtered through Whatman 1 fat free paper. Samples were mixed with 0.2 vol. of 0.017% MgCb/dF^O thoroughly and spun for approximately 5 minutes at 2000 rpm. The upper, protein containing, layer was removed. The sides and interface were washed with 49 upper solvents 3 times (upper phase of a mixture of chloroform-methanol-0.034% MgC^ water solution at 8:4:3 v/v/v), without disturbing lower phase. At least 0.5 ml of the lower, lipid containing phase was removed to a clean glass tube and the solvent was evaporated by a stream of N2 or by leaving samples in the fumehood over night at room temperature. The dried lipid was then dissolved in 10 pi of 2:1 chloroform-methanol (v/v). TG and FFA separation was performed by via thin layer chromatography. Briefly, 5 pi of the dissolved lipid was spotted on silica sheets and run in a solution of hexanes:ether:acetic acid at 83:16:1 v/v/v for approximately 1 hour (10 cm). Iodine crystals were used for development. Pictures were taken immediately and the spots circled with a pencil. After most of the iodine was gone, the spots were cut out, inserted into 24 well scintillation plates and counted for H radioactivity. Controls were oleic acid and triolien at 5 pg per lane. 50 Chapter 3: In vitro assessment of Adenovirus-LPL efficacy The initial studies done in the HepG2 cell line were published as described below and represent a collaboration between myself and Dr. Guoqing Liu, a post-doctoral fellow in our lab. I participated in or performed all experiments presented in this chapter with the exception of the VLDL-hydrolysis experiment (section, Figure3.8), which was work performed by Dr. Liu and is provided herein as additional proof towards the function of Ad-mediated LPL gene transfer and subsequent hepatic-derived expression. Technical support for the majority of the LPL activity, mass and lipid measurements was provided by Li Miao. Relevant publications: 1. Liu, G , Ashbourne Excoffon, K.J.D., Benoit, P., Ginzinger, D.G., Miao, L., Ehrenborg, E., Duverger, N., Denefle, P., Hayden, M.R. and Lewis M.E.S. 1997. Efficient adenovirus-mediated ectopic gene expression of human lipoprotein lipase in human hepatic (HepG2) cells. Human Gene Therapy 8, 205-214. 51 3.1 Introduction Manipulation of the genetic regulators of gene expression is a primary focus for gene transfer. Often, the original tissue specificity of the transgene to be delivered will dictate whether ubiquitous expression is acceptable or if tissue specific expression is required. LPL is normally synthesized in extrahepatic differentiated tissues, primarily skeletal and heart muscle, adipose tissue, macrophages and lactating mammary gland {69}. LPL activity has also been observed in a variety of other tissues including kidney, lung, ovaries and brain with unknown cellular origins and functions in these tissues. A notable exception is the expression of liver LPL, which is uniquely a characteristic of relatively undifferentiated fetal or neonatal hepatocytes only. LPL expression is rapidly extinguished in the newborn liver, most likely by transcriptional regulatory mechanisms, such that LPL is normally absent from adult hepatic cells and tissues {132}. Furthermore, besides the natural absence of functional liver-derived LPL, circulating LPL is efficiently bound and ultimately degraded in the adult liver, the major site for LPL degradation in humans {64}. The physiologic processes that lead to the specific cellular or tissue regulation of LPL are not well understood. Both inactive and active forms of LPL are secreted and normally bind to the membrane-associated proteoglycans. LPL release from the proteoglycans is mediated by the addition of heparin to the plasma or cell culture medium {1,133} allowing the direct measurement of plasma or culture medium for LPL activity and human specific immunoreactive protein mass (mass). Although adenoviral vectors have been widely used in gene transfer studies relevant to lipoprotein metabolism and atherosclerosis in vivo {100-102}, intravenous delivery significantly targets the liver (95-100% {105,134,135}) and thus needs to be carefully evaluated for its utility in LPL gene transfer to various tissue compartments, especially the liver. Our initial goal was to evaluate LPL expression in a wide range of primary and immortalized cell lines in vitro in order to determine if expression, via an adenovirus vector system, was feasible. For these and subsequent in vivo experiments we hypothesized that a well characterized, highly efficient and ubiquitously expressing promoter would answer our initial questions. 52 Most first generation adenoviral vectors that are primarily designed for ubiquitous cellular expression contain one of two promoters: the cytomegalovirus (CMV) or Rous sarcoma virus (RSV) promoter. The CMV promoter is characterized by short term but extremely high activity peaking at around day 5 in vivo and diminishing to near baseline by day 14 {94,97,98}. The RSV promoter generally has longer term expression peaking around day 14 and has been seen as long as 2-6 months in immune deficient {135} or immunosuppressed {136} animals, but sustains lower activity. Accordingly, literature that compares the two adenovector promoters shows markedly higher levels of transgene expression from the CMV promoter {94}. We have created adenoviral vectors containing the LPL gene under either the RSV promoter in order to facilitate longer term expression (Figure 3.1) or the CMV promoter for high levels of expression (Figure 3.2). 3.2 In Vitro Evaluation of Ad-LPL Efficacy 3.2.1 Pilot study on Cell Type versus LPL Activity The efficacy of the RSV based vector was assessed in vitro via infection of cultured cell lines including hepatocytes (human HepG2), preadipocytes (murine 3T3L1), myoblasts (murine C2C12), fibroblasts (primary feline fibroblast explants) and renal cells (Monkey Cos) (see Table 3.1). Subsequently, expression of the CMV based vector was assessed in the human HepG2 model. For most of these experiments, heparin-exposed media supernatants from infected cells were tested for LPL immunoreactive mass and bioactivity by methods well established in our laboratory {68,127,129}. 3.2.2 Hepatocyte Target Due to the in vivo hepatotropism of the adenovirus and the normal lack of LPL expression in the adult liver, initial studies focused on characterizing the expression and function of Ad-RSV-LPL from hepatic cells. We selected HepG2 cells, a human hepatoma-derived cell line expressing many traits attributed to normal human hepatocytes, to investigate whether human LPL could be ectopically and normally expressed in this hepatocellular model after adenoviral-mediated gene delivery. 53 54 < 55 Cell Line Tissue Origin Animal Origin Transformed HepG2 Liver Human C 2 C 1 2 Muscle Mouse 3T3-L1 Adipose Mouse COS Kidney Monkey 293 Kidney Human Primary CF Skin Cat MF Skin Mouse Table 3.1 List of cell lines, immortalized and primary, used or developed within this thesis 56 We hypothesized that the machinery for secretion and post-translational modification of LPL in mature hepatic cells could still be utilized if exogenous LPL was delivered to the liver using recombinant adenoviruses. Furthermore, over-expression of LPL would be able to override the liver's natural role in LPL degradation and thereby enable the HepG2 cells to accommodate LPL-derived hydrolysis of TG-rich lipoproteins, such as chylomicrons and VLDL, normally seen at very low to absent levels in human hepatic cells. Initial studies were performed using the RSV-LTR based vector that has previously demonstrated high activity in hepatocytes in vivo {134,137,138} with subsequent studies using the CMV-promoter. Efficiency of Adenovirus-mediated Gene Transfer in HepG2 Cells HepG2 cells are resistant to most transfection methods (unpublished observations). However, several studies have demonstrated that adenoviral vectors can efficiently transfer genes into hepatocytes in vitro and in vivo with high levels of expression {139-141}. Using a previously described Ad-RSV-LacZ ((3-gal) reporter construct {140}, HepG2 cells were infected for 1 hour (MOI=100) and further incubated for 24 hours. Staining of these cells with B-gal substrate confirmed that the majority of the cells were stained blue (positive) relative to uninfected controls, indicating a transduction efficiency close to 100% (Figure 3.3). Quantitative measurement of basal 13-gal activity in HepG2 cells was compared and, after respective 30 min substrate incubations, endogenous 6-gal activity was undetectable (data not shown). Similar results were observed with a CMV-driven LacZ reporter construct (data not shown). Over-expression of ectopic LPL in HepG2 cells The adenoviral-mediated expression of the human LPL gene in HepG2 cells was confirmed by Northern analysis (Figure 3.4). The 1.6 kb human LPL transcript was abundant in the cells infected with Ad-RSV-LPL harvested 48 hours post-infection (MOI=100), whilst it was completely absent in both uninfected and Ad-RSV-LacZ infected (MOI=100) HepG2 control cells (shown in Figure 3.4 is the Ad-RSV-LacZ control). In the heparin-exposed medium harvested from HepG2 cells infected with Ad-RSV-LPL, the total human LPL protein mass was readily detected by ELISA using the human 57 Figure 3.3 In vitro detection of adenovirus-mediated (3-Gal expression in HepG2 cells. The cells were infected with A d - R S V - L a c Z at moi 100 for 1 hour and incubated in regular medium for 24 hours. Staining with X - G a l for 30 minutes was performed on uninfected control cells (A) and Ad-RSV-LacZ-infected cells (B). 58 Figure 3.4 Northern blot of total cellular R N A from Ad-RSV-LacZ-infected control (Ctl) (moi = 100) and Ad-RSV-LPL transduced (moi = 100) HepG2 cells. The membrane was hybridized to a probe for the human L P L cDNA. Transcript size in kilobases is indicated. 59 LPL-specific monoclonal antibodies (mAbs) 5D2 and 5F9. No exogenous or basal LPL mass or activity was detected in either uninfected or Ad-RSV-LacZ infected (control) HepG2 cells over the range of MOI examined. Figure 3.5 A shows that after 24 hrs incubation, the concentration of human LPL protein (ng/105 cells/ml) in the medium increased in a linear fashion with increasing multiplicity of infection (MOI) of Ad-RSV-LPL between 100 to 600 pfu per cell. The mass did not increase further beyond 600 MOI. A vector-mediated increase in LPL activity in these cells correlated directly with measures of total LPL protein mass. Significant cellular toxicity was observed in parallel with increases in MOI, defined by the degree of cellular lifting and rounding. HepG2 infections with lower MOI, varying between 12.5-100 Ad-RSV-LPL pfu were evaluated after a 48 hour incubation. As shown in Figure 3.5B, this yielded significantly increased levels of LPL protein to 2032.1± 274.5 ng/105 cells/ml at MOI 100 and also reflected a dose-dependent increment from 493.7 ± 102.8 ng/105 cells/ml at MOI 12.5, in each case above negligible levels determined from either uninfected or Ad-RSV-LacZ infected controls. Parallel increases in LPL activity from 32.3 ± 9.9 mU/105 cells/ml at MOI 12.5 to 92.7 ± 22.6 mU/105 cells/ml at MOI 100 were also observed. Moderate cellular toxicity was seen at MOI > 100 only, and at lower MOI (MOI < 100) no discernible changes in cell morphology were seen (data not shown). It is also notable that at the same MOI of 100, LPL mass and activity 48 hours post-infection were approximately 10 times higher than respective determinations at 24 hours. Based on these findings we established an Ad-RSV-LPL infection of HepG2 cells at MOI of 50 and monitored LPL expression over a 5-day period. Figure 3.6 again demonstrates the profound increase in LPL mass and activity between 24 to 48 hours, which peaked at Day 3 and remained essentially unchanged to Day 5. In order to avoid the effect of interval heparin exposure on the cellular release of LPL over time, separate dishes were used for each time point tested over the 5 day period, at 37°C. The maximum observed in vitro LPL activity from HepG2 cells, per 105 cells grown in 1 ml of media over 48 hours, is significant particularly when compared to normal human levels of post-heparin plasma (PHP) LPL activity. An average mU/ml measure for normal human PHP LPL bioactivity determined from our laboratory is 293.9 mU/ml ± 38.2. Therefore one ml of media from as few as 1 x 105 HepG2 cells produced an average 48 hour mU/ml measure of LPL activity at 60 A) 1 > VP u < B) > '•3 U < 60 40 h 24 hours post-infection • Activity • Mass 0 Ctl 100 200 400 600 800 Multiplicity of Infection 48 hours post-infection • Activity • Mass 800 600 400 200 3 3 c 3 = tQ o 3 3 3. o> o r* <" CD fi> Ui Ui 100 - ^ ~ \ ~ -I fafcllt Ctl 12.5 25 50 100 Multiplicity of Infection 2000 1000 3 3 _ c 3 3 (Q O 3"*" 3 _ ft) o r* <" <D 0> Ui Ui Figure 3.5 Efficient expression of the human LPL gene in HepG2 cells. Cells were infected with Ad-RSV-LPL for 1 hour, incubated for a period of 24-48 hrs, as indicated, and subsequently 15 minute heparin-exposed medium was harvested for LPL assay. A) LPL mass and activity at high moi (100-800) determined 24 hours post-infection. The control cells were uninfected HepG2 cells. B) LPL mass and activity at low moi (12.5-100) 48 hours post-infection. The control cells were infected with Ad-RSV-LacZ at moi 100 and demonstrate similar negligible endogenous basal levels of LPL activity and protein mass in either Ad-RSV-LacZ -infected or uninfected control HepG2 cells. The recorded data are the mean of two dishes of cell and in B represent the mean + SD from four dishes of cells. All measurements were performed in duplicate. 61 80 60 H E D E > 40 o < n 20 0 LPL Activity LPL Protein r2000 M500 MOOO h500 3 §• < O 1 2 3 4 5 Days post-infection (MOI = 50) Figure 3.6 Time course of human LP expression in HepG2 cells. Cells were infected with Ad-RSV-LPL at a moi of 50 for 1 hour and thereafter 15 minute heparin-exposed supernatants were harvested for LPL assay of protein mass (ng/ml) and activity (mU/ml) at each indicated time interval. Medium from separate dishes was used to determine LPL mass and activity at each times point to avoid the effect of interval heparin exposure on the distribution and release of cellular LPL levels. The recorded data are the mean + SD from four dishes of cells. All measurements were performed in duplicate. 62 a level nearing 27% of normal PHP LPL activity per ml of plasma from the human circulation. Similar LPL expression results were observed using Ad-CMV-LPL and are shown in Chapter 6 for comparison with an alternative version (Ser447Ter) of the human LPL gene described in that chapter (Figures 6.1-5). Briefly, at an MOI of 50, LPL activity was approximately 29.3 mU/ml/105 cells with protein levels of approximately 973.7 ng/ml/105 cells as measured using LPL chicken polyclonal antibodies (Ab) and the 5D2 mAb 24 hours after infection. Over a time course however, in contrast to Ad-RSV-LPL, expression was extinguished by day 4. Characterization of Expressed Human LPL Activity HepG2 cells can synthesize and secret hepatic lipase, an enzyme that also hydrolyzes TGs in emulsion and is involved in plasma lipoprotein metabolism. In our assay system the substrate emulsion is able to measure both LPL and HL catalytic activities. Although there was negligible LPL activity in the medium from either control HepG2 cells or cells infected with adenovirus containing a reporter gene (Ad-RSV-LacZ), the total lipase activity present in the medium of the cells infected with Ad-RSV-LPL was further analyzed for susceptibility to known LPL inhibitors to ensure it was derived from the expressed LPL transgene. Figure 3.7 shows the 98.1% reduction in total lipase activity seen after the addition of 1 M NaCl, the LPL inhibitor used in our assay to differentiate LPL from HL in post-heparin plasma. In comparison, incubation of the medium with the human-specific mAb 5D2, resulted in a similar decrease of total lipase activity to 5.1% of the original level. These results confirm that the lipolytic activity derived from Ad-RSV-LPL infected cells was almost totally vector derived. VLDL-TG hydrolysis by LPL-expressing HepG2 Cells After confirming that the measured lipase activity in HepG2 cells was indeed derived from expression of the transferred LPL gene, a functional assay of LPL-mediated TG hydrolysis from VLDL was performed by Dr. G. Liu. This is included in this thesis in order to demonstrate the functionality of HepG2 derived LPL. The absence of heparin in the 63 Figure 3.7 Characterization of Ad-RSV-LPL transduced HepG2 cell-derived LPL activity. Heparin-exposed medium from cultured HepG2 cells infected with Ad-RSV-LPL at a moi of 100 for 48 hours was either incubated with 1 M NaCl for 15 minutes on ice or with 1:8 diluted 5D2 mAb for 2 hours on ice. The activities obtained were compared with those measures from the same untreated medium. Results indicate the mean + SD of four dishes of cells measured in duplicate. 64 media minimizes LPL release to the culture supernatant, thereby attributing any observed in vitro alteration in TG hydrolysis to proteoglycan-bound LPL functioning at the cell surface. Figure 3.8A demonstrates the progressive and consistent decrease in VLDL-derived TG from the medium of Ad-RSV-LPL infected HepG2 cells, noted initially at 15 minutes. At the end of a 4 hour incubation, more than half of the VLDL-TG was hydrolyzed by the cells over-expressing LPL relative to time 0. Conversely, VLDL-TG levels in the media from Ad-RSV-LacZ infected (MOI=100) control cells did not change during this period of incubation. Medium collected at the end of the experiment was further analyzed by FPLC to determine the changes in TG concentration and in particles size (if detectable) in the partially hydrolyzed VLDL. As shown in Figure 3.8B, TG levels in the VLDL fraction were markedly decreased in Ad-RSV-LPL transduced cells versus respective Ad-RSV-LacZ infected controls. However, the resultant specific change in the respective VLDL particle size is undetected by the FPLC system. The VLDL in the medium from both LPL over-expressing and control HepG2 cells was eluted at the same position as the void volume. This suggests that the size of the partially hydrolyzed VLDLs were still in a range that was larger than the pore size of the gel matrix of the columns, which would allow normal LDL to be separated. Collectively, these experiments serve to demonstrate the ability of hepatocytes to express bioactive human LPL transferred via an adenoviral vector. 3.2.3 Adipocyte target In vitro LPL expression within the adipose tissue compartment is very highly regulated and has consequently been well studied. Several cell lines, such as 3T3L1 {84,85}, 3T3-F442A {74,86}, and OM771 {74,86} have served as excellent adipocyte cell culture models yielding significant insight into LPL regulation in this highly differentiated tissue. In vitro, adipose cell lines commence LPL synthesis and secretion upon differentiation from fibroblast like pre-adipocytes to adult adipocytes {84,86}. Generally this process requires the addition of hormones such as insulin to the cell culture {74,84}. In the early stages of cell differentiation, the level of LPL transcription and translation is increased significantly. Shortly thereafter, the cells begin accumulating fat droplets. Considering the natural 65 o — o o C O c o o (0 E O H I I J — I CD 00 g •a "§ "I S 3 c 0 , 0 o g © g 13 g ° - fc J J 5 m CD f | I g i « CD O OT in m GO - CQ o 00 C3 00 00 > s g 0 O^  W < SPl cu CD U CO OT si CO CU S N 1 ~0 OT < S >-, O U . C 2 0 0 c >> o S ' * « S fe s § CU 7 1 % s a c a a.S E 6 6 course of expression, most observations on the effects of many chemicals on normal LPL expression, glycosylation, intracellular transport and secretion have been performed in these types of cell lines (review {87}). Unfortunately these cells are extremely difficult to transfect and adenovirus shows no exception to this rule. Even at very high doses of virus, these cells proved resistant to infection (Figure 3.9) and show minimal alterations in LPL activity after either Ad-RSV or CMV-LPL infection (Figure 3.10). If an effective transfection system was developed, several interesting questions could be asked including if a dose response is possible upon successful infection or if the post-translational regulation of LPL tight enough to limit secretion of functional LPL. Can expression from an exogenous promoter alter the natural differentiation process observed in this cell type (i.e. is differentiation accelerated)? Additionally, would any kind of drug modulation alter the ability of the Ad-derived LPL to be secreted in a manner different to the endogenous expression profile (e.g. insulin{84,85}, fibrates{32},NO-1886{29,30}, FFA{73,74}, heparin{69,122,142})? Ex vivo Although infection proved to be difficult in the immortalized cell line 3T3L1, we were interested in assessing and comparing the ability of Ad-RSV-LPL versus Ad-CMV-LPL to infect and express from primary adipose tissue. Epididymal fat pads were isolated from several LPL +/- mice after a 4 hour fast and cut into pieces of approximately 0.5 g. The tissue portions were directly injected with 25 pi of one of PBS (control), Ad-RSV-LacZ or Ad-RSV-LPL (5xl09 pfu), or Ad-CMV-LacZ or Ad-CMV-LPL (3xl08 pfu) and incubated in 12 well dishes containing 1ml DMEM with 10%FBS for 36 hours. Medium treated with 5U/ml heparin for 15 minutes was assessed for LPL activity and mass with only a slight elevation in activity seen in the media collected from Ad-RSV-LPL and Ad-CMV-LPL treated tissue (0.01 + 0.6 mU/ml controls, 0.62 ± 0.6 RSV-LPL (p=0.05), 3.22 + 0.6 CMV-LPL (p<lxl0'7), Table 3.2). Despite the significance associated with these numbers, they are very low and close to the background of the LPL activity assay; thus it is difficult to definitively compare these activities. 67 Figure 3.9 Staining for P-galactosidase activity in Ad-RSV-LacZ (A) or Ad-RSV-LPL (B) infected 3T3-L1 cells at a MOI=500. Only approximately 5-10% of the cells appear to be infected with Ad-RSV-LacZ in panel (A) as indicated by the blue staining. 68 6 4 H 2 H 0 X 10 50 100 500 Ad-RSV-LPL 500 Ad-RSV-LacZ 0 Figure 3.10 Expression of the human LPL gene in 3T3-L1 cells. Cells were infected with Ad-RSV-LPL for 1 hour, incubated for a period of 48 hrs and subsequently 15 minute heparin-exposed medium was harvested for LPL assay. LPL mass and activity at moi 10-500 determined 48 hours post-infection. The control cells were either Ad-RSV-LacZ (moi 500) or uninfected 3T3-L1 cells and demonstrate some endogenous basal levels of LPL activity. The recorded data are the mean+ SD from three dishes of cells. All measurements were performed in duplicate. *p<0.01 versus controls. 69 Tissue Expression Media Expression Adenoviral LPL Activity LPL Mass LPL Activity vector (mU/ml/lOOmg tissue) (ng/ml/100 mg tissue) (mU/ml/100 mg tissue) PBS Control 8.24 + 5.7 36.00 ± 11.2 0.00 + 0.1 Ad-RSV-LacZ 2.46 + 2.7 23.67 + 3.9 0.00 ±0.1 Ad-RSV-LPL 5.61 ±3.0 5 1 . 2 8 + 1 2 . 3 0 . 6 2 + 0 . 6 Ad-CMV-LacZ 5.69±4.2 31.60 ± 5.1 0.01 ±0.1 Ad-CMV-LPL 3 1 . 0 2 + 2 . 8 1 0 3 9 . 9 3 + 1 8 6 . 1 3 . 2 2 + 0 . 6 Table 3.2 LPL tissue based activity or protein levels after either PBS (control), Ad-RSV or CMV-LacZ or LPL injection directly into isolated murine fat pads (n=4 per group). Bold represents statisitically significant elevations compared to PBS or LacZ control injected tissues, (p < 0.05) 70 The tissues were then washed with PBS, homogenized and assessed for cellular LPL activity and mass. Despite very small changes in LPL activity or mass secreted into the media where the tissue portions were incubated, there were significant changes extant within the tissues. There was a significant increase in intracellular mass (51.3 ng/ml + 12 vs. 23.7 ± 4 Ad-RSV-LacZ and 36.0 ± 11 PBS control treatment, p<0.01) but not activity in the tissues receiving Ad-RSV-LPL (Table 3.2). Upon treatment with Ad-CMV-LPL, there was both a significant increase in activity (31.0 mU/ml + 3 vs. 5.7 + 4 Ad-CMV-LacZ and 8.2 + 6 PBS control, p<0.0002) and a dramatic increase in mass (1039.9 + 186 ng/ml vs. 31.6 + 5 Ad-CMV-LacZ and 36.0 + 11 PBS control, p<0.00005). It is interesting to note that although the tissue based LPL activity in the adipose tissue receiving Ad-CMV-LPL was only approximately 3-6x greater than that found in the control or Ad-treated tissues, the immunoreactive mass levels were approximately 30 times greater than control or Ad-CMV-LacZ and 20 times higher than Ad-RSV-LPL levels. This indicates that under the present conditions, Ad-CMV-LPL has a much higher expression potential even in a tissue, such as adipose, that has very tightly regulated post-translational LPL expression. The very low secreted media based LPL activity indicates that although gene transfer and subsequent transcription and translation within adipocytes in vivo may be possible, the resulting plasma (if one may liken media expression levels to plasma levels) expression from that site may be nominal unless in vivo regulatory mechanisms capable of increasing LPL release are defined. 3.2.4 Myocyte and other cell targets Several studies on LPL regulation within the muscle compartment, as well as other tissues, have also been performed {70,143-145}. Unlike most adipocyte based cell lines, other cell lines do not generally produce LPL spontaneously. However, one advantage to the other cell lines employed in this thesis is that they tend to be more susceptible to transfection. Although slightly resistant to Ad-infection, almost 100% of the cells in the mouse muscle cell line C 2 C 1 2 were infected at a MOI of 500 using the control Ad-RSV-LacZ (Figure 3.11). When infected by Ad-RSV-LPL at a similar MOI, large amounts of LPL 71 Figure 3.11 Staining for p-galactosidase activity in Ad-RSV-LacZ (A) or Ad-RSV-LPL (B) infected C 2 C 1 2 cells at a MOI=500. Virtually all cells infected with Ad-RSV-LacZ stained blue indicating almost 100% infection. Ad-RSV-LacZ contains a nuclear localizing signal resulting in the observed punctate staining. 7 2 specific RNA could be detected by Northern blot analysis (Figure 3.12) as could heparin releasable LPL activity (125 mU/ml/105 cells) in comparison to negligible background expression and activity from Ad-RSV-LacZ treated cells. When expression was compared across several cell lines (Figure 3.13) including human HepG2 hepatocytes, mouse 3T3-L1 pre-adipocytes, mouse C2C12 myoblasts, monkey COS renal cells, and primary cat fibroblasts, it appeared that the transformed HepG2, C2C12 and Cos cells supported the highest LPL expression at a MOI of approximately 50-100. Our in vitro studies with Ad-RSV-LPL also suggest species-specific tropisms for different adenoviral plaque preparations or stocks (Table 3.3). No two viral stocks have the same • expression from all cell lines. For this reason, we subsequently screened 5 different viral stocks of Ad-RSV-LPL in CD1 mice for in vivo activity (see Table 3.3). Although a statistically defined, consistent elevation of LPL activity was not seen, the two stocks, 1 and 4, showed the highest overall expression of activity both in vivo and in vitro and were therefore chosen for amplification. Although the specific nature of this tropism is unclear, it is possible that even slight changes in viral configurations may result in species specific tropism. Characterization of viral preparations in vitro can reveal many features and idiosyncrasies about the vector in question. Although the selection of highly expressing adenoviral vector stocks deemed able to infect primary tissue culture cells from the model organism may not consistently ensure successful in vivo gene transfer, results should infer a greater likelihood of success. 3.3 Discussion In vitro data is essential to the assessment of both expression potential, some levels of toxicity and feasibility of tissue target systems. Data presented in this chapter confirm the ability of human LPL, transferred by an adenovector, to produce bioactive LPL as measured by lipolytic activity in several cell lines of various origins. Due to the hepato-trophism of adenovirus when administered intravenously in vivo, the major target of the majority of our studies was a hepatic cell line HepG2. Adenoviral-mediated LPL gene delivery and expression in differentiated HepG2 liver cells provided a novel investigational tool to study biologic, physiologic and, in this 73 Ctl Ad-RSV LPL Figure 3.12 Northern analysis of Ad-RSV-LPL or control Ad-RSV- LacZ infected C 2 C 1 2 cells 7 4 Cos HepG2 C2C12 3T3L1 CF Control Figure 3.13 Comparison of LPL activity across several cell lines 48 hours after a 1 hour incubation with Ad-RSV-LPL (moi 50-100). The recorded data are the mean+ SD from three dishes of cells except for the control which isrepresenative of the average + SD of all 5 cell lines. All measurements were performed in duplicate. 75 LPL Activity (mU/ml/105 cells or mU/ml) Cell Line Viral Batch 293 C 2 C 1 2 Cat Fibroblasts In Vivo 1 41.2 55.5 74.2 752.2 2 48.6 73.3 104.1 444.2 3 49.1 17.1 33.3 499.9 4 28.3 63.7 51.8 568.7 5 43.0 72.2 65.1 396.3 Control 0.6 0.3 0.9 463.0 Table 3.3 In vitro and In vivo screening of 5 separate CsCl purified viral plaques or batches for LPL activity. In vitro (n=3, MOI=100-200) activity is measured inmU/ml/105 cells and in vivo (n=2) activity is measured in mU/ml of mouse plasma. 76 model, nonconstitutive functions of LPL in dissecting its complex role in lipoprotein metabolism. Overall, our results were novel as this was the first report describing high efficiency in vitro gene delivery, expression and function of human LPL ectopically expressed in an hepatocellular HepG2 model as a result of adenoviral-mediated gene transfer. Our in vitro data was critical to further in vivo analyses and preclinical considerations of liver-directed adenoviral/LPL gene delivery. In the initial studies we utilized a recombinant adenoviral vector containing a human LPL cDNA under the control of a RSV promoter to express LPL in HepG2 cells, a human hepatocellular carcinoma-derived cell line. Although variable chromosomal abnormalities have been described in this cell line, HepG2 cells retain well-differentiated function, expressing many traits attributed to normal human hepatocytes {146}. HepG2 cells are therefore an accepted model for studying lipoprotein metabolism in human hepatocytes {147}. In our experience, this cell line demonstrates poor transfection efficiencies by many other gene transfer methods such as lipofection, electroporation and calcium-phosphate precipitation. However, the present study demonstrated effective gene transfer and successful LPL expression in this naturally non-LPL-expressing, differentiated human hepatic cell line via adenoviral vector gene delivery. From as few as 105 infected HepG2 cells per ml of media supernatant, LPL activity approaching 27% of levels measured from 1 ml of normal human plasma was achieved. A similar level was also attained from murine C2C12 and monkey COS cells after Ad-RSV-LPL gene transfer, however other cell types were only able to accommodate much lower expression levels. Comparison to human plasma activity is meant to provide a more tangible comparison of the expression capacity of our in vitro cell system after adenoviral-mediated gene transfer. The mechanisms which induce liver LPL expression during embryonic development, as well as those leading to its extinction shortly after birth are poorly understood. Despite reports which suggest that liver-specific LPL transcription may be down-regulated by trans-acting hepatocyte-derived extinguisher factors {69}, the HepG2 host cells were clearly amenable to efficient exogenous LPL gene transfer, expression and function in this experimental system. Interestingly, we have also demonstrated that the biological function of the exogenously delivered LPL was fully preserved in the HepG2 hepatic cell system despite the known role adult liver cells perform in LPL turnover, through uptake and 77 irreversible degradation of the enzyme {140}. Our data suggests that the LPL-transduced differentiated HepG2 cells do possess the machinery for appropriate transcription, translation, modification and normal secretion of human LPL. These basic results are not altered despite an inherent Ela activating ability in HepG2 cells, which transactivates Ela-dependent viral genes in viral infection {148}. This feature in fact may have enhanced the in vitro utility in characterizing the cellular, biochemical and potential pathologic effects resulting from non-replicative adenoviral-mediated, ectopic over-expression of LPL in a human hepatic cell line and highlights the potential relevance of this cell line in other Ad-mediated gene expression studies. Overall, these findings imply that the liver could serve as a feasible target for adenoviral mediated LPL gene delivery and expression. Hepatocellular systems have been used to study lipoprotein conversion and uptake after the addition of LPL to HepG2 cell supernatants {149,150} or perfused rat livers {151}, both non-physiologic approaches, due to the immediate LPL-mediated TG hydrolysis in the culture or perfusion medium. In contrast, expression of intracellularly delivered LPL in HepG2 hepatocytes via gene transfer underlies a more ideal model of lipoprotein hydrolysis, occurring at the cell surface rather than directly in the culture medium (for nonheparin-conditioned medium), potentially applicable to other models of lipoprotein metabolism. It is possible that this model system could predispose to increased intracellular fatty acid uptake and immobilization. However, hepatocellular morphology remained unchanged in these in vitro studies. The genetically-modified Ad-RSV-LPL transduced HepG2 cell populations consistently demonstrated high levels of LPL RNA, LPL immunoreactive mass and activity compared to uninfected and adeno-RSV-LacZ controls. The levels of LPL protein and activity peaked at Day 3 and remained stable when re-evaluated again at Day 5. Cellular toxicity, marked by variable degrees of cell rounding and lifting, appeared to be exclusively correlated with MOI >200, with persistent high level expression and normal cellular morphology noted at MOI's of 12.5,25, 50 and 100. Functional assessment using human VLDL as substrate demonstrated rapid VLDL-TG hydrolysis via adenoviral-mediated hepatocellular over-expression of ectopic LPL in this experimental system. Greater than 50% of the added TG was hydrolyzed after 4 hours, consistent with FPLC evidence of a marked decrease in VLDL-TG, versus Ad-RSV-LacZ 78 infected controls. Therefore, our findings reveal that naturally non-LPL producing HepG2 cells can function in lipoprotein/TG processing after LPL gene transfer, in contrast to the poor LPL degradation of VLDL and LPL-derived VLDL remnants documented in normal HepG2 cells {152}. In contrast to most cell lines, including HepG2, fibroblast (preadipocyte) cell lines that can differentiate into mature adipocytes generally spontaneously commence expression and secretion of LPL upon differentiation. This change is induced in cell lines such as 3T3-L l and Ob 1771 by media supplementation. For example, 3T3-L1 cells differentiate when small amounts of a cocktail including insulin, dexamethasone and 3-isobutyl-l-methylxanthine are added to the cells. After 48hrs, the media is changed to include only insulin as a supplement and differentiation occurs. Peak LPL mRNA and activity levels occur about 4-5 days after treatment with full differentiation by about day 7 {84}. Even without inducing differentiation, if cells are left confluent in culture for several weeks, some LPL expression can occur. Due to the endogenous LPL expression in adipocytes, this has been the in vitro model of choice for many of the studies on LPL transcriptional and post-transcriptional regulation. Likewise, studies in cultures of primary adipocytes or biopsied adipose tissue have led to further in vivo clarification. Early studies on diabetic rats and humans revealed the interaction between insulin and LPL expression such that adipose tissue LPL activity is low in the insulin deficient state and is increased after insulin treatment {70}. Similar effects have been seen in culture using both primary and transformed cell lines {71,72}. Effects on regulation have also been observed with many other factors (reviewed in {1,69}). Major regulators include FFA (increased transcription associated with decreased amounts of active dimeric LPL, likely via posttranscriptional feedback inhibition and/or physical dissociation from endothelial heparan sulfate) {73-76}, heparin (increased secretion) {69,122,142}, cytokines such as tumor necrosis factor (decreased transcription) {77,78}, and hypolipidemic drugs such as fibrates and thiazolidiones (increased transcription and activity) {32}. Despite this work, the true complexity of LPL regulation is not well understood. LPL is controlled at the transcriptional level as well as during mRNA processing, transport, translation, posttranslational modification (e.g. glycosylation and processing in the endoplasmic recticulum), dimerization, protein trafficking and finally secretion {69,87,153}. 79 / Additionally, little is known about LPL transport and regulation once outside the cell. LPL is able to bind to the cell that synthesized it and in adipose, up to 80% of this newly synthesized LPL is degraded by the adipocytes under basal conditions {1}. Some LPL escapes the cell surface, possibly via a concentration gradient, moving from the parenchymal cells where it is synthesized, through the vascular endothelial cells that do not appear to degrade LPL, finally to its site on the proteoglycans on the luminal surface of the vascular endothelium {154,155}. Subsequently, LPL is able to bind to lipoproteins and mediate TG hydrolysis. Eventually this LPL leaves the endothelium and moves to the liver where it is degraded {64}. With all of the possible steps for regulation, it is challenging to ascertain which step is most important, even when employing an in vitro model system. In ex vivo experiments it was demonstrated that the majority of synthesized LPL, likely derived from the adenovector, was caught within the tissue. However, the exact location was not determined. Despite the potential benefits to over-expressing LPL in the adipose compartment, gene transfer to this site may not be feasible considering the tight regulation found within this tissue. Comparative studies for Ad-RSV-LPL were also carried out in several other cell lines including muscle (C2C12), kidney (COS), and primary fibroblasts (cat and mouse). Similar to HepG2 cells, both the COS and 293 cell lines are able to transactivate viral genes and thus were able to sustain elevated levels of LPL expression {67,156}. Interestingly, the cell line C2C12 was able to sustain elevated expression at high MOI but this was variable depending on the plaque from which the batch o f adenovector was grown. Even at lower doses of virus (MOI=50-100), this muscle cell line was able to sustain significant levels of LPL expression not observed in control and control infected cells (Figures 3.12, 3.13). Taken together with the development of muscle specific LPL transgenic mice {25,57} and successful expression after naked plasmid DNA delivery {111}, this may suggest that the regulation within the muscle compartment may be less than that observed in the adipose tissue compartment. Thus the muscle may serve as a better site for exogenous LPL gene expression. Of course the caveat to keep in mind is the potential myopathy that may ensue from high local LPL concentrations. Comparison between in vitro and in vivo expression was significant from the perspective that although in vitro expression was not a direct predictor of in vivo potential, it 80 was strongly suggestive. Initially we attempted to use LPL expression and activity derived from 293 cells for picking viral plaques for propagation and in vivo experiments. This cell line transcomplements the missing El A activity in the first generation adenovectors and thus is the cell line of choice for viral generation {67}. Infection of this cell line is very sensitive to MOI since the virus rapidly replicates and high levels of gene expression can be rapidly observed (personal observation). Thus the initial data was misleading and although it allowed us to choose viral stocks able to express LPL, it could not predict the ability to express in vivo. When stock were tested in cell lines without any transactivating genes, we then were able to choose potential candidates for successful in vivo gene expression. In vitro models provide a venue for determining the critical points of regulation beyond transcription by using an exogenous promoter and a delivered gene. Although many potential experiments await a vector system capable of transfecting adipose type cell lines that maintain a high level of LPL regulation, these experiments will be pivotal to in vivo manipulation of gene expression post gene transfer. In summary, the results presented further provide an important and fundamental basis for predicting successful in vivo liver, muscle or fibroblast-directed gene transfer and expression of LPL via an adenoviral gene delivery approach. 81 Chapter 4: In vivo assessment of Adenovirus-LPL efficacy after intravenous, liver targeted delivery The majority of the data in this chapter comes from two publications, as described below, and represents a collaboration between myself and Dr. Guoqing Liu, a post-doctoral fellow in our lab. I participated in or performed all experiments presented in this chapter with the exception of any FPLC analysis, which was work performed by Dr. Liu and is provided herein as additional proof towards the lipoprotein profile alterations in vivo after intravenous Ad-mediated LPL gene transfer. Technical support for the majority of the LPL activity, mass and lipid measurements was provided by Li Miao and Fudan Miao. Relevant Publications: 1. Ashbourne Excoffon K.J.D., Liu, G., Miao, L., Wilson, J.E., McManus, B.M., Semenkovich, C.F., Coleman, T., Benoit, P., Duverger, N., Branellec, D., Denefle, P., Hayden, M.R. and Lewis, M.E.S. 1997. Correction of hypertriglyceridemia and impaired fat tolerance in lipoprotein lipase-deficient mice by adenovirus-mediated expression of human lipoprotein lipase. Arteriosclerosis, Thrombosis and Vascular Biology 17,2532-2539. 2. Liu, G., Ashbourne Excoffon K.J.D., Wilson, J.E., McManus, B.M., Miao, L., Benoit, P., Duverger, N., Branellec, D., Denefle, P., Hayden, M.R. and Lewis, M.E.S. 1998. Enhanced lipolysis in normal mice expressing liver-derived human lipoprotein lipase after adenoviral gene transfer. Clinical and Investigative Medicine 21,172-185. 82 4.1 Introduction Two important aspects for potentiating maximum ectopic gene expression after gene delivery in vivo are specific to the route of vector administration and the modulation of the immune response. There are many aspects to both of these factors and extensive studies employing first generation adenovectors have been reported in the literature. Depending on the route of administration, the adenovirus is capable of varying extents of infection in different tissues {99,105}. For example, intravenous injection largely targets the liver while nasal administration targets the lung {99}. Establishing the most effective route of administration in uncompromised mouse strains allows for a more effective treatment of homozygous deficient patients by targeting the most productive site for LPL synthesis, secretion and function. Human LPL mass and activity, as well as lipid and lipoprotein parameters including TG level, HDL-cholesterol, total cholesterol, and FPLC analysis of the lipoproteins were the main parameters used to characterize successful gene transfer. Additionally, immunological responses were assessed histologically as well as by in vitro and in vivo means. In this present chapter, expression has been characterized after intravenous administration. Adenoviral-mediated gene transfer into mammalian liver is a highly efficient process that can result in transduction frequencies approaching 100%, by access to virtually all hepatocytes through the fenestrated liver endothelium {105,134,135}. The hepatic synthesis of proteins that are either biologically active within liver cells or secreted into the circulation from their site of synthesis in the liver has been augmented through adenoviral-mediated gene transfer. Some of these proteins are important regulators of lipoprotein metabolism including the LDL-R and receptor-associated protein (RAP), apo AI, apo E, cholesterol-7oc-hydroxylase and apo B messenger RNA editing protein {100,101}. LPL, however, is normally synthesized in extrahepatic differentiated tissues {132}yand circulating LPL is efficiently bound and degraded in the adult liver, the major site for LPL degradation in humans {64}. Therefore, liver-directed LPL gene transfer presents as a more complex process. In addition to the ectopic delivery of the LPL gene to the liver followed by de novo synthesis, post-translational modification and secretion, mature active LPL must ultimately be transported and bind to appropriate sites where it is active in the hydrolysis of circulating TG-rich lipoproteins. 83 In chapter 3 it was shown, in vitro, that despite a poorly understood mechanism that rapidly extinguishes LPL expression at the transcriptional level in the newborn liver, the machinery for secretion of active functional LPL from typically non-expressing mature hepatic (HepG2) cells can be utilized after LPL gene delivery and will facilitate hepatocellular-mediated VLDL-TG hydrolysis {68}. Encouraged by these earlier observations, in vivo studies in normal CD1 and heterozygous LPL deficient mice were initiated. These studies aimed to address the following fundamental questions intrinsic to liver-targeted LPL gene transfer and expression; (1) is the mature adult liver in situ able to efficiently express LPL and will LPL undergo appropriate post-translational processing, secretion and transport to any site, where it is active? Furthermore, (2) does ectopically expressed LPL in the mature liver, in vivo, participate in plasma lipoprotein metabolism in a manner similar to LPL derived from its normal sites of synthesis? Using a first generation E1-/E3- adenoviral vector system, our studies have demonstrated that targeted delivery of hLPL to the mouse liver allows normal expression and functional activity of LPL. For initial studies an RSV-LTR promoter was selected. This promoter is characterized by a low level, relatively long term expression that possesses high activity in hepatocytes in vivo {134,137,138}. A high expressing CMV driven adenovector was developed for later studies involved in targeting muscle and adipose tissues as well as the feline studies presented in chapters 5 and 7 respectively. However, initial studies following our well documented intravenous protocol were performed in order to characterize the expression profile of these new adenovectors. This chapter has been divided into several sections with the first section presenting a detailed study of the RSV-driven virus in vivo, followed by a brief comparison to the CMV-driven vector. This latter section will mainly highlight the differences observed between these two vectors upon intravenous delivery. 4.2 RSV Promoter Encouraged by earlier in vitro observations, we proceeded to in vivo studies in normal CD1 mice and heterozygous LPL deficient mice. The importance of exploring LPL gene transfer in normal mice was in part to compare adeno-derived liver based LPL 84 expression to both normal as well as ubiquitously expressing transgenic LPL models. Subsequent exploration of gene transfer in various murine models of lipoprotein abnormalities was then initiated commencing with studies in the slightly dyslipidemic heterozygous LPL deficient mice. Due to the similarity of the data achieved between these two distinct groups of mice, data from only the heterozygous study will be presented with a final comment indicating the only difference observed between these studies. 4.2.1 Ad-RSV-LPL Gene Transfer to Heterozygous (+/-) LPL Deficient Mice Due to the lethality of complete LPL deficiency in mice, presumably due to congestion of the peripheral and pulmonary circulation by large TG-rich lipoproteins, mice heterozygous for LPL deficiency due to a targeted carboxy-terminal disruption {33}, were employed. LPL heterozygous mice have an approximate two to three-fold elevation in TG levels associated with decreased post-heparin plasma (PHP) LPL activity levels, in comparison to their wild type (WT) litter mates. We adopted this model of murine heterozygous LPL deficiency {33} to evaluate the alteration in hypertriglyceridemia, lipoprotein profile, lipoprotein composition and tolerance to intravenous fat loading, in vivo, after liver-targeted, adenoviral-mediated LPL gene transfer. Expression Levels Dose-response analysis of Ad-RSV-LPL expression in C57BL/6 mice Initially C57BL/6 WT litter mates from the murine LPL gene targeted kindred were used to evaluate the ideal dosage and in vivo utility of our Ad-RSV-LPL vector. The mice (n=3 per group) were intravenously injected with Ad-RSV-LPL at three different doses: lxlO9, 2.5xl09 and 5xl09 pfu; or Ad-RSV-LacZ at 5xl09 pfu. The dose-response relationship at Day7 according to both human-specific LPL activity and immunoreactive mass in PHP is shown in Figure 4.1. Clearly, human specific LPL mass and activity increased proportionally with increasing dosage of Ad-RSV-LPL, relative to negligible levels from the Ad-RSV-LacZ injected control mice. Peak levels of activity (388.7±68.0 mU/ml) and mass (7136.9±1193.6 ng/ml) were achieved in these WT mice using 5xl09 pfu of Ad-RSV-LPL. The latter dosage was therefore chosen as the maximum desirable dose in all subsequent 8 5 • Human LPL Activity • Human LPL Immunoreactive Mass Figure 4.1 Dose response analysis (1, 2.5 and 5xl09 pfu) of adenovirus-mediated expression of human LPL immunoreactive mass (ng/ml) and activity (mU/ml) in Ad-RSV-LPL and Ad-RSV-LacZ infected wild type mice. 86 mouse infections for inducing meaningful enhancement of LPL function and lipoprotein profile. Hepatic expression of human LPL in vivo We next systemically administered 5x109 pfu of Ad-RSV-LPL (n=9) or Ad-RSV-LacZ (n=9) by tail vein injection to mice heterozygous for targeted disruption of the LPL gene. Livers, lungs and spleens from both Ad-RSV-LPL and Ad-RSV-LacZ animals were harvested 7 and 42 days after the administration of each vector and at the end of the experiment on Day 60. The adenoviral-mediated expression of the human LPL gene in mouse liver was initially confirmed at Day 7 by in situ hybridization analysis using an LPL riboprobe (Figure 4.2B) at which time it was obviously positive in the majority of hepatic cells. By comparison, negligible hLPL transcript was detected from either spleen or lung tissues at Day 7, and no signal was detected using the sense control LPL riboprobe in any tissues (data not shown). No endogenous mouse LPL mRNA was detectable in Day 7 Ad-RSV-LacZ infected mouse liver, confirming the riboprobe to be human-specific (Figure 4.2A). At Day 42 (Figure 4.2C), LPL expression was significantly decreased and upon sacrifice at Day 60 (Figure 4.2D), no detectable Ad-RSV-LPL message was apparent in the liver. 1 Total lipase activity determined from the plasma and liver is due to the combination of hepatic lipase (HL) and LPL. The respective activity for each enzyme is measured by performing the assay in the presence or absence of 1 mol/L NaCl which, when present, differentially inhibits LPL but not HL activity. As seen in Table 4.1, the total lipase activity from Day 7 post-infection liver tissue homogenates of mice receiving Ad-RSV-LPL was significantly elevated in comparison to Ad-RSV-LacZ controls (432.8±140.9 vs. 188.7± 17.3 mU/g liver; p=0.02). The liver homogenate of both Ad-RSV-LacZ and Ad-RSV-LPL mice contained similar amounts of lipase activity not inhibited by 1 mol/L NaCl (p>0.05) and represents HL. Therefore, the additional inhibited activity of 255.4±131.0 mU/g liver in the livers of mice receiving Ad-RSV-LPL is consistent with this being human-specific LPL activity (p=0.015). In association with this finding, human LPL immunoreactive mass from Ad-RSV-LPL treated mice was 1306.9± 814.4 ng/g liver versus undetectable LPL mass in Ad-RSV-LacZ controls (p=0.02). 87 Figure 4.2 ISH analysis of Ad-RSV-LacZ and AD-RSV-LPL-infected mouse livers. The adenovirus-mediated expression of the human LPL gene in mouse liver was confirmed at day 7 by ISH analysis using an LPL riboprobe (B). In contrast, no endogenous mouse LPL message was detectable on day 7 Ad-RSV-LacZ-infected mouse liver, confirming the riboprobe to be human specific (A). On day 42 (C) and at sacrifice on day 60 (D), minimal to absent Ad-RSV-LPL message was apparent, in keeping with episomal loss of the LPL transgene and/or extinguished LPL gene transcription. 88 Lipase Activity, mU/g Liver* After 1 mol/L NaCl Inhibition Total Residual (HL) (LPL + HL) Inhibited (LPL) Human LPL Protein Mass, ng/g Livert Ad-RSV-LacZ 188.7± 17.3 160.3 ± 14.2 28.4+ 18.8 0.0 Ad-RSV-LPL 432.8+ 140.9 177.4 + 15.0 255.4+ 131.0 1306.9 + 814.4 0.02 0.38 0.015 Table 4.1 Lipase activities (Total [LPL + HL], HL and LPL) in liver homogenates from mice injected with Ad-RSV-LacZ (n=4)or Ad-RSV-LPL (n=4). Values are mean + Standard deviation. * 1 mU= lnmol free fatty acid per minute, t As detected by monoclonal antibodies 5D2 and 5F9 89 Detection of Human-specific LPL in Mouse Post-Heparin Plasma It should be noted that ectopic expression of human LPL in the liver is also accessible to the plasma compartment, since intravenous injection of heparin resulted in a rapid release of human LPL into the blood stream. As shown in Figure 4.3A,B, on Day3 after Ad-RSV-LPL administration, both human LPL mass and activity in PHP could be detected (Figure 4.3B), yet no significant change in total LPL activity (mouse + human; Figure 4.3A) versus Ad-RSV-LacZ controls was apparent. Subsequently, peak human LPL expression was achieved on Day7. Total LPL activity in PHP was 833.8± 132.5 mU/ml in Ad-RSV-LPL mice, 2.7 times higher than Ad-RSV-LacZ controls, of which 70% was human-specific. HLPL immunoreactive mass was 9343± 1974.7 ng/ml in PHP, 8.5 times higher than levels typically detected from normal human PHP measured in this laboratory by the same method (1100± 100 ng/ml; n=10). Both human LPL activity and mass decreased to 393.7 mU/ml and 2600 ng/ml respectively at day 14 in accordance with reduced total LPL activity of 573.2±304.9 mU/ml. The difference in total LPL activity in PHP was indistinguishable at or beyond day 28 between Ad-RSV-LPL and Ad-RSV-LacZ treated animals, due to the disappearance of detectable human-specific LPL activity. However, human LPL immunoreactive mass remained significantly elevated at 478.0 ng/ml (p<0.05) until at least 42 days after the administration of Ad-RSV-LPL (Figure 4.3B). At Day 60, no detectable human LPL activity or protein was observed in PHP samples. Lipid Changes Correction of hypertriglyceridemia in LPL +/- mice As a direct consequence of hepatic expression of human LPL, the moderate hypertriglyceridemia in heterozygous LPL +/- mice was corrected and inversely correlated with levels of PHP LPL (Figure 4.4A). Plasma TG levels dropped sharply from 234 to 48 mg/dl 3 days after administration of Ad-RSV-LPL and decreased maximally to 15 mg/dl at Day 7 (p<0.001). Peak expression of human LPL began to decline after Day 7 and as a result TG levels rose gradually, returning to the endogenous levels seen in Ad-RSV-LacZ control mice by Days 42 and 60, correlating with the disappearance of human-specific LPL activity or mass. It should be noted that plasma TG levels were unexpectedly reduced by approximately 30% in Ad-RSV-LacZ mice 3 days after viral administration, and plateaued 90 0 10 20 30 40 50 60 Days post-infection B 0 0 0 10 20 30 40 50 60 70 Days post-infection Figure 4.3 Total and human LPL (H-LPL)immunoreactive mass (ng/mL) and activity (mU/mL) from Ad-RSV-LacZ-and Ad-RSV-LPL treated mice.A, Total (human +mouse) plasma LPL activity from Ad-LPL and Ad-LacZ mice (n=5, both groups) over 60 days. B, Human-specific (H-LPL) activity and immunoreactive mass in Ad-RSV-LPL mice (n=5) was assessed by 5D2 monoclonal antibody inhibition and ELISA. Total plasma LPL activity was significantly increased from day 3 to 28 after administration of Ad-LPL relative to Ad-LacZ controls (A). By day 28, H-LPL activity had declined to undetectable levels, even though LPL immunoreactive mass was still significantly elevated (P<0.005, n=5), which persisted to at least day 42 after infection (P<0.005, n=5). 91 A 5 E O Ad-LacZ Ad-LPL 10 20 30 40 50 60 Days post-infection B 60 h O) E 9 30 Iff* Q x T3 mg/ 100 o s-Q) 75 W (V O 50 .c O 2 25 o h-Ad-LacZ Ad-LPL 10 20 30 40 50 60 Days post-infection ^ •* ==--1 Ad-LacZ Ad-LPL 0 10 20 30 40 50 60 Days post-infection Figure 4.4 Lipoprotein profile in A d - R S V - L P L and Ad-RSV-LacZ infected mice heterozygous for L P L deficiency. A , At day 7 after A d - R S V - L P L delivery (n=5), plasma T G decreased maximally to on tenth of corresponding T G values from Ad-RSV-LacZ infected controls (n=5). B , In addition to lowered plasma T G levels, H D L - C in A d - R S V -L P L mice was stable relative to a decrease in Ad-RSV-LacZ infected mouse control. C, TC values were not different between the two groups throughout the 60 days. 92 thereafter. Surprisingly, plasma HDL-C levels were also reduced nearly 40% (from 107 to 70 mg/dl) at Day 3 after Ad-RSV-LacZ injection, yet remained unchanged at baseline after Ad-RSV-LPL administration (Figure 4.4B). Similar changes observed in total cholesterol (Figure 4.4C) likely reflect this alteration in HDL-C since most of plasma cholesterol in mice is carried in the HDL fraction. As revealed by FPLC (Figure 4.5), the plasma lipoprotein profile of mice heterozygous for targeted disruption in the LPL gene, in comparison with WT litter mates, is characterized by a much larger TG peak in the VLDL fraction (Figure 4.5A,B). This peak was profoundly decreased 7 days after administration of Ad-RSV-LPL (Figure 4.5D) relative to uninfected wild-type mice (Figure 4.5A), or LPL +/- mice receiving Ad-RSV-LacZ (Figure 4.5 C,E). This confirmed that the correction of hypertriglyceridemia by administration of Ad-RSV-LPL was due to an enormous reduction of CM/VLDL-derived TG. The alteration of HDL in mice receiving Ad-RSV-LPL, as analyzed by FPLC, was consistent with the levels of HDL-C measured from plasma by the quantitative precipitation method. A broader and higher peak correlated with a relative enrichment of a larger sized isoform in the HDL fraction and was evident at Day 7 after Ad-RSV-LPL administration (Figure 4.5D), resolving by Day 60 (Figure 4.5F). The nature of this larger HDL species, whilst compatible with increased LPL-mediated conversion to HDL2, was not identifiable by our current assay procedure and is under further investigation. Correction of impaired fat tolerance by hepatic expression of hLPL Although the moderate hypertriglyceridemia observed after an overnight fast in heterozygous LPL knockout mice was alleviated via adenoviral-mediated, liver-targeted expression of human LPL, it was of further interest to test whether tolerance to an acute fat load could be modified in these mice. In preliminary experiments it was found that the response to intravenously delivered fat was greatly impaired in heterozygous LPL mice in comparison to WT littermates (data not shown). Subsequently, an intravenous fat tolerance test was performed on WT untreated, or LPL +/- mice receiving either Ad-RSV-LacZ or Ad-RSV-LPL (n=4, each group). Figure 4.6 confirms that the clearance of lipids injected I.V. via bolus infusion was significantly prolonged in LPL +/- mice administered Ad-RSV-LacZ, compared to WT litter 93 c o +5 O 3 100| 50 c o '+3 O §> 100 50 Wild Type ^ v B +/- Day 0 100 50 0 10 20 30 40 50 Fraction No C Ad-LacZ, Day 7 i B • | iod re 10 20 30 40 50 Fraction No Ad-LPL, Day 7 ™ 50 o> 3 10 20 30 40 50 Fraction No 10 20 30 40 50 Fraction No c o <3 O re 1001 C 50 o> 3 Ad-LacZ, Day 60 10 TG Choi 20 30 40 Fraction No % 100 2 14-- 50 o 3 Ad-LPL, Day 60 50 10 20 30 40 Fraction No 50 Figure 4.5 FPLC analysis of lipoprotein composition for plasma sampled from Ad-RSV-LacZ and Ad-RSV-LPL injected LPL heterozygous (+/-) mice at 0, 7, and 60 days. Lipoprotein profile at day 7 from Ad-RSV-LPL-infected +/- mice (D) showed a significant decrease in the TG content of VLDL and CM relative to baseline values from day 0 WT mice (A) or day 0 LPL +/-mice (B). The decrease in the TG fraction of Ad-RSV-LPL infected +/- mice relative to Ad-RSV-LacZ-infected +/- controls was most significant at day 7 (C and D). A broadening of the HDL-C peak was demonstrated at day 7 in Ad-LPL +/- mice (D), which may represent HDL conversion to an alternate, larger HDL isoform such as H D L 2 . At day 60 (E and F) the peaks normalized, in agreement with the time course observed for hepatic LPL expression. 94 «•••> Ad-LacZ (+/-) Ad-LPL (+/-) • - WT (+/+) 5 E O CS) o 1000 100 10 Hours 15 20 25 Figure 4.6 Intravenous fat-load tolerance in A d - R S V - L P L and Ad-RSV-LacZ injected L P L +/- mice and uninfected wild type (+/+) littermates. Fat loads consisted of 250 pi of 20% Intralipid administered via intravenous bolus infusion at day 7. Plasma T G levels (mg/dL) were measured at the indicated time intervals. 95 mates. However, the clearance curve of Ad-RSV-LPL treated heterozygous mice remarkably parallels that observed in WT mice. Thus, the capacity to restore normal postprandial LPL-mediated hydrolysis of serum triacylglycerols is clearly apparent from these studies in heterozygous LPL mice after a single injection of Ad-RSV-LPL. 4.2.2 Comparison of Gene Transfer to Normal CD1 Mice versus LPL +/- Mice Although the results from these two studies were remarkably similar, it is interesting to note the one major difference. Whereas Ad-RSV-LPL gene transfer to LPL +/- mice was able to sustain the levels of HDL and TC in comparison to the small drop observed in the control group that received Ad-RSV-LacZ, the opposite was true in normal CD1 mice. When comparing the Ad-treated cohorts of CD1 mice, a 20 to 30% reduction in HDL-C values was observed in the group that received Ad-RSV-LPL at Day 14 (p=0.024) and Day 28 (p=0.0001) post-infection, in comparison to the control Ad-RSV-LacZ cohort. A significant decrease was noted in TC in the Ad-RSV-LPL group (versus Ad-RSV-LacZ) only at Day 28 (p=0.003). The cause of this difference is not completely understood but may partially represent genotype or strain-related differences in metabolism or immune response. 4.2.3 Ad-RSV-LPL Gene Transfer to Human LPL Transgenic Mice It has been shown several times in the literature that gene expression from first-generation adenovectors may be prolonged in the absence of an immune response to the exogenous gene product {104,136,157-159}. It was hypothesized that an immune reaction in mice to hLPL would likely be observed since several monoclonal antibodies (mAbs) were generated to bovine LPL in mice {129}. These antibodies cross react with bovine and human LPL in a specific way that does not recognize mouse LPL. Thus these Abs have been employed in a multitude of experiments, including this thesis, especially when differentiation between human and mouse LPL is desired. To test the potential length of LPL expression using our Ad-RSV-LPL, a muscle expressing human LPL (hLPL) transgenic mouse model back crossed onto a C57BL/6 background was employed {57}. If an immune response to the hLPL gene was a major mechanism leading to the short term expression, exposure to the human protein throughout life should make these mice tolerant 96 to gene transfer of the hLPL gene and thus prolong expression using our Ad-RSV-LPL vector. LPL Expression and Plasma Lipid Levels When a small cohort of hLPL muscle transgenic mice (n=4) were given 5x109 pfu of either Ad-RSV-LPL or LacZ, LPL activity and immunoreactive mass increased over baseline levels of 417 + 145 mU/ml and 337 + 129 ng/ml, respectively in both groups, peaking at Day 7 (Figure 4.7A,B). The group receiving Ad-RSV-LPL had approximately a 3-fold increase in both activity and mass over baseline (1230 + 203 mU/ml and 1283 + 496 ng/ml respectively). Interestingly, Ad-RSV-LacZ experienced an approximate 2-fold increase in LPL activity and mass (934 + 80 and 741 + 160, respectively). It was unexpected that the group receiving Ad-RSV-LacZ would observe such an increase in both LPL activity and mass, and may represent a non-specific activation of either the endogenous LPL gene or the muscle expressing transgene. The reason for this is unclear but serves to highlight the importance of including a control virus in all experiments of gene transfer. The increase in LPL expression was transient with LPL activity and mass levels decreasing and becoming similar to the control LacZ levels by Day 28. This suggests that the transient expression observed in these, the normal and the heterozygous LPL deficient mice is not due to an immune reaction to the hLPL gene product. This transient expression may have been due to an immune reaction to the adenovirus, promoter shut-off, elimination of adeno-infected cells or some combination of these or alternative mechanisms. When investigating the lipid profile of these LPL muscle transgenic mice, it was observed that the plasma TG levels in the mice receiving Ad-RSV-LPL dropped to approximately 40% of baseline at Day 7 (22.4 + 3 mg/dl) (Figure 4.8 A). Plasma TG also dropped slightly to 71% of baseline in mice given Ad-RSV-LacZ. At this time there was still a significant difference between these two cohorts (p=0.03) but both groups returned to background and equivalent levels by Day 28. Similar to previous studies, total cholesterol (TC) dropped by 32% at day 7 in the group receiving LacZ (p=0.05 vs. baseline), with the majority of this drop being in the HDL-C fraction (p=0.02 vs. baseline) (Figure 4.8B,C). Both TC and HDL-C did not change from baseline in mice receiving Ad-RSV-LPL. This again illustrates a possible compensatory effect of the exogenous LPL in maintaining HDL 97 A Q. 0 H 1 1 1 0 10 20 30 Time (days) B 2000 n > S3 .—* ra E 2 o> O C C — ' 2 .Ei 000 E o> _l Q. a. • Ad-RSV-LPL Ad-RSV-LacZ Time (days) Figure 4.7 Changes in LPL activity (A) and human specific immunoreactive protein (B) after Ad-RSV-LPL or LacZ gene transfer to human LPL muscle transgenic mice at a dose of 5xl09 pfu. Due to an unexpected increase in background (represented by control Ad-RSV-LacZ mice) LPL levels, significantly elevated mass and activity was only observed at day 7 post-gene transfer (p<0.05). This data reveals the importance of a control virus group in studies of viral gene transfer. 98 •Ad-RSV-LPL Ad-RSV-LacZ 10 20 Time (days) 30 B ~ 75 -| O) E £ tt> a> o £ 9 _ i Q 50 i 25 *— Ad-RSV-LPL • • Ad-RSV-LacZ s - i 10 20 Time (days) 30 B in .—. 5 f 2 o l -Ad-RSV-LPL •^•Ad-RSV-LacZ 10 20 Time (days) 30 Figure 4.8 Changes in plasma triglycerides (A), HDL-cholesterol (B) and total cholesterol (C) after Ad-RSV-LPL or LacZ gene transfer to human LPL muscle transgenic mice at a dose of 5x109 pfu. A significant decrease from baseline TG levels was observed in both groups of mice at day 7, however the level in Ad-RSV-LPL mice was significantly lower than that of Ad-RSV-LacZ mice. As observed in LPL +/- mice, there was only a significant decrease in the HDL-C of the Ad-RSV-LacZ cohort of mice (p<0.05). No change in total cholesterol was observed. 9 9 levels in the face of a cholesterol lowering viral burden in another murine model in a manner similar to that seen in heterozygous LPL deficient mice. 4.3 CMV Promoter Similar studies were performed with a different adenovirus utilizing the CMV promoter to drive expression of the human LPL cDNA. These adenovectors were mainly developed for studies within more difficult target tissue compartments and for the feline model for LPL deficiency. Due to the similarities of these studies, only the significant differences from the RSV studies are included herein. Three different vectors were developed each containing a different gene. Ad-CMV-LPL contained the same cDNA construct as the RSV-LPL virus and it was hypothesized that these two vectors would then be comparable. Two control viruses were also developed. Ad-CMV-LacZ was slightly different since although it was the same bacterial 0-galactosidase gene, this adenovector did not contain the nuclear localizing signal that was included in the RSV construct. Because LPL can be monitored in the plasma, it was desirable to have a secretable marker gene, not involved in lipoprotein metabolism. Thus Ad-CMV-AP was developed using a secretable human alkaline phosphatase gene. The normal function of this enzyme is to convert pyridoxal 5'phosphate (PLP) to pyridoxal. Elevated serum AP has been observed in humans and is not associated with any disease {160}. This corresponds well with our observations that mice over-expressing AP do not demonstrate any abnormal behaviors or disease-like characteristics. 4.3.1 Ad-CMV-LPL Gene Transfer to Heterozygous LPL Deficient Mice Initial dose response characterization of this group of vectors proved them to be highly effective at relatively low doses (data not shown). As demonstrated in Figure 4.9A,B, a 10-fold decrease in viral load, in comparison to 5xl09 pfu of Ad-RSV-LPL, could produce slightly higher activity and mass levels (Day 7,955 + 181 mU/ml and 4451 + 1469 ng/ml, respectively) with peak levels occurring at day 3 (1454 + 39 mU/ml and 7406 + 3433 ng/ml, respectively). Interestingly, despite similar activity and mass, a dose of 5x10 pfu per mouse somehow resulted in a virtual elimination of plasma TG (0.45 + 2.24 mg/dl, Figure 4.10). Correspondingly, there was a very significant reduction of both total and HDL cholesterol 100 Figure 4.9 Time course of A) LPL activity and B) human specific immunoreactive mass in LPL +/- mice receiving 5xl08 pfuAd-CMV-LPL or Ad-CMV-AP. All time points, except for baseline measures and the LPL activity measure 42 days after treatment, were significantly increased in mice receiving Ad-CMV-LPL (n=3) over mice receiving Ad-CMV-AP (n=3), p<0.05. 101 Time (Day) Figure 4.10 Time course of changes in plasma triglycerides (TG) in LPL +/-mice receiving 5x108 pfu Ad-CMV-LPL or Ad-CMV-AP. TG levels were significantly reduced in mice receiving Ad-CMV-LPL (n=3) over mice receiving Ad-CMV-AP (n=3) (p<0.05) up to day 28 post-injection. Although a trend is obvious for lower TG in the Ad-CMV-LPL cohort beyond day 28, an unexpected decrease was observed in the Ad-CMV-AP group of mice at this time. The etiology of this decrease is currently unknown. 102 (17 + 4 and 9 + 1 mg/dl respectively at day 3 vs. 45 + 4 and 35 + 2 mg/dl at baseline, pO.OOOl) indicating that this decrease in plasma TG likely affected all lipoprotein classes. Surprisingly, expression from these CMV-driven vectors lasted for more than 60 days with mass and activity remaining significantly elevated at the end of the time course experiment (60 days, p<0.05, Figure 4.9A,B). Plasma TG was also significantly decreased at this point (p=0.04, Figure 4.10). This was unexpected in consideration of the short time course of expression (~2 wks) for most published CMV-driven vectors {94,97,98}, and was at least 2 weeks longer expression than the RSV-based viruses. This may have been a direct result of a reduction in toxicity due to a reduced viral load producing a lower hepatic based immune response leaving a larger number of healthy hepatocytes expressing bioactive LPL for a longer period of time. Alternatively, the CMV-promoter may have had a higher transcription or translation rate resulting in a higher rate of secretion, or perhaps more efficient secretion or stability of bioactive dimeric LPL. Nevertheless, these results demonstrate the highly efficient and effective nature of this group of viral vectors. Similarly, plasma expression after control Ad-CMV-AP delivery showed a prolonged time course (Figure 4.11). Although the level had dropped from a peak level of approximately 55 U/ml sustained from day 3 to day 14, it was still significantly elevated at day 60 (2.36 vs. 0.13 U/ml control, p=0.03). These results indicate that there was likely no immune response to either the hLPL protein product or human secreted AP. It is unknown how long a significant level of expression, with a corresponding effect on plasma TG, would have been sustained in these mice, however at 60 days it was near baseline and control levels and likely would have extinguished shortly thereafter. 4.3.2 Ad-CMV-LPL Gene Transfer to LPL Transgenic Mice In order to confirm the results observed in the muscle LPL transgenic mice with Ad-RSV-LPL in another transgenic mouse model, similar experiments were carried out in the CMV-driven LPL cDNA transgenic mice developed in our lab in the early 1990's {26}. Additionally, this time course involved a comparison of Ad-CMV-LPL and AP. 103 Ad-CMV-LPL Ad-CMV-AP 0.01 H 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 Time (Day) Figure 4.11 Time course of changes in plasma alkaline phosphatase activity (AP) in LPL +/- mice receiving 5x10s pfu Ad-CMV-LPL or Ad-CMV-AP. Except for baseline (Day 0), AP levels were significantly increased in mice receiving Ad-CMV-AP (n=3) over mice receiving Ad-CMV-LPL (n=3) (p<0.05) for the entire period of the experiment (60 days). 104 The results were almost identical to those found in the section above. Peak LPL plasma activity and mass were observed at day 3 (1284 + 65 mU/ml and 6135 + 1453 ng/ml, respectively, Figure 4.12). Activity remained significantly elevated until at least day 42 (p<0.01). A trend towards elevated activity and significantly elevated mass (p<0.05) was still observed at day 60. Plasma TG levels were still dramatically decreased at day 60 however, TG in the control group receiving Ad-CMV-AP also dropped unexpectedly by this time point (Figure 4.13). This alteration in plasma TG was not observed in previous experiments and the etiology is currently unknown. As described above, plasma levels of AP after Ad-CMV-AP gene delivery also remained elevated at day 60 (Figure 4.14, p=0.04). 4.4 FFA clearance and VLDL clearance/distribution One of the major concerns of a hepatic target for LPL gene therapy is the lack of normal expression within this site beyond neonatal stages. The reason for this is most logically because this is the site of remnant uptake and VLDL production. LPL is implicated in not only the breakdown of TG within the VLDL particles but also consequently their uptake {161}. LPL has also been implicated in the uptake of other lipoprotein particles such as LDL through a postulated bridging activity between the heparan sulfate proteoglycans or receptors on the cell surface (e.g. LDL receptor related protein (LRP)) and the lipoprotein particle {162-164}. This suggests that LPL expression on the hepatic endothelium could potentially have profound effects on the overall lipoprotein metabolism of the entire organism. This correlates well with the fact that LPL expression in the liver can be reinduced in periods of disease such as cancer {165,166}, bacterial infection {167} or after resection of the liver (i.e. hepatectomy) {168} when the dietary needs of the liver and the body change. Over-expressing LPL in the liver in a healthy individual may cause some disease like symptoms such as cachexia and hypothetically the creation of a fat laden liver due to the creation of a virtual VLDL trap. The effect of this state may be more exaggerated in a completely deficient LPL patient where a VLDL trap is almost inevitable due to the lack of LPL in other tissues to hydrolyze the TGs in any VLDL that may escape the liver derived LPL. Merkel et al. directly tested this by creating a liver based LPL transgenic mouse {42}. On a normal background there was a significant decrease in plasma TG but no apparent 105 --.1500 -I A) Time (Day) Figure 4.12 Time course of A) LPL activity and B) human specific immunoreactive mass in LPL muscle transgenic mice receiving 5x108 pfuAd-CMV-LPL or Ad-CMV-AP. All time points, except for baseline measures and the LPL activity measure 60 days after treatment, were significantly increased in mice receiving Ad-CMV-LPL (n=4) over mice receiving Ad-CMV-AP (n=4), p<0.05. 106 100 i 0.1 ~\ i 1 1 1 1 1 1 0 10 20 30 40 50 60 70 Time (Days) Figure 4.13 Time course of changes in plasma triglycerides (TG) in LPL muscle transgenic mice receiving 5x108 pfii Ad-CMV-LPL or Ad-CMV-AP. TG levels were significantly reduced in mice receiving Ad-CMV-LPL (n=3) over mice receiving Ad-CMV-AP (n=3) (p<0.05) up to day 60 post-injection. The etiology of the unexpected decrease observed in the Ad-CMV-AP group of mice at later time points is currently uriknown. 107 0.01 H 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 Time (Days) Figure 4.14 Time course of changes in plasma alkaline phosphatase activity (AP) in, LPL muscle transgenic mice receiving 5x10s pfu Ad-CMV-LPL or Ad-CMV-AP. Except for baseline (Day 0), AP levels were significantly increased in mice receiving Ad-CMV-AP (n=4) over mice receiving Ad-CMV-LPL (n=4) (p<0.05) for the entire period of the experiment (60 days). 108 histological alterations indicating that fat accumulation in the liver was not occurring. When bred onto a completely LPL deficient background the histological picture was significantly different with high levels of lipid droplets found in most hepatocytes. Only the mice expressing 10-20% of normal LPL levels survived whereas mice expressing higher liver derived levels died shortly after birth (personal communication). The mice that survived experienced a cachexic like state prior to weaning with elevated plasma TG and lack of weight gain. After weaning these mice survived fairly well although a futile cycle of VLDL uptake and production by the liver was strongly indicated. This suggests that the liver may be a suitable site for LPL gene transfer in normal or partially deficient patients but may cause some toxicity in the complete deficient state. To further evaluate this utilizing our adeno-based gene transfer system, a radioactively labeled lipid approach was attempted. Using 3H-oleic acid in a group of mice, H-VLDL was able to be obtained. The clearance of this labeled VLDL was then monitored in the plasma of mice given either Ad-CMV-LPL or Ad-CMV-AP. Additionally it was hypothesized that accumulation of this label may occur in the livers of the mice receiving Ad-CMV-LPL but not AP. 4.4.1 FFA clearance: +/+ mice vs. +/- mice Initial experiments were performed to observe the production/degradation rate of 3 H -VLDL after the injection of 50-100 uCi of 3H-oleic acid. Approximately 60-70% of the radiolabeled FFA was successfully dissolved in an aqueous BSA-saline solution. A volume of 150 ul containing 50-100 uCi of label was then injected via the tail vein and blood samples were taken 2, 30,45, 60 and 90 minutes after injection. Both plasma and extracted lipid separated by thin layer chromatography (TLC) were monitored for 3 H label. Pilot experiments revealed that more counts in the form of TG could be found in the plasma of heterozygous LPL deficient mice than in wild type littermates 30-60 minutes after injection (Figure 4.15). The peak amount of TG based label occurring at 30 minutes but the peak TG/FFA ratio occurring at 90 minutes. Because a large amount of VLDL was desired, blood was isolated by heart puncture from a group of 10 LPL +/- mice between 30-40 minutes after oleic acid injection and placed in tubes containing approximately 50ul of 100 mg/ml EDTA. After a brief spin, the plasma was then isolated and subsequently 3 H-VLDL was 109 A Time (min) B Time (min) 2 -I Time (min) Figure 4.15 Clearance of 3Holeate after injection into wild type or LPL +/- mice. Lipids were extracted, separated by thin layer chromatography and theft-counts were measured in a scintillation counter. Rate of clearance and appearance of oleate (FFA) and triglycerid^TG) respectively were determined in A) LPL +/- mice and B) wild type littermates. C) The ratio of TG to FFA in the groups of mice (n=5). Peak plasma radiolabeled TG levels were determined to occur approximately 30 minutes after oleateinjection. 110 isolated by two rounds of ultracentrifugation. Greater than 90% of lipid isolated in the VLDL fat cake, after ultracentrifugation, was in the TG band upon TLC analysis indicating that most of the label should be in the form of VLDL-TG (data not shown). 4.4.2 VLDL clearance/distribution: Ad-CMV-LPL vs. AP treated mice Two groups of 5 heterozygous LPL deficient mice were injected with either Ad-CMV-LPL or Ad-CMV-AP. At day 3, pre-heparin plasma only was sampled and expression was confirmed by monitoring plasma TG and AP levels (data not shown). Seven days post injection, mice were fasted for 2 hours and then injected with approximately 300,000 cpm of H-VLDL in a 300ul volume. Blood was then sampled 2, 5, 15,30 and 60 minutes after injection. When crude counts were measured directly in the plasma, at 2, 5 and 15 minutes, there was approximately 50% as much radiolabel in the group receiving LPL versus the AP group (Figure 4.16A). This is likely due to the lower initial plasma level of TG such that the addition of the VLDL intravenously was either diluted or sequestered into pools outside of the plasma compartment or cleared more quickly than the AP mice with normal plasma levels of TG. When calculated as a percentage of radioactivity at 2 minutes the statistical difference was lost and clearance curves appeared to be very similar (Figure 4.16B) indicating that there was no difference in clearance rate beyond 2 minutes post injection. Due to the small amounts of blood taken at each time point and the small amount of radioactivity detected by directly measuring the plasma, lipids were not extracted from these plasma samples. However, total lipids were extracted from the liver at sacrifice 60 minutes post injection. The rationale was that if a VLDL trap was present, most of the radioactivity in the mice receiving LPL should be present in the liver of these mice. Unfortunately there was no difference in 3 H between these two groups within the liver compartment indicating that if there is a VLDL trap present, it is not obvious by these methods (data not shown). 4.5 Discussion The studies presented within this chapter demonstrate the therapeutic potential of adenoviral-mediated delivery of exogenous LPL to the liver. In the groups of mice studied, the lipid profile was altered to one associated with a decreased risk of CAD. Additionally, 111 2000-•2 1500 -1000-500-0-n E (0 J2 Q. "5 o t o a. O Ad-LPL Ad-AP "lO 20 30 40 50 60 70 Time (min) p<0.05 O) c "E « E cu i— E a u 100 H 50 Ad-CMV-LPL Ad-CMV-AP i— 10 - i — 20 30 40 Time (min) —i— 50 —i— 60 —i 70 Figure 4.16 Clearance of 3H-VLDL in LPL +/- mice after receiving 5xl08 pfuof either Ad-CMV-LPL or AP. A) Raw counts per minute (CPM) in the plasma of adenoviral treated mice (n=5 per group). B) Percentage of CPM remaining, assuming 2 minutes to be 100%. 112 Ad-LPL was able to correct the lipolytic defects associated with hypertriglyceridemia and impaired fat tolerance caused by heterozygous LPL deficiency in mice. Utilizing the RSV promoter, expression of immunoreactive human LPL enzyme was maintained for 42 to 60 days in treated animals despite undetectable human LPL catalytic activity after Day 28. Fasting determinations of plasma triglyceride levels remained lowered up to the end of the experiment at Day 60. In the absence of measurable human LPL catalytic activity, this slow return of TG to baseline levels may reflect the noncatalytic effect of LPL protein on removal of lipoproteins from the circulation. In vitro studies have revealed that LPL may function as a bridging ligand to the low-density lipoprotein receptor (LRP), which serves to enhance the uptake of CM and VLDL remnants independent of LPL lipolytic activity {163,164}. In contrast to findings in other models systems, our CMV-driven vector was able to sustain activity for greater than 60 days. The extremely high expression levels produced by this vector, likely due to the high expressing CMV promoter, allowed a significant reduction of viral load delivered to the mice. This may have reduced the immune reaction and hepatotoxicity such that infected hepatocytes were able to live and consequently express LPL for a longer period of time. Additionally, LPL expression is affected by the cytokines produced during an immune response. For example, normally the LPL plasma level is inversely related to levels of TNF, a cytokine produced by immune cells {77,167}. A lower immune response would likely translate into less TNF and thus potentially less down regulation of the LPL gene. Although the effects of TNF have primarily been shown to be transcriptional, whether the effects of TNF, or any other cytokine or hormone produced in response to Ad-infection, would affect only the endogenous LPL gene or may affect a cDNA driven by an alternative promoter has yet to be investigated. Alternatively, perhaps the LPL produced was more bioactive, stable or secreted more rapidly due to the powerful CMV promoter. Although this has not been directly tested, this latter hypothesis is not likely since both adenovectors contained the same human LPL cDNA and should produce identical proteins. Interestingly, the time course of expression was not altered significantly after either Ad-RSV or CMV-LPL delivery to transgenic mice expressing human LPL. This indicates that no immune response to human LPL was evident and the diminishment of expression 113 was not specific to a xenogenic response but due to some other process (i.e. promoter shut off, hepatocyte destruction/apoptosis). In support of this, general immunological responses were observed histologically in all groups of mice likely representing an antibody response to adenovirus (data not shown). A lack of immune response to the human LPL protein was surprising since antibodies cross reacting with human LPL, but not mouse LPL, were produced in mice after immunization with bovine LPL {129}. Additionally, the human and mouse LPL proteins only have approximately 85% identity {35} suggesting that, although limited, there should be enough diversity for an immune reaction to occur. One element to keep in mind is that the C57BL/6 strain of mice appears to be resistant to immune reactions after adenovirus administration making strain variation a potential reason for not observing an immune response {159}. Lipid analyses and FPLC characterization of lipoproteins confirmed that the major contribution to significantly decreased plasma TG levels was derived from ectopically expressed liver LPL. In the heterozygous LPL model, in contrast to the HDL lowering seen with the Ad-RSV-LacZ vector, HDL-C levels remained at baseline after Ad-RSV-LPL administration. Presumably the over expression of LPL appropriately compensated for the HDL-C lowering effect which possibly represents a transient adenovector-mediated effect on the hepatic or circulatory transport of TG-rich lipoproteins. FPLC analysis of HDL composition in heterozygous mice receiving Ad-RSV-LPL was consistent with a relative enrichment of a larger sized HDL isoform. The nature of this larger HDL species is compatible with enhanced LPL-mediated conversion to HDL2, a possibility that is in keeping with observations by Shimada et al {27} in LPL over-expressing transgenic mice, in which HDL2 cholesterol was increased 1.4-fold versus non-transgenic controls. The lack of compensation in the normal CD1 mice may represent strain related differences in the immune response to Ad administration or possibly genotypic differences in metabolism. Peak human LPL gene expression in the liver was noted at Day 7 for the RSV viruses and Day 3 for the CMV-driven viruses following parenteral adenoviral-LPL infection. At this time, with both promoter systems, plasma TG levels in fasted animals decreased dramatically with RSV decreasing approximately to one-eighth and one-half of control levels and CMV almost 100% in comparison to Ad-LacZ injected mice and uninfected litter mates, respectively. The magnitude of this finding was unexpected since the 114 increased mobilization and hepatic influx of fatty acids derived from the hydrolysis of TG-rich lipoproteins normally stimulates increased hepatic VLDL-TG assembly and secretion back into the circulation {161}. Conversely, transgenic animals over-expressing human LPL in extrahepatic tissues such as adipose and muscle hydrolyze TG locally in those tissues {24,25}. Consequently, mobilized fatty acids are either re-synthesized into TG for storage in adipose tissue, or utilized for energy consumption in muscle, ultimately reducing plasma TG levels. Whether these results represent increased uptake and storage/bile secretion of TG in the liver alone or in other tissues as well is currently unknown. Upon gross and histological examination of adipose arid muscle, these tissues appeared normal (see Table 5.3) and despite evidence of inflammation in the liver, little fatty infiltration was obvious. The correction of hypertriglyceridemia enacted by liver-expressed LPL was also consistent with the improved tolerance to intravenous fat loading in the LPL-deficient mice. The amount of fat administered intravenously to the mice (3.5g/25g mouse) was proportional to 980g consumed by a 70 kg human. Under such an extreme fat load, the clearance time of exogenous TG was greatly prolonged in LPL deficient mice relative to WT litter mates. However, over-expressed liver-derived LPL in heterozygous mice receiving Ad-LPL restored normal postprandial TG clearance, in keeping with the response elicited from uninfected WT mice. Upon oral fat load, the clearance time and curve was found to parallel that from Ad-RSV-LacZ injected heterozygous mice. However, the Ad-RSV-LPL mice manifested 10-fold lower plasma TG levels in the range of normal uninfected litter mates. After intravenous fat load, the clearance time and curve was completely normalized to that demonstrated by WT litter mates. Evidence suggests that the relationship between LPL and atherogenesis may not necessarily be mediated only by the changes that occur in plasma lipoproteins. The multi-functional nature of LPL, coupled with the variable tissue-specific expression of this enzyme, had previously confounded a clear determination of a pro- or anti- atherogenic role for LPL in atherosclerosis. For example, studies of macrophage-derived LPL expression imply a causative role for LPL in atherogenesis, especially since macrophage-derived foam cells are a major source of LPL in atherosclerotic plaques {169,170}. Studies of inbred murine strains have shown an association between high levels of LPL synthesis and secretion in macrophages with increased susceptibility to atherosclerosis {63}. A recent 115 - report from our laboratory also supports that the expression site may be more important than the plasma LPL expression level since both LPL transgenic and heterozygous LPL deficient mice had decreased atherosclerosis in comparison to normal litter mates {61}. When present in the arterial wall, LPL may promote binding and retention of LDL to the subendothelial matrix where these lipoproteins may be converted to more atherogenic forms {162}. The atherogenicity of LPL could depend on which cells are proportionally affected by increased LPL-mediated lipoprotein uptake; for example, liver cells or cells at the arterial wall. Future research with a vector capable of long term expression will allow the study of atherosclerosis post gene transfer. Our studies have demonstrated that the metabolic effects of LPL in the regulation of plasma TG metabolism are not necessarily dependent on co-ordinated secretion from its constitutive peripheral tissue sources. Our findings also suggest that ectopically-expressed LPL from the liver can participate in plasma lipoprotein metabolism in a manner similar to its action at physiological sites in peripheral tissues. Furthermore, liver-expressed LPL will remove atherogenic TG-rich lipoproteins from the circulation, theoretically compatible with ultimate protection against atherosclerosis. The demonstration of successful LPL gene expression and function resulting from adenoviral targeting of the murine host liver also supports the prospect for modifying alternate pathways of lipid metabolism by enhanced LPL expression. The feasibility of such a strategy has been previously demonstrated in mice presenting with an improved lipoprotein profile and decreased susceptibility to atherosclerosis, on a genetic background created from crossing gene-targeted LDL-R or apo E deficient mice with LPL over-expressing transgenic mice {28}. The major limitations of currently developed adenoviral vectors, such as transience of expression and related immunogenicity, have been previously well described {134,171}, but a number of recent developments appear promising for this vector system {136}. Prevention of the adenoviral-induced immunogenic host response is the main obstacle to be overcome before this technology can render prolonged transgene expression and be applied therapeutically. Nonetheless, these studies have demonstrated that the level and duration of expression of the adenoviral-delivered human LPL transgene upon IV administration is sufficient to have significant effects on lipoprotein metabolism. The evidence provided by these studies for correcting the lipolytic defects due to LPL deficiency by liver-targeted LPL 116 gene delivery is encouraging, and support the assessment of other vector systems which might allow longer term expression and be less immunogenic. 117 Chapter 5: vivo assessment of Adeno-LPL efficacy via alternative routes of delivery 118 5.1 Introduction Adult individuals normally synthesize LPL in extrahepatic differentiated tissues, primarily skeletal and heart muscle, adipose tissue, macrophages and lactating mammary gland {69}. Although LPL is expressed at low levels in several other tissues, one notable exception is the liver where LPL expression is rapidly extinguished shortly after birth, such that LPL is normally absent from adult hepatic cells and tissues {132}. Additionally, circulating LPL is efficiently bound and ultimately degraded in the adult liver {64}. Initially these issues sparked a great debate over the ideal site to target and perhaps limit LPL gene based expression to upon gene transfer. In consideration of the unknown factors associated with hepatic-derived LPL, such as the ability to produce bioactive LPL and metabolic consequences, it was predicted that liver-directed LPL gene transfer may be a more complex process than directing expression to the muscle or adipose tissue compartments. Thus, one of our goals was to determine the feasibility of adenoviral-mediated gene transfer targeting either the adipose or the muscle tissue compartment where LPL is normally expressed throughout the life of the individual. By targeting these sites, it was hoped that some level of post-translational regulation would be maintained resulting in an improved lipoprotein profile while sustaining a more normal expression pattern consequently limiting the disruption of TG partitioning and nutrition. 5.2 Adipose Our first target tissue was the epididymal fat pads in heterozygous LPL deficient mice. This site was chosen since these large fat pads could be easily identified and isolated in mice greater than 6 weeks of age by a simple surgery involving a small incision in the inguinal region of the mouse. Direct injection of the fat pad could then be observed (Figure 2.2). Due to the minimal vascularization of this tissue, we hypothesized that escape of adenovirus to other tissues would not happen to a significant extent and any plasma alterations would largely be a result of the virus injected within this tissue. 119 5.2.1 LPL Expression: plasma vs. tissue activity A pilot study was done to look at the variability in tissue specific LPL activity in the fat pads of several mice. Most atherosclerosis studies done in mice utilize the C57B1/6 strain background due to their elevated susceptibility to this disease {172,173}. Concurrent to this study, the heterozygous LPL knockout mice were being bred onto this background for atherosclerosis studies ongoing in the lab. The mice employed in this study were crossed between 4-6 times to C57BL/6. Some significant differences were observed depending on the background of the mouse (Table 5.1) after a 4 hour fast. Thus it was determined that maintaining one level of back-crossing was essential in order to minimize background variability for any LPL gene transfer studies to the adipose tissue. The reason for this variability is not understood although it is clearly related to background strain differences. The generation 4/5 background was selected due to the similar adipose and muscle background LPL activity levels in LPL heterozygous mice and siblings of this generation were bred to obtain sufficient numbers for these studies. Pilot intravenous (IV) injections revealed a highly effective dose of 5xl08 pfu of Ad-CMV-LPL per mouse (Chapter 4), thus the epididymal fat pads of male mice (n=10) were injected bi-laterally with 2.5xl08 pfu/side of either Ad-CMV-LPL or AP. Control IV injections of 5xl08 pfu/mouse were performed concurrently and confirm the activity of the virus. Unfortunately, despite a significant increase in LPL activity and decrease in plasma TG after IV treatment, no differences were obiserved in IA treated mice (Table 5.2). Although at day 7 there was a modest decrease in plasma TG from baseline levels, a similar decrease was also observed in Ad-CMV-AP treated mice suggesting that this observed decrease may be related to viral toxicity issues (control data for day 7 shown). Similarly, despite massive amounts of human specific LPL immunoreactive mass in the plasma after IV injection, there was no obvious alteration over the background observed in this ELISA system (Figure 5.1). When expression was investigated using a sensitive RT-PCR procedure that can differentiate between human and mouse LPL RNA, transcripts of both origins could be readily observed (Figure 5.2). After IA injection, no human LPL expression could be observed in the muscle compartment while a slight human background could be observed in the liver suggesting that some adenovirus likely escaped the adipose after injection and 120 Generation N6 N4 or 5 N4 or 6 Genotype +/-+/+ +/-+/+ +/-+/+ n 10 2 8 4 2 4 Adipose LPL Activity (mU/mL) 16.3 ±4 .9 46.1 ±3.1 21.8 ±5 .0 31.2 ±6 .6 10.9 ± 1.9 42.8 + 5.9 Muscle LPL Activity (mU/mL) 32.7 ±4.20 39.9 ± 10.1 26.7 ± 10.7 44.6 ± 5.80 28.1 ± 1.50 37.9+1.70 Table 5.1 Endogenous LPL activity in muscle and adipose tissue. 121 Vi ^ c i VO ,_ © IT) CN <N <—' CN .& II S3 S + 1 + 1 +1 • + 1 T> 0 0 ON ON oo i n i n B jvel ON oo CN CN CN de L< S II • + 1 i n c -+1 o + 1 VO vo + 1 r -• mm u a u .Sf c i CN '—i H (n=4 + 1 o + 1 CN + 1 + 1 c i (n=4 o *-1 © m .2- H s fl 3 w O IT; w II 0 0 C l VO oo >n C J + 1 +1 + 1 + vo VO i n CN 0 0 c i (--C l C l CN C l ON ON C l CN ON CN + 1 + 1 + 1 + 1 oo m ON C l CN ON r-- O C l C l CN > II C l i n 0 0 -—1 + 1 + 1 + 1 + CN CN m 0 0 CN ON C l C l CN o C l ii s H o C l o r->> >. >-> ca ro ro a Q Q Q Co Q T3 CU -*-» <U 00 cu CO O ro ro o cu « >> ro s cu cu o ro i—l DH fl1 cu s cu cu i T3 |> ii *5 co ,< ro < cu S3 Is Is CO CU -a •c "OO 03 ' S o S c <u co ro o cu - o a" cu a cu s cu > u i < CU <C ro co cu J 3 > co CU o ro PH hJ ro e £ ^ cu ° o § > "S - Pi fl cu •> cu CO £ o C ro | Q S i ~ ctl CO ~ ro S co o ro t -§ flv) 00 2 "to ro O c cu T3 o u CN i n s u ro ^ 3 CU T 3 •3 t^—i co O > c o cu > cu 122 * LPL AP LPL AP LPL AP Liver Muscle Adipose Figure 5.1 Comparison of plasma LPL immunoreactive mass four days after liver, muscle or adipose targeted Ad-CMV-LPL or Ad-CMV-AP delivery. A signiricanlncrease was only observed after intravenous (liver) delivery. (p<0.001) 123 ts O oo Q i c ca I 8 a &^  7 "S s s g ca CO O Xi co ON OO ON OO &o 2 " H NO c3 <U 0 u > £ -S ffl ° w c 01 O « 1 1 a X> ™ OT M s s Cj N g 3^ m OT cu rn I A PQ rs o oo vo w i-H i-H VO ) I ft 'I' T CQ in CN a y a .s .s .& PH OT OT -a o S « fi. a, ea .Er a> C3 C3 1 2 <^ I o 1 g a-OT <w< i rs T 3 ,5 ^ 8 c s ^ - S £ •£P o i ca rj <u Is I 124 reached the liver. Interestingly there seemed to be some background mouse LPL RNA expression within the liver as well. Likely this was either an artifact of the system or the endogenous mouse gene was somehow transactivated, perhaps due to elevated cytokines. However, this level of hepatic transcription was not sufficient to alter lipid or lipase levels. RT-PCR analysis of control IV mice revealed some level of human LPL transcript in both the adipose and muscle compartments in addition to very strong expression in the liver. This may suggest that part of the efficacy of IV liver targeted gene transfer may be due to disparate expression within other tissues in addition to the liver. Tissue homogenates were prepared from the fat pads after necropsy to determine if the transcription observed via RT-PCR had translated into any additional LPL protein within the adipose tissue itself. Interestingly there was a significant increase in LPL mass at both days 4 and 14 (p=0.02 and 0.04 respectively, Figure 5.3A). Although there was a trend towards increased activity this was not significant indicating that the majority of this human specific LPL mass was in an inactive form within the adipocytes (Figure 5.3B). When investigated histologically, several changes associated with inflammation and toxicity were found within the adipose compartment (Table 5.3). This included a focal loss of adipocytes, some local fibrosis and adipocyte collapse as well as the appearance of lipophages (lipid filled macrophages). Within the liver there was mild inflammation and alterations in cellular morphology consistent with a low amount of adenovirus infection. The spleen presented with well-developed follicles and germinal centers, clonal selection and apoptosis, all factors consistent with a vigorous immune response. The exact nature of this immune response was not investigated. Although successful infection and transcription was apparent within the epididymal adipose tissue compartment after Ad-CMV-LPL gene transfer, this expression did not translate into obvious alterations in plasma LPL and lipoprotein profiles. This suggests at least two possibilities. A larger number of adipocytes may require targeting (i.e. several adipose tissue compartments). This would require investigations with an alternative vector system capable of a higher level of infection with less associated toxicity/immunogenicity. Alternatively, the adipose tissue compartment is truly not a viable target for gene transfer for 125 A €8 > 2 s 2 ~Si> S ft s s N H -1 2500" 2000" 150" 100" 50" * LPL AP Day 4 LPL AP Day 14 B E N H N - 3 90" 60" 30" 1_ LPL AP Day 4 LPL AP Day 14 Figure 5.3 LPL immunoreactive mass (A, p = 0.02 and 0.04 at D4 and 14 respectively) and activity (B, not significant) in adipose tissue homogenates from Ad-CMV-LPL or Ad-CMV-AP adipose targeted mice sacrificed at Day 4 or 14. 126 4> CO o fl, -3 < -rt CU CU i§ S 6 <^ Vi Vi Vi CU O 2 "a . f l * CO ro IJ o £ 2 ro 6 CO fl > cu CU co 00 — c f O >> OH <U O O fl 9H h-1 00 < 00 O ox s. CS H o» s cw CW CO s ro. O co cu CU c u u M £ to 'co O -4-» co C  Ji ° SH .2 13 ra =3 fl C cu S o S u m > _ 2 '-fl « c o c ro o fl o 00 fl a w so co CU CU "o CO fl CU Vi O & < c cu cu "B, c/3 «3u 127 LPL since the regulatory mechanisms override the ability of exogenously derived LPL to escape. Discernment between these two possibilities may be the focus of future studies. 5.2.2 AP plasma and tissue activity In contrast to the lack of plasma LPL expression, significant levels of AP could be readily observed in the plasma of Ad-CMV-AP intra-adipose treated mice at days 3 and 7 (p<0.001, Figure 5.4A). Surprisingly, the levels attained were approximately 30% that of IV treated mice suggesting that a reasonably good infection and secretion rates were achieved (Figure 5.4C). Interestingly this plasma expression was extinguished by day 14 while expression after IV administration was readily observed and continued well beyond this time point (Figure 4.11). Also in contrast to the results with LPL, AP expression within the tissue was only elevated approximately 2 fold at day 4 (p=0.009), returning to baseline by day 14 (Figure 5.4A). This suggests that the majority of synthesized and active AP was immediately secreted out of the cell while LPL is somehow retained. The turnover of AP within the plasma compartment was not investigated and can not currently be ruled out as a contributing factor to attaining this level of plasma expression. These results suggest that the adipose tissue compartment may be a suitable target for rapidly secreted gene products such as AP or hormones but may not be a suitable target for highly post-transcriptionally regulated genes such as LPL. 5.3 Muscle The next target tissue was muscle. The tibialis anterior and quadriceps muscle in heterozygous LPL deficient mice were chosen since they could easily be identified and injected by limiting the needle depth to approximately 2 millimeters. Direct injection with trypan blue revealed the efficacy and spread of solution upon injection. Although other adenoviral-mediated gene transfer studies have identified some vector in other tissues after intramuscular (IM) injections {79}, we hypothesized that the majority of gene expression would be from the treated sites and not from other regions such as the liver which might have a significant effect. 128 Time (Days) cu CO S s ta c Q. 3 s § i * c 2 1 < 150 100 50 Ad-CMV-LPL Ad-CMV-AP Time (days) Time (Days) Figure 5.4 Plasma alkaline phosphatase (AP) activity over time after Ad-CMV-LPL or Ad-CMV-AP gene transfer to LPL +/- mice. A significant increase in AP acitivityvas observed for less than 14 days in Ad-CMV-AP treated mice after A) adipose targeting or B) muscle targeting, p<0.05. A significant elevation in plasma AP activity was sustained for longer than 14 days after C) intravenous (liver) gene targeting, p<0.05. 129 P=0.009 o w a (0 JZ a. CO o JC a. 0) c 15 10.0 7.5 H > o < 5.0 H 2.5 H 0.0 Ad-CMV-LPL Ad-CMV-AP Figure 5.5 Tissue derived alkaline phosphatase (AP) activity 4 days after intra-adipose injection of a total of 5xl08 pfu of either Ad-CMV-LPL or Ad-CMV-AP into LPL +/- mice (n=4/group). In contrast to the nearly 100 fold increase in plasma based AP expression, very little (approximately 2 fold over control Ad-CMV-LPL treated mice) was in the adipose tissue of Ad-CMV-AP treated mice. 130 5.3.1 L P L Expression: plasma vs. tissue activity In a manner similar to adipose, the 4 hour fasted tissue based LPL activity was systematically examined in the tibialis anterior muscle of LPL heterozygous knockout mice backcrossed onto various degrees of the C57BL/6 background. In contrast to the adipose findings, the level of LPL activity was reasonably consistent in all groups showing a background level of approximately 26-33 mU/ml in LPL heterozygous mice (Table 5.1). To maintain consistent breeding groups between adipose and muscle experiments, the generation N = 4 or 5 background was chosen for study. o Pilot injections with 2.5x10 pfu Ad-CMV-LacZ per side revealed an infection efficiency of approximately 5% in the tibialis anterior. This is consistent with what has generally been reported in the literature {79-81}. In initial studies, the dose of 5xl08 pfu was divided between the two tibialis anterior compartments. Follow up studies used 5x108 pfu per side dividing the dose into 4 doses and injecting the tibialis anterior once and the quadriceps in 3 different but distinct sites for a total dose of lxl0 9 pfu/mouse. This was done with the hope that targeting more muscle may have a larger influence on the plasma compartment. Disappointingly, upon either treatment with Ad-CMV-LPL, there was no alteration in plasma TG, LPL activity or immunoreactive mass levels (Table 5.2, Figure 5.1). RT-PCR was subsequently employed to determine if any human LPL RNA was present within this muscle compartment. (Figure 5.2) In contrast to only mouse LPL being present in the adipose tissue, both mouse and human LPL expression was observed in the muscle. Interestingly, although there was also some detectable human RNA expression in the liver, this level was insufficient to alter the plasma LPL activity and lipoprotein profile. Tissue homogenates on the tibialis anterior muscles were done on day 14 and showed a striking elevation of both LPL activity (2.3x, p<0.001) and mass (42x, p=0.002) over Ad-CMV-AP treated control mice (Figure 5.6). Despite this significant increase in tissue based expression, this did not translate into any plasma alterations. Histologically, the muscles in these mice had evidence of mild inflammation and localized fibrosis (Table 5.3). The livers revealed mild inflammation consistent with a small amount of adeno infection. Correspondingly, the spleen, with variable follicle size, partly reactive germinal centers and a slightly expanded red pulp, revealed that although there was 131 Figure 5.6 LPL imrnunoreactive mass (A, p = 0.002) and activity (B, p < 0.001) in muscle homogenates from muscle targeted mice sacrificed 14 days after Ad-CMV-LPL or Ad-CMV-AP gene delivery via direct muscle injection. 132 a definite immune response occurring, this was to a lesser extent than after either IA or IV adenovirus delivery. 5.3.2 A P plasma and tissue activity Similar to adipose tissue, when the plasma of Ad-CMV-AP IM treated mice was analyzed for AP activity, a very significant increase was observed at both days 3 and 7 (p<0.001 for both, Figure 5.4B). In contrast to the elevation of LPL activity and mass still present at day 14 within the tissue, AP plasma and tissue based activity was extinguished (data not shown). Whether this represents an immune response or toxicity within the tissue specific to the transgene was not determined. However, it remains obvious that the muscle is a suitable site for secreted gene products such as AP and perhaps a less suitable site for gene products regulated post-transcriptionally by this tissue. 5.4 Discussion In the adult mouse, LPL is synthesized in extrahepatic differentiated tissues {69}. LPL expression is normally absent from adult hepatic cells and tissues {132} and circulating LPL is ultimately degraded in the adult liver {64}. Therefore, it was our initial prediction that liver-directed LPL gene transfer would present as a more complex process than directing expression to the muscle or adipose tissue compartments. Initial in vitro experiments revealed the acute ability of a liver derived cell line, HepG2, to successfully synthesize and secrete bioactive LPL (Chapter 3). Complementary experiments in mice confirmed the ability of the liver to express human LPL producing the desired plasma TG lowering effect in vivo. However, due to the limited length of expression, the results of long term LPL biosynthesis in the liver could not be evaluated using our current adenovector gene transfer system. Results from the LPL transgenic mouse model expressing hepatic-derived LPL indicate that on a normal background, LPL gene delivery to the liver may be feasible and cause little long-term toxicity {42}. However, when placed on a completely deficient background in mice, expression solely from the liver presents several complexities such as increased lipid within the liver. The long term effect on metabolism, especially if challenged with disease, as well as the resultant tissue 133 composition and possible atherosclerotic complications have yet to be evaluated with liver based LPL expression in both the normal and rescued models. Pilot in vitro studies showed that a muscle derived cell line, C 2 C 1 2 , was also able to successfully synthesize and secrete bioactive LPL (Chapter 3). In contrast, infection of the mouse pre-adipocyte cell line, 3T3-L1, was limited, resulting in very little bioactive LPL being secreted into the medium. Similarly, injecting isolated mouse epididymal fat pads produced significant intracellular LPL activity and mass but limited amounts of media based LPL activity. One important consideration was that these two compartments have traditionally been difficult to target by first generation adenovectors. In contrast to the liver, which can be efficiently infected (>90%) producing subsequent high level gene expression after a simple non-invasive intravenous injection {105,134,135}, direct injection of adipose and muscle tissue result in approximately a 5% infection/gene expression rate {79-81}. In the muscle compartment, several modalities have been applied to try to increase the rate of infection including damaging the muscle either physically (forcep pinching) or chemically (toxins). This increases the number of new myocytes, which seem to express a higher concentration of the adenovirus receptors (avP3, avPs and Coxsakievirus-adenovirus receptor (CAR) {174,175}), and potentially disrupts some of the tight junctions allowing the adenovirus access to a greater number of cells {176}. Although this increases the percentage of infected myofibers, the treatment largely remains localized to the site of injection. Not many studies targeting the adipose tissue have been performed but those that have suggest an infection rate similar to muscle {88}. This may be due to similar reasons as the muscle tissue including a low density of receptors or limited access to the adipocytes upon direct injection of the tissue. In order to overcome some of this limitation, a high expressing CMV-driven adenovector was developed. It was hoped that despite low infection rates of approximately 5% in both of these organs, the level of expression from the virus able to enter the cells could attain enough expression for therapeutic alterations in plasma lipoprotein metabolism. Unfortunately, although expression within the tissues themselves was easily identifiable using both RT-PCR and LPL activity and mass analysis of tissue homogenates, this expression did not translate into a plasma alteration of LPL or lipoprotein profile. 134 Complexities associated with LPL expression in the muscle and adipose tissue compartments have been studied both in vitro and in vivo for a long time and yet are not clearly understood {87,144,177}. As described in Figure 6.6, there are many points of regulation for LPL ranging from transcription to transport to the vascular endothelium once escaping the cell of synthesis. From studies performed in this thesis it would appear that LPL production in the adipose tissue appears to be more highly regulated than that of muscle. It could be hypothesized that this is necessary due to the differential roles these cell types play in overall body homeostasis. White adipose tissue is specially designed to store fat for times of limited nutrition. When fasting conditions arise, it is important that this tissue release the TG stores in order to supply the other organs of the body with the necessary energy for survival. If LPL were not highly regulated in this tissue, this switch would not be as dramatic allowing the adipose to go from a storage depot to a major nutritional supplier. Muscle, on the other hand, takes up FFA mainly for oxidative purposes. Although some TG storing may occur, the majority of the fat within this cell type is used for energy purposes and thus plays a much more selfish role in TG metabolism {75,177}. As long as there is some level of control of LPL such that levels do not greatly exceed those of adipose under fed conditions, it perhaps is not as critical to nutritional homeostasis. Although not reported, it is possible that the cause of myopathy in the high expressing LPL muscle transgenic mouse models is due to the level of muscle derived LPL exceeding the level of adipose LPL. This would cause a higher level of local lipolysis and FFA uptake at all times manifesting in the toxicity observed. At lower muscle expression levels, the myopathy is reduced perhaps due to a more even balance between adipose and muscle tissue LPL levels. Some of the advantages of targeting the adipose tissue are that this is a natural site for high levels of LPL expression and thus this cell type is designed to accommodate significant levels of FFA flux, both in and out, without any toxicity. Additionally, if targeting subcutaneous regions, this is often a large, readily discernable target site that is easy to access. Disadvantages include the possibility that over-expression from the adipose may result in obesity {77}, or alternatively, too much expression may alter the delicate nutritional and hormonal balance provided by this tissue. This latter scenario was observed in initial IA experiments but could not be consistently replicated. Additionally, some studies 135 suggest that more than 80% of the surface LPL is internalized and degraded by the adipocyte {1}. Expression levels would have to be huge to overcome the tremendous ability of adipocytes to degrade LPL. Similarly, this region is not highly vascularized and thus may have limited access to plasma lipoproteins in comparison to a highly perfused region such as the liver. Focal expression may also be an issue since injected adenovirus tends to infect the locally treated region. Partially for this last reason, the adipose has not been a well studied target site for gene transfer. Targeting the muscle also has several potential advantages including being a large, natural site for LPL gene expression that can be targeted via non-invasive mechanisms (e.g. direct injection). It has also been suggested that over-expression from this site may reduce diet-induced obesity {57} - a significant issue in the North American society. In both confirmation and contrast, muscle specific transgenic mouse models have revealed that if sufficient expression able to affect lipoprotein metabolism and reduce obesity is attained, a possible disadvantage is the significant toxicity that appears to be dose or expression dependent {25,55,56}. Focal expression from one or two muscle categories may result in significant myopathy instead of a less atherogenic lipoprotein profile and weight loss. Additionally, this site has been traditionally difficult to target using adenovectors, likely due to the size of the virus impeding its ability to move between the tight junctions of the myocytes, as well as the lower amount of receptors present on mature myofibers, both prohibiting the infection of a large number of these fibers {176}. In the studies presented, expression within the muscle of Ad-CMV-LPL treated mice was elevated over Ad-CMV-LacZ however, this level of expression did not translate into an obvious decrease in plasma TG. This is contradictory to results published by Schlaepfer et al {111} where IM plasmid injections were able to yield a significant correlation between increased quadriceps LPL activity and decreased plasma TG. It should be noted that the plasmid injections in the muscle also resulted in expression in the opposite limb. Other studies on plasmid gene transfer have shown this phenomenon as well as a significant amount of DNA present in the liver {178}. Additionally, a significant increase in plasma LPL with decreased TG were observed with intraperitoneal plasmid injections. Although not documented, if sufficient plasmid was able to get into other tissues, expression within the liver or other tissues may 136 have provided the major contribution to the TG lowering effect observed in these groups of mice. These studies focused on gene transfer directed to the adipose and skeletal muscle categories because they are well defined, large, easily manipulated target sites that are relatively well isolated. Several other sites could potentially be targets for LPL gene transfer including the heart, kidney or perhaps the vascular endothelium. Each of these sites is associated with advantages and disadvantages. For example, transgenic studies suggest that the heart seems to be able to accommodate elevated expression and lowered plasma TG without the myopathy observed in skeletal muscle. The kidney, due to its highly vascularized nature, as well as the vascular endothelium may be able to also produce the desired alterations in plasma TG. One disadvantage common to these sites is the difficulty in targeting and confining the vector and subsequent expression. Potentially a significant amount of the virus could reach the liver or other tissues producing confounding results. This may not truly be a disadvantage since perhaps the best target may be to increase LPL gene expression in several tissue compartments. Normally LPL is expressed in many tissues throughout the body. Thus, disseminating the gene transfer vehicle to several organs, whether naturally expressing LPL (muscle, heart, adipose) or not (liver), may serve to increase the plasma LPL and alter the lipoprotein profile in a favorable manner (decreased TG at least) but help maintain overall homeostasis. This hypothesis will likely be tested in future gene transfer studies performed with a vector, such as AAV, capable of high level transfection and long term expression in all of the above mentioned, and potentially more, sites. Particularly if regulation of LPL was more clearly understood, including certain signals or promoter/enhancer elements, in addition to producing an anti-atherogenic lipoprotein profile, LPL gene transfer may also help to alter the body composition favorably (i.e. as a treatment for obesity). 137 Chapter 6: In vitro and in vivo assessment of the LPL Ser447Ter alteration by adenoviral-mediated gene transfer The work presented in this chapter will contribute to a publication currently in preparation. Relevant Publications: 1. Ashbourne Excoffon, K. J.D., et. al. Adenoviral-mediated gene transfer of the common human lipoprotein lipase variant, Ser447Ter: Potential benefits over wild type lipoprotein lipase. In preparation. ) 138 6.1 Introduction More than 70 mutations have been described within the LPL gene {8}. The majority of these mutations occur in exons 4, 5 and 6, the exons involved in forming the catalytic triad of this enzyme. Some of these mutations occur at polymorphic rates within the general population while others have been found only in a few people. Almost all of these mutations, such as those within or close to the catalytic triad of LPL, result in LPL with a complete or near-complete loss of catalytic function. However, it is now appreciated that a few mutations within the LPL gene result in partial defects in catalytic activity and a mild clinical phenotype. These include three common polymorphisms, Asn291 Ser, Asp9Asn and Ser447Ter, which occur in approximately 1 -7%, 2-4% or 17-22%, respectively, of the general population {42}. In several studies, the former two mutations have been associated with a mild decrease in HDL-C and an increase in TG. Both of these characteristics are well known risk factors for cardiovascular disease and studies on the relationship between 291 or 9 polymorphisms and CAD indicate that individuals with these alterations have an elevated risk of manifesting heart disease {15}. Conversely, although there were initial reports suggesting that the Ser447Ter (447) mutation might be responsible for type one hyperlipidemia {179}, it is now generally accepted that the 447 mutation is associated with a mild increase in HDL-C and a decrease in TG thus predisposing to a reduced risk of CAD {18,21,180,181}. Several in vitro studies have been performed in order to determine the effects of these mutations on the resulting bioactivity and stability of the LPL dimer {11,122,182,183}. Although there have been several contradictory results, studies done in our lab {122} in COS cells revealed that the 291 and 9 mutations result in a decrease in catalytic activity by altering secretion and/or stability of the LPL homodimer. Conversely, the catalytic activity and stability of the 447 mutant dimer appeared normal. The significant difference was that this mutation presented with a higher total secreted mass level (131%) than control LPL and it was suggested that this may be due to an enhanced secretion of the inactive monomeric form of LPL. The mechanism behind the significant effects, in vitro and in vivo, of deleting the last two amino acids (a serine and a glycine) of a 448 aa long protein is not well understood. 139 Enhanced secretion of LPL protein with or without activity may prove beneficial to the biological system if the protein is still able to bind both the heparan sulphate proteoglycans as well as the lipoprotein particles. Mounting in vitro and in vivo evidence suggests that this could provide a bridge between the cell and the particle bringing them in close association and allowing subsequent uptake and removal of the lipoprotein particle {58,164,184-187}. Although this event may be predicted to contribute to the development of atherosclerosis if uptake occurs in the vasculature, the TG lowering effects may be protective if it is in the peripheral tissues or the liver {61,172}. Recent studies done in our lab involving Apo E deficient mice, as well as normal C57BL/6 mice, expressing LPL at varying levels in different tissue have provided evidence that this may be true {61}. Additional studies in vivo have substantiated the validity of the LPL/lipoprotein bridging hypothesis as well. These include indirect evidence such as the results presented in chapter 3 of this thesis where a decrease in plasma TG was maintained in the absence of elevated LPL activity but in the presence of LPL immunoreactive mass {119}. Direct evidence stems from a transgenic mouse model developed to express catalytically inactive LPL solely from the muscle {58}. The transgenic mouse model was developed containing a human LPL minigene mutated in one of the residues within the catalytic triad (Aspl56Asn) under the control of the muscle creatine kinase promoter. These mice produced significant levels of post-heparin plasma dimeric LPL mass but no increase in human specific LPL activity. The major result was a reduction in VLDL-TG. Regardless of the mechanism, the association with improved lipoprotein profiles suggests that the 447 mutant may prove to be a desirable form of LPL for gene transfer to people. Additionally, these studies allow further investigation of the mechanism behind the beneficial effects of the 447 mutation. Thus, a serotype 5 adenovirus containing the 447 mutant LPL gene under the control of the CMV-promoter was developed. The major goal of this work was to compare and contrast the effects of this alteration, both in vitro and in vivo, to the similar wildtype LPL containing adenovirus. 140 6.2 Ad-CMV-447 In Vitro Preliminary in vitro studies were done in HepG2 cells and indicated a dose response relationship for LPL activity for Ad-CMV-447 of a similar magnitude to the adenovirus containing the wild type LPL (Ad-CMV-LPL, Figure 6.1). However, there was a striking difference in the amount of LPL immunoreactive protein produced subsequent to infection. Even at a MOI of 5, Ad-CMV-447 treated cells produced more LPL mass than Ad-CMV-LPL at an MOI of 50. When comparing the two viruses at an MOI of 50, which results in the infection of essentially 100% of the cells, the amount of secreted LPL protein from Ad-CMV-447 infected cells was roughly 4-fold greater that that of Ad-CMV-LPL (pO.OOl). 6.3 Ad-CMV-447 In Vivo 6.3.1 Dose Response and Time Course The animal model employed for all of these studies was the +/- LPL knock-out mouse model {33}. When the dose response relationship of the Ad-CMV-447 virus was initially evaluated in a small cohort of mice via intravenous injection, unexpected results were obtained. Neither the level of LPL activity or protein were responsive to 2 fold increases of virus (Table 6.1). The level of activity was similar to that seen in the cohort of mice given 5x10 pfu of Ad-CMV-LPL. However, the level of LPL protein seen in these mice was higher than ever previously seen in any of our studies. Initially we believed that this might represent saturation of our ELISA; however, despite activity apparently starting to decrease at day 7, the immunoreactive mass continued to increase at all three doses to a level of approximately 35-40,000 ng/ml (Table 6.2). The majority of this protein at day 7 was observed in pre-heparin plasma and the small variation in lipolytic activity indicated that it was largely in an inactive form (Table 6.2). Approximately 5000ng/ml was found exclusively in post-heparin plasma perhaps indicating the saturation level of available heparin sulfate proteoglycan binding sites. Over a time course of 70 days, LPL activity levels in mice given Ad-CMV-447 followed those of Ad-CMV-LPL closely and returned to baseline levels between 6 and 10 weeks post-injection. (Figure 6.2A) However, LPL immunoreactive protein levels were significantly elevated in mice receiving either Ad-CMV-LPL or 447 with levels in the 447 cohort maintaining profoundly elevated levels over 141 LPL Activity (mU/ml EH3 LPL Protein (ng/ml) 100 1500 50 H "0 h 1000 3 25 500 3 50 UL 50 100 Ad-CMV-LPL Ad-CMV-447 Figure 6.1 LPL activity and immunoreactive protein per 105HepG2 cells 24 hours after infection with either Ad-CMV-LPL (MOI 50) or Ad-CMV-447 (MOI 5, 50 or 100). 142 ON •St- CM + 1 +1 +1 + 1 00 ON ON CN in ON + 1 +1 + 1 + m < *-H < <-H cn in in +1 + 1 + 1 + o o O o o ON 00 in m CO m + 1 + 1 + o NO cn NO cn CO o cn •«* »—i 00 cn NO oo NO ON |2 * - 1 ON ^ " ^ . CN + 1 +1 +1 r~ "2" cn r-» NO CN + 1 in U h-1 a a a PH a OH ft OH PH 00 00 O O O ' O u X X X X CN >n 1 m < 143 Vi 'a, 3 E .5 a L . a (V JS • a. 03 a Vi o CM U 5 iz, OO §1 •a ^ iz, 00 H a .fl OH W '-fl fl i—i e OH W h-) CL, W I—) '-fl fl < £ OH W Vi e o ON + 1 NO cn r -+1 NO CN cn + 1 CN CN cn r -+ 1 cn cn NO CN CN + 1 CN ON 00 ON CN CN + 1 Os ON cn oo cn in CN . +1 Os 00 in ON r--i •o + 1 oo in cn oo cn + 1 00 CN "1 +1 ON +1 ON ir i CN r--cn NO in + 1 cn o IT) NO i r i O + 1 CN o cn oo o , + 1 oo NO CN o , +1 oo NO cn in +1 00 NO © CN + 1 in NO NO in cn + 1 NO iri -3-cn ON Os + 1 00 NO NO CN •+ + 1 cn oo cn 144 B .E 1000000 -i Time (days) Figure 6.2 Time course of LPL expression in mice heterozygous for LPL deficiency receiving 5x10s pfu of Ad-CMV-LPL, Ad-CMV-447 or Ad-CMV-AP. Expression was measured by A) LPL activity and B) LPL immunoreactive protein. Although LPL activity was significantly elevated over Ad-CMV-AP control mice for only 14 days, plasma protein levels were significantly elevated over the duration of the experiment with extremely high levels of protein present in the plasma of receiving Ad-CMV-447. 145 the wt LPL group even 10 weeks post-injection (Figure 6.2B). Unfortunately the duration of this expression beyond this time point was not evaluated. TG levels were significantly reduced at all three doses of Ad-CMV-447 in a manner similar to 5xl08 pfu Ad-CMV-LPL (Figure 6.3 A). Similar to previous observations, this plasma TG decrease was significant for the duration of the experiment despite the normalization of post-heparin LPL activity levels within these mice and correspond to the time course of LPL protein expression. Both total and HDL cholesterol were significantly reduced until day 14 (Figure 6.3B,C). Although not directly tested, considering the profound decrease in both TG and cholesterol, this likely represents a significant decrease in all categories of lipoprotein particles with some equilibrium being attained after day 7. 6.3.2 Ad-CMV-447 at a Reduced Dose The lack of dose-response and profound TG decreases observed at the levels tested suggested that the system was likely saturated potentially by infection of virtually all of the hepatocytes and/or efficient expression due to the CMV promoter. We hypothesized that a 10-fold reduction in dose might allow us to further investigate the dose response relationship of the LPL activity, protein and lipoprotein changes. Thus, LPL +/- mice (n=5/group) were given either 5x10 or 5x10 pfu of either Ad-CMV-447 or LPL. The goal of this study was to evaluate whether differences in lipoprotein levels might be observed at a much lower dose. It should be noted that in these experiments a dose 5xl07 pfu is equivalent to approximately 5xl09 particles - a dose rarely effective in vivo in experiments employing adenoviral vectors. As previously observed, at a dose of 5x10 pfu of either Ad-CMV-LPL or 447 per mouse, there was a significant 2.7 fold increase in plasma LPL activity accompanied by a significant increase in LPL protein levels (Figure 6.4) at day 5 post gene transfer. Corresponding TG, HDL-C and Total-C levels also dropped significantly (data similar to Figure 6.3 A,B,C). At this dose, the only significant difference between the adenovector containing the wild type LPL cDNA versus the 447 version was the plasma LPL protein level, which was elevated in the 447 group. Interestingly, at a dose of 5x10 pfu, although the significant protein level difference was maintained (Figure 6.4), we were able to observe other differences between Ad-CMV-LPL and 447. LPL activity in the mice receiving 146 -25 H 1 — , , , , , ^ , 0 10 20 30 40 50 60 70 80 Time (days) 50 -i 0 -I 1 1 r 1 1 1 1 1 0 10 20 30 40 50 60 70 80 Time (days) Figure 6.3 Time course of alterations in lipoprotein profile observed in mice heterozygous for LPL deficiency receiving 5xl08 pfu of Ad-CMV-LPL, Ad-CMV-447 or Ad-CMV-AP. A significant decrease in plasma triglycerides (A) was observed for the entire duration of the experiment (p<0.04) while significant decreases in both HDL (B) and total cholesterol (C) were only observed for 7 days post gene transfer (p<0.04). 147 CH3 LPL Activity (mU/ml EHH3 LPL Protein (ng/ml) 2000 -i 4000 E E, • > o < 1000 H I - 3000 2000 1000 •o r I * I i II =• CD i o <" (D Baseline 0.0 Ad-CMV-447 Ad-CMV-WtLPL Figure 6.4. LPL activity and protein level alterations after the administration of 5xl07 pfu of Ad-CMV-wtLPL or Ad-CMV-447 via tail vein to mice heterozygous for LPL deficiency. Day 5 post-heparin LPL immunoreactive mass levels were elevated in both groups with much higher levels being observed in the Ad-CMV-447 group (p<0.03). Interestingly, LPL activity was significantly elevated in both groups at day 5 although to a more significant extent in mice receiving Ad-CMV-LPL versus Ad-CMV-447 (p<0.0001 vs. p=0.02 respectively as compared to baseline). 148 Ad-CMV-LPL was approximately 90% of the full dose, a 2.5 fold increase over background while the Ad-CMV-447 cohort observed only approximately 50% of the activity of the higher dose. Although significantly elevated in both groups over baseline or control mice, post-heparin LPL protein levels were still most profoundly elevated in the Ad-CMV-447 group (p<0.03). The most provocative differences observed only at this lower dose were in TG and cholesterol measures. At 3 days post-gene transfer, TG levels were significantly decreased in both Ad-CMV-LPL and 447 groups (Table 6.3, Figure 6.5A), indicating the efficacy of the transferred LPL in both groups of mice. However, there was a significant increase in both HDL-C and total-C only in the Ad-CMV-447 group (Table 6.3, Figure 6.5B,C). The magnitude of these alterations, when compared to baseline levels, indicate that the majority of the increase in total-C is within the HDL-C fraction. At this same dose in the Ad-CMV-LPL cohort, there was a slight decrease in both total-C and HDL-C with only the decrease in total-C achieving significance (p=0.04). This is the first time we have observed an increased HDL-C content after adenovirus-mediated gene transfer of the human LPL gene in mice. A similar significant elevation of the HDL-C fraction was observed at day 7, resolving by day 14 (Figure 6.5B). 6.3.3 Human Associations Several investigations into populations of people with and without this Ser447Ter mutation have been performed revealing a strong association with increased plasma HDL, decreased TG and a decreased risk of CAD. Although many mechanisms have been proposed, supporting evidence has been limited. When the above data revealed a profound elevation in LPL protein levels in a murine model, we initiated further investigation of this phenomenon with Dr. John Kastelein, a collaborator in Amsterdam, Holland. Pre- and post-heparin plasma samples from 565 subjects with proven coronary artery disease were provided by Dr. Kastelein and were analyzed within our lab for LPL activity and protein levels. These subjects were patients involved in the placebo-controlled Regression Growth Evaluation Statin Study (REGRESS) which was originally designed to evaluate the effect of P-blocking agents in a large group of male patients with angiographically proven CAD {19,188}. As shown in Table 6.4, when compared to non-carriers, carriers of the 447 149 OT 8 s PH h-1 00 c OT CO CO OH ii h4 PH o +1 o CO N O +1 CO CO o + 1 C N O N O N O N O N + 1 O N C N O N ^ + 1 00 r-» o +1 o N O OT .zL £-\ -s 'fi r t i e OT « P 5 BJ 1 0 OT '-a >, a o H m in +1 + 1 + 1 N O o m in « .2; CU CO OT -o 'SH hJ 1 m NO NO h-1 + 1 +1 + 1 Q NO CN NO CO CO CU CU O H o C N + 1 O N CO +1 CO +1 C N ^ <-> — 1 > cu "3 OT CO OQ hJ PH hJ u I < PH 150 ioo n 0 H 1 1 1 1 0 10 -20 30 40 Time (days) Figure 6.5 Time course of lipid changes after a low dose (5xl07 pfu/mouse, n=5) of either Ad-CMV-LPL or Ad-CMV-447. A significant decrease from baseline was observed in both groups of mice for the duration of the time course (p<0.01). An increase above background and Ad-CMV-LPL in both HDL-C and total-C was observed at days 3 and 7 post-gene transfer of Ad-CMV-447 (p<0.05). 151 LPL Activity (mU/ml) LPL Immunoreactive Protein (ng/ml) Mean N Median Range N Non-Carriers 106.3 ±42.9 494 201 15-2692 458 447 Carriers 122.7 ±56.8 116 1777 94-6598 107 P-Value <0.001 <0.001 Table 6.4 LPL activity and immunoreactive protein levels in human subjects from the REGRESS study (Regression Growth Evaluation Satin Study) comparing Ser447Ter carriers to non-carriers. 152 mutation presented with a significant increase in LPL activity levels and a highly significant eight-fold increase in LPL protein concentration. Previously this population of patients carrying the 447 polymorphism had been shown to have an anti-atherogenic lipoprotein profile with elevated HDL-C and decreased TG levels {19}. This data confirms the relevance of gene transfer studies in animal models prior to and allowing insight into the manifestations of human genetics and biology. 6.4 Discussion Although advances in the methodology of gene delivery are being well studied, the genetic material to be transferred through gene therapy techniques is also still under vigorous examination. Both animal and human data is presented within this chapter suggesting the utility of the Ser447Ter variant of the human LPL gene not only for patients with complete and partial LPL deficiency but also for persons at risk of cardiovascular disease (CAD). The 447 variant is common in the human population and results in a two amino acid truncation of the LPL protein. In humans it is associated with increased plasma HDL, decreased triglycerides (TG) and a decreased risk of CAD {15-21}. We now show that it is also associated with a significantly increased amount of plasma LPL protein. Complementary to this, gene transfer studies utilizing an adenovector containing this variant, in a murine model for LPL deficiency, reveal a similar phenotype to above. These beneficial effects are not observed with the wild type version of the human LPL gene and thus are unique to this variant suggesting preferential utilization of this variant in future gene based therapeutic strategies. LPL has multiple functions in the metabolism of lipids and lipoproteins. Not only does it catalyze the hydrolysis of triglycerides, it functions as a ligand allowing the bridging of intravascular lipoproteins mediating the clearance of these triglyceride-(TG) rich lipoproteins (including atherogenic species) via receptor-mediated uptake by the liver {58,164,184-187}. Through both mechanisms, plasma TG concentrations can be lowered. Additionally, it is generally accepted that LPL also plays a role in HDL metabolism. Current models suggest that LPL-mediated hydrolysis of the TG component of TG-rich lipoproteins allows the transfer of surface remnants (including phospholipids, free cholesterol and apoproteins) from these large, TG-rich lipoprotein particles to nascent HDL 153 particles {27,44}. Although this has been invoked to explain the positive correlation between LPL and HDL levels {2,189}, the exact mechanism is not well understood. However, from the data presented within this chapter, it appears that this effect may be at least partly mediated by an LPL protein effect rather than relying solely on catalysis. The increase in LPL protein levels may somehow alter either the processing of TG-rich lipoproteins affecting HDL generation rates, or alter the catabolism of HDL in vivo. In the present studies a profound and significant elevation in LPL protein level was discovered after gene transfer of an adenoviral vector containing a Ser447Ter alteration in the human LPL gene. A similar phenotype was then observed in a defined human population of 447 carriers versus non-carriers. When lipid levels were re-evaluated in mice after receiving a lower dose of virus producing physiologically relevant levels of LPL protein and only slightly, albeit significantly, elevated LPL activity levels, a significant increase in HDL-C was observed in addition to decreased TG. This alteration in HDL was likely not observed at the higher doses due to a potential saturating effect of the LPL present in the plasma which significantly reduced the majority of the plasma lipoproteins. This effect may have stemmed from either efficient hepatocyte infection, such that the majority of the liver was infected and expressing LPL, and/or the efficiency of the CMV promoter in the viral construct. This murine data corresponds with previous, and now present, descriptions of human 447 carriers. The nature of this relationship is not currently understood and will surely be elucidated in future studies. It is interesting to note that despite the differences between mouse and human lipoprotein metabolism, the mechanism relating LPL protein concentration and HDL is likely common to both mouse and human. It will be important to determine whether this effect is due to LPL protein concentration or whether the deletion of the last 2 amino acids from this protein is somehow altering LPL function or degradation and mediating the observed effects. One important caveat when interpreting the murine studies is that the majority of the adenovirus-mediated expression was targeted to the liver, a non-physiological site for LPL expression. Thus, we are not creating a true representation of the in vivo situation. For example, it has been shown that an inactive mutant version of LPL expressed in the muscle is capable of lowering TG {58} but part of the mechanism for this observation may require heterodimerization resulting in one side capable of lipolysis while the inactive side is 154 capable of binding. This transgenic line of mice was not able to rescue the homozygous lethality of LPL (activity) deficiency. When expressed in the liver, there is no endogenous LPL for heterodimerization and similarly, the effect and/or mechanism of the 447-LPL protein may be altered or enhanced. The effect of truncating the last 2 amino acids of LPL appears to be pleiotropic. Potentially there are many places where a small change such as this can have a large impact on the regulation of LPL (Figure 6.6). LPL is highly regulated both transcriptionally and post-transcriptionally and this change potentially could affect LPL synthesis and processing right from RNA stability to LPL protein secretion to degradation by the liver. Certainly many different steps have been implicated in the literature (e.g. dimerization and intracellular trafficking {122}, altered interaction with lipid substrates {179}, altered binding to endothelial surfaces {16,21}, altered interactions with LRP/LDLR/lipid binding/apoCII {190,191}, and altered remodeling of nascent lipoproteins in and near the space of Disse in the liver affecting lipoprotein concentration and uptake {190}). LPL mRNA levels are regulated by several transcription factors {1,69}, some of which in turn are regulated by metabolic products or hormones (e.g. FFA, insulin, or TNF). Once transcribed, the RNA is edited and transported out of the nucleus. Although not likely, a missense mutation (TCA -> TGA) near the 3' terminal end of the mRNA hypothetically could alter this process or rate of LPL mRNA turnover if the cellular mediators of this process were to have altered binding. Quantitative RT-PCR or mRNA turnover studies may be useful in determining if the transcriptional efficiency or RNA degradation rates are implicated. LPL is then synthesized as an inactive proenzyme in the endoplasmic recticulum and is transported to the Golgi apparatus where multiple step glycosylation leads to dimerization and activation. Once synthesized, there are three potential routes for the LPL protein to take: degradation, constitutive secretion and regulated secretion. Hypothetically, any of these steps could be affected by the truncation of the last 2 amino acids. There is some in vitro evidence to suggest that the secretory pathway may be affected. When compared to wild type LPL, protein produced containing mutations that abolish LPL activity often get shuttled into the degradation pathway {192}, while one study from our lab suggested that 155 Site of Action DNA 1 RNA 1 Protein • Transcription rate • Editing • Stability/turnover • Synthesis (inactive monomer) • glycosylation/processing • Dimerization (active dimer) • Sorting/ turnover/ degradation • Transport • Secretion - Constitutive - Regulated • Binding to cell surface • Transport to vascular endothelium • Binding to HSPG • Binding to lipoproteins/function • Dissociation • Uptake/ degradation by liver Nucleus Nucleus Cytoplasm Endoplasmic Recticulum Golgi Vesicles/lysosomes Interstitial space Vasculature Figure 6.6 Points of regulation for the LPL gene and protein. 156 the 447 mutant had increased secretion rates likely in the monomelic, inactive form {122}. This latter study might be interpreted to suggest that there is a problem in both dimerization as well as the regulation of secretion. Both of these possibilities are consistent with our in vivo murine data considering the profound levels of inactive pre-heparin as well as active post-heparin 447 protein. Future protein trafficking studies may be able to evaluate the location of cellular LPL and whether the transport of the newly synthesized 447-LPL is altered from the normal process. Once secreted out of the cell, LPL can either bind to the HSPG on the cell surface, potentially allowing re-uptake by that cell followed by degradation, or be translocated to the vascular endothelium where it attaches to the HSPG on the luminal surface {153}. LPL then comes in contact with its TG substrate in TG-rich lipoproteins. At this point, LPL can bind to and be taken up by the circulating TG-rich lipoproteins or hypothetically become inactive and fall off the endothelium. Either way, it is generally accepted that the site of vascular LPL degradation is the liver. The profound level of the 447-LPL in pre-heparin plasma after gene transfer suggests that there may be a problem with receptor recognition and degradation. In support, the c-terminus of LPL has been implicated in receptor recognition and binding {35,182}. This may be reliant on conformation or sequence recognition. Mutational analysis of the LPL c-terminus revealed the importance of the last 11 or 12 amino acids, with activity dropping rapidly upon truncation of amino acid 436 (Val436Ter) {182}. Truncations of more than this in humans result in a totally inactive allele {8,193}. Plasma 447-LPL protein turnover and receptor binding studies will likely provide further insight into the potential mechanisms behind the in vivo effects of this alteration. In summary, the Ser447Ter mutation significantly affects the biology of the LPL gene and protein. Studies of this altered protein may lead to insight into many important fields of research including protein trafficking and lipoprotein, in particular HDL, metabolism. In addition to previous epidemiological studies, the data presented strongly suggests the use of this "super-LPL" in all gene-based therapeutic strategies for LPL deficiency and related hyperlipidemias. 157 Chapter 7: Adenovirus-mediated correction of feline LPL deficiency The majority of the data in this chapter comes from one publication, as described below, and represents a collaboration between myself and Dr. Guoqing Liu, a post-doctoral fellow in our lab. I participated in or performed all experiments presented in this chapter with the exception of any FPLC analysis, which was work performed by Dr. Liu and is provided herein as additional proof towards the lipoprotein profile alterations in vivo after intravenous Ad-mediated LPL gene transfer. Technical support for the majority of the LPL activity, mass and lipid measurements was provided by Li Miao and Fudan Miao. Relevant Publications: 1. Liu, G., Ashbourne Excoffon K.J.D., Wilson, J.E., McManus, B.M., Rogers, Q.R., Miao, L., Kastelein J.J.P., Lewis, M.E.S. and Hayden, M.R. 2000. Phenotypic correction of feline lipoprotein lipase deficiency by adenoviral gene transfer. Human Gene Therapy 11, 21-32. 158 7.1 Introduction The lethality of homozygous LPL deficiency in mice makes the investigation of studies in gene transfer in larger animal models capable of sustaining complete deficiency essential for the development of therapeutics for this genetic disease. Currently there are two animal models described for complete deficiency including a colony of cats maintained here at UBC and a colony of mink maintained in Sweden. Due to the availability of the colony of cats here at UBC, this has been the chosen model for the studies on adenovirus-mediated gene transfer described herein. The additional benefits of using the feline model are that it is well characterized and is quite similar in phenotype to the human deficiency, including chylomicronemia, subcutanous xanthomatas, lipemia retinalis, and a decreased tolerance for fat loads. Furthermore, cats are a nice model to work with from the perspective of temperament. If handled correctly from an early age, management in studies such as these is uncomplicated requiring a minimum of sedation or physical manipulation. Please see Chapter 1.3.1 for a more complete discussion on models of LPL deficiency. Although there are benefits to using larger model organisms such as the cat, two major limitations include the number of animals available and the large amount of viral preparation required. Thus the numbers of animals used in these studies were small. However the results, especially utilizing the high expressing CMV driven vectors are convincing of the therapeutic value of LPL gene replacement for the correction of chylomicronemia within this animal model. 7.2 RSV promoter With the success of our initial adenovirus vector containing the RSV promoter both in vitro and in murine systems, the simple intravenous injection liver targeted approach was applied to the feline model for LPL deficiency. Initial experiments were performed in heterozygous deficient cats due to availability of these animals as well as an exaggerated phenotype of elevated TG (50-150mg/dl). A normal cat has very low TG levels (10-20mg/dl) making the determination of a change in TG levels more challenging than observing a reduction in TG levels of a +/- cat to "normal" levels. 159 7.2.1 Ad-LacZ Pilot study in normal cats An initial dose response study was performed by Drs. S. Lewis and G. Liu in normal cats with the Ad-RSV-LacZ virus. Three cats were injected intravenously with Ad-RSV-LacZ at a dose of lxlO 1 0, 2xl010, or 4xl01 0 pfu/kg (Table 7.1). Clinical chemistry was done on each cat weekly. Ultrasound guided liver biopsies were taken at the indicated time points and measured for LacZ activity. At all three doses there was elevated LacZ activity at day 7 however at the lowest dose expression was extinguished by day 14. Activity was not followed further than day 7 in cat 2 due to complications not related to this study although expression at day 7 was almost 2-fold higher than the expression at lxl0 1 0. Expression in the liver of the cat receiving 4x1010 pfu/kg was also elevated more than 2-fold over the cat receiving 2x1010 pfu/kg and expression was sustained for at least 4 weeks. At all 3 doses the hemoglobin count decreased in correspondence to a change in the Coombs test, which indicates red blood cell hemolysis, from negative to positive. These results indicated that there is toxicity associated with all 3 doses of virus however the highest and most prolonged expression was observed with the highest dose. 7.2.2 Ad-RSV-LPL Pilot study in LPL +/- cats Considering the apparent increase in level and length of expression associated with increasing the dose of virus coupled with similar levels of toxicity, two LPL +/- cats were treated with Ad-RSV-LPL or LacZ at a dose of 5.5x1010 pfu/kg. Unfortunately there was significant toxicity associated with this dose and the animal receiving LacZ was sacrificed at day 3 and the cat receiving LPL was sacrificed at day 17. However, this dose of Ad-RSV-LPL resulted in a profound over-expression of LPL increasing the plasma activity levels from a baseline measure of 116.5 mU/ml to 2570.9 mU/ml, An increase of over 22 times the original level. In situ hybridization done on the liver of these animals revealed a significant number of hepatocytes expressing LPL only in the animal receiving Ad-RSV-LPL (Figure 7.1). Due to the toxicity of the high dose of adenovirus, the next pilot study employed a significantly reduced dose of virus. Four cats were injected with 1.15xl010 pfu/kg of either Ad-RSV-LPL or LacZ (n=2 for each group). 160 Cat Dose (pfu) Time (Days) Hgb Coombs LacZ specific Activity (mU/mg) 1 l x l O 1 0 0 104 - ve 0.00 7 79 + ve 6.01 14 70 + ve 0.00 2 2x l0 1 0 0 72 - ve 0.00 7 78 + ve 11.00 3 3x l0 1 0 0 94 - ve 0.00 2 ND ND 16.28 7 84 + ve 29.52 28 52 + ve 14.30 Table 7.1 B-galactosidase activity in liver biopsies taken from cats injected with varying doses of Ad-RSV-LacZ 161 162 Expression levels At a dose of 1.15x1010, the plasma LPL activity in the cats receiving Ad-RSV-LacZ remained approximately at the baseline level (Figure 7.2). In the cats receiving Ad-RSV-LPL, there was approximately a 2-fold increase in plasma LPL activity. However, LPL activity appeared depressed in comparison to baseline in these same cats at both the day 21 and day 28 time points. This may have been due to the amount of toxicity or the antibody response observed in these animals (Table 7.2, Figure 7.3) versus those receiving Ad-RSV-LacZ. Lipid changes The cats in the group receiving Ad-RSV-LPL were chosen to have higher baseline TG levels as well as a slightly more impaired rate of TG clearance. This was so that in our pilot studies, with small numbers of cats, changes in lipid metabolism could be more readily observed. It is not fully understood why the cats, despite the LPL genotype, within the colony vary except that these cats are not significantly inbred and natural variations exist within the population of cats. The baseline lipid levels of the two groups were certainly within the range normally seen for the heterozygous genotype. Administration of Ad-RSV-LPL revealed a significant decrease in TG levels at day 7 when LPL activity was approximately doubled in these same cats (Figure 7.4). TG levels then remained low for the duration of the study despite the return of plasma LPL activity to baseline levels or lower. Over the same time period, the TG levels in the cats receiving Ad-RSV-LacZ did not vary. Intravenous fat tolerance was evaluated in these cats prior to virus administration and again at 10 days post adenoviral delivery (Figure 7.5). At baseline, the cats in the LPL group had a slower fat load clearance rate than those in the LacZ group. At 10 days post gene transfer, the clearance rate in the cats receiving LPL was greater than those receiving LacZ whereas the clearance rate of the latter group was not altered despite viral administration. 163 Time (days) Figure 7.2 Time course of plasma L P L activity in L P L +/- cats treated with an intravenous injection of 1.15x1010 pfu/kg of A d - R S V - L P L or A d - R S V - L a c Z . A significant increase in L P L activity was only observed at day 7. 164 Virus Dose Time Hgb Coombs sGPT (pfu/kg) (Days) (norm 80-150) (norm 5-65) Ad-RSV-LacZ 1.15 x lO 1 0 0 129/113 -/- 32/29 3 129/120 -/- 88/32 7 104/104 -/- 52/41 14 99/113 -/- 43/125 28 118/111 ND 26/44 Ad-RSV-LPL 1.15 x 1010 0 137/150 +/- 39/27 3 129/154 +/- 87/36 7 123/127 '+/- 82/86 14 93/122 +/- 151/113 28 118/122 ND 36/36 Table 7.2 Markers of vector toxicity in LPL +/- cats receiving either Ad-RSV-LPL (n=2) or Ad-RSV-LacZ (n=2) at a dose of 1.15 x 1010 pfu/kg. (Hgb = hemoglobin, sGPT = serum transaminases) 165 L7T7J Ad-RSV-LPL mmW Ad-RSV-LacZ 100 -i 0 7 14 21 28 60 Time (day) Figure 7.3 Development of inhibitory anti-LPL antibodies in LPL +/- cats after administration of Ad-RSV-LPL but not Ad-RSV-LacZ. Post-injection serum samples on the days indicated were incubated in vitro with a human LPL tissue culture source for 2 hr on ice. At a 1:8 dilution, more than 50% inhibition of human LPL activity was observed 2 weeks after Ad-RSV-LPL injection, and nearly complete inhibition was observed by 4 weeks. 166 Time (days) Figure 7.4 Plasma triglyceride (TG) alterations in LPL +/- cats treated with 1.15xl010 pfu/kg Ad-RSV-LPL or Ad-RSV-LacZ. The groups (n=2) of cats were selected to have initially elevated levels of plasma TG. A significant decrease from baseline was observed in cats treated with Ad-RSV-LPL from baseline for the duration of the study. Due to the lower initial levels of TG in control Ad-RSV-LacZ cats, no difference was observed between groups over the study. 167 Time (hours) Figure 7.5 Intravenous fat load tolerance at baseline (pre-gene transfer) and after treatment with Ad-RSV-LPL or Ad-RSV-LacZ. In order to observe a more exaggerated response, the group (n=2) of cats receiving Ad-RSV-LPL were selected to have a more prolonged fat load clearance rate. After injection of Ad-RSV-LPL, a significantly increased rate of clearance was observed while Ad-RSV-LacZ treatment had no effect on fat load clearance. 168 Toxicity The antibody response to human LPL was evaluated in cats receiving either Ad-RSV-LPL or LacZ (Figure 7.3). Plasma from the indicated time points was pre-incubated with a known amount of purified bovine LPL protein and subsequently assessed for the amount of remaining LPL activity. Although there was a slight depression in LPL activity remaining after plasma from the cats receiving Ad-RSV-LacZ, this was likely a non-specific reaction. However, in cats receiving Ad-RSV-LPL, there was a severe inhibition of LPL activity indicating a strong neutralizing immune response to the expressed human LPL protein. Clinical chemistry measures done on these cats indicated that the toxicity of the injected virus was minimal (Table 7.2). Only one cat receiving Ad-RSV-LPL, which was coombs positive prior to gene transfer, remained coombs positive. In all cats there was a mild elevation of serum transaminases indicating a mild liver toxicity that was resolved in all cats by day 28. 7.2.3 Pilot study in LPL -/- cats A dose of 1.15xl010 pfu/kg of either Ad-RSV-LPL or Ad-RSV-LacZ was administered to two LPL -/- cats. For unknown reasons, there was no obvious change in LPL expression in either cat. Neither cat was sacrificed at this time due to the limited numbers of homozygous LPL deficient cats. Several untested hypotheses were generated such as the high levels of circulating TG-rich lipoproteins adsorbing to or inactivating the virus after injection, or alternatively inactivation of the virus prior to administration. At this point controls, including the injection of a small cohort of mice or infection of cells in culture, were implemented for all future studies. Further experiments done in LPL -/- cats were limited to the higher expressing Ad-CMV-LPL virus and controlled with a secretable alkaline phosphatase marker gene (Ad-CMV-AP). 7.3 CMV promoter Due to the inability of the RSV-driven adenovectors to express efficiently in LPL -/-cats, the efficacy of the CMV-driven adenovectors were evaluated. This virus was very 169 effective at a greatly reduced dose which in turn results in minimized amounts of toxicity. Although Ad-CMV-LacZ was available, alkaline phosphatase (Ad-CMV-AP) was used as a control virus. The major advantage was that gene expression could be determined without sacrificing the animals to assess them histologically. 7.3.1 Pilot study in LPL +/+ cats Expression levels and Lipid changes Ad-CMV-LPL and Ad-CMV-AP stock was tested iv in mice prior to initiating studies in our feline kindred and confirmed a profound elevation of LPL activity, immunoreactive mass, specific activity and decrease in plasma TG (see Chapter 4). Subsequently, 2 wild-type cats were chosen for initial study. Two days after infusion of Ad-CMV-LPL at 8X109 pfu/kg to one cat, the level of PHP LPL activity increased more than 3-fold compared to baseline at Day 0 or the Ad-CMV-AP treated cat at Day 2 (Table 7.3). LPL activity increased further to >2400 mU/ml 7 days after infusion. This correlated directly with increased levels of LPL immunoreactive protein mass reaching 1,369 ng/ml at Day 7, representing a >10 fold increase from controls. Although plasma TG was initially low in both cats after an overnight fast, a further decrease of TG in the cat receiving Ad-CMV-LPL persisted over the 7 days post gene transfer. It is worth noting that LPL protein mass and activity decreased substantially in the cat receiving Ad-CMV-AP, thereby highlighting the importance of such a viral vector control in this type of study. Nonetheless, this pilot experiment confirmed the efficacy of the adenoviral vector system in a feline model and the pilot was further extended to include cats with complete LPL deficiency. 7.3.2 LPL-/-cats LPL Expression levels Two LPL deficient cats were initially recruited for a pilot study to compare Ad-CMV-LPL and Ad-CMV-AP. Significant elevation in LPL activity, immunoreactive mass and normalized plasma TGs were seen up to Day 5 (Table 7.4). The sacrifice of the 2 LPL deficient cats 5 days after gene transfer revealed a high level of LPL activity in the liver homogenate (Table 7.4), and human LPL message in the liver and spleen by 170 Plasma LPL Day Virus Injected Activity (mU/ml) Mass (ng/ml) Plasma TG's (mg/dl) 0 Ad-CMV-AP 448.5 227.7 24.6 Ad-CMV-LPL 282.2 112.2 20.1 2 Ad-CMV-AP 310.6 141.3 43.7 Ad-CMV-LPL 996.8 414.5 6.5 7 Ad-CMV-AP 117.3 114.6 48.3 Ad-CMV-LPL 2477.4 1369.9 4.6 Table 7.3 Efficacy of Ad-mediated LPL gene transfer in normal cats. Viral vectors at 8 X 109 PFU/kg were infused via the cephalic vein into two wild-type cats. Both the pre- and post-heparin blood were drawn prior to and on day 2 and day 7 after infusion for analysis of TGs in pre-heparin plasma and LPL activity and protein mass in post-heparin plasma. 171 Plasma LPL Virus Injected Activity Mass Plasma T G LPL activity (mU/ml) (ng/ml) level In liver (mg/dl) (mU/g wet weight) 0 Ad-CMV-AP 0 14.1 1138.9 Ad-CMV-LPL 3.9 11.3 830.3 2 Ad-CMV-AP 5 12 1064.7 Ad-CMV-LPL 160.1 202 10.4 7 Ad-CMV-AP Ad-CMV-LPL 1011.5 18.2 11.3 421.2 Table 7.4 Pilot study of Ad-mediated LPL gene transfer to cats with complete LPL deficiency. Viral vectors at 8 x 109 PFU/kg were infused via the cephalic vein into two homozygous LPL-deficient cats. Both the pre- and post-heparin blood were drawn prior to and on day 2 after infusion for analysis of TGs in pre-heparin plasma and LPL activity and protein mass in post-heparin plasma. The two cats were sacrificed on day 5 after infusion of viral vectors for analysis of LPL activity in the liver and tissue distribution of exogenous human LPL. No heparin was given on day 5 to avoid confounding the measurement of LPL activity in the liver, performed concurrently. Therefore, no PHP was sampled on day 5 and only TGs were measured. 172 RT-PCR analysis (Figure 7.6) in the cat given Ad-CMV-LPL. There was a very faint band corresponding to human LPL in the lung. However, no human LPL transcript was detected in other tissues examined. The control cat did not show a human LPL specific band in any tissue. Four other cats with complete LPL deficiency underwent a similar study. Cats 1 and 2 received Ad-CMV-AP, while cats 3 and 4 received Ad-CMV-LPL. In these latter 2 cats PHP LPL activity and protein mass readily increased 2 days post Ad-CMV-LPL gene transfer to 200 and 500 mU/ml and 230 and 290 ng/ml from zero endogenous LPL levels respectively (Figure 7.7A,B). The increase of LPL in both cats reached normal wild type levels. In cat 3, both LPL activity and protein mass were increased at day 7 post-treatment while those of cat 4 began to decline from Day 3. At 2 weeks post gene transfer, LPL activity and protein in both cats 3 and 4 fell to undetectable levels, similar to the 2 control cats, and remained undetectable until the end of the experiment (4 weeks). Lipid changes Upon the expression of active LPL, a profound reduction in plasma TG was evident. Due to the variable clearance of TG-rich lipoproteins in the LPL -/- cats, great variations of TG in the fasted state have been observed. Prior to gene transfer, lipid levels in these cats were ascertained on at least three separate occasions. TG levels were then converted to a percentage value of original TG level. Figure 7.8 shows that gene transfer of Ad-CMV-LPL resulted in more than a 90% drop in plasma TG in each of the cats 2 days after administration of this vector. The reduction in plasma TG persisted up to Day 7 in cat 4, increasing thereafter. In cat 3 plasma TG increased from Day 14 onwards. Four weeks after gene transfer, the plasma TG levels were elevated to baseline levels noted prior to Ad-CMV-LPL gene transfer. Although natural fluctuation of plasma TG was observed in 2 control cats no such effect was significant. Therefore, the nearly 10-fold reduction in plasma TG is solely dependent upon exogenous LPL expression in these animals. 173 j ' I I Ik in co CN C O OO CD i n CO CN _ to T - OO fJQ CN CN T— I OO CO m CO T— CN CO r--CD LO co CM CD < CU >> a -a c c "3 U o o cn C3 J3 < E .* "0 t3 S> £ -a <u > +5 CO P* OT +-> T CU cu CU CO . ? J cl —1 o o 3 -a o 0, 5 —i P* o cn c cci E "3 « CJ cu 9 c Xi "3 * I 3 | S g l i s CO ° U3 _g j ^ * J jB £ c« -3 o 00 c _3 -o C CCS C CU y OT •-" > 00 T3 3 S 3 OT j-OT Ed U «* rt rt OT • op 3 o - c d cs rt g OT S CU - C 1- rt > ° rt "o rt ±3 O S cu X3 1 C E 3 -3 3 o ' O T OT CU s-c X a NO <u E 3 00 OT 3 o a > c c "3 CU <u g, CU . 3 v1". *3 w ca rt cd cu OT O rt CJ CJ 3 3> T3 > O rt a, 55 > o U "3 CL, « I H a: CD M i OT ccj i U PU I f-T3 , c r ca £ 3 J3 OT CU _00 OT rt £ « J H ^ PL* xi OT "73 rt CU OT CU — OT i rt CU 3 CU ••5 T3 OT OC CU rt 3 | 2 «5 CM i-3 a-U > u -a < Ofi c cS cu 59 CU OT OT •-. CU — u PH cu i a . S o t! O CQ ri o cu OT S-CU > cu s-c2 *o "S — rt rt 3 3 O O o CJ 174 C N C O ^ -*—> +-> CO CO o o (0 o CO Q_ Q_ O i l +3 H P " D " O o o < < CQ |ui/6u 'u|e;ojd n d - ! c ro a> c <D I re (0 >» CO Q <4H O * 2 CO . £ ) CU H-» o t-H OH cu > OH PQ roue; n o CU i C ! ro o g co CM a ro o t= cu '3 cu -a JH OH <+H O ro 6 co "EH c •c ro DH u <2 O T 3 CU CO ro OH CO ^ <H -5 a o OH o ro 1 co co CO s cu M-H o cu ro cu eg co O -d ro t o ro o C H cu *5 i ^ FJ I-H . S o ro cu B cu 6 cu |iu/nui 'JtyAiov Idi P-H se in se in se in u ro cu *> u , i> o < cu bH 7.71 tors. CO ure vec iple 00 -a PtH < ro CO 175 CC CO O O O O _ . Q_ Q_ Q_ Q_ ^ ^ ^ ^ oooo "D"D"D"D < < < < TT)I c CO Q) c CD o I (0 (0 >» ro Q o o 0 A B Q J.O O/ 0 ' | 9 A 9 | Q I l-l u OT CN OT CS O OT CO OT o tu T3 OH hJ O -=3 g — ' <y a ts f—i JU s OT o <u l-l cu 2 ^ - - H 1-1 g > rt o co cj s ^ OT NO o ON A o o 4> •s * 1 •o 3; OT 3 S § 2 <5 <3 +1 DO OT H MI'S CU a, • OH OT 00 OT *^  «> | » 2 * * .SP 8 "2 P* oo .5 176 Analysis of plasma lipoproteins by FPLC revealed a huge TG peak at the fractions equivalent to CM/VLDL in homozygous cats before Ad vector delivery, despite removing the plasma fat cake prior to FPLC analysis (Figure 7.9A,B). Subsequently, this peak of TG-rich lipoproteins at the VLDL/CM position (Figure 7.9B) essentially disappeared 7 days post Ad-CMV-LPL transfer (Figure 7.9D), yet persisted in those control cats receiving Ad-CMV-AP (Figure 7.9C). The prominent peak of HDL-C did not change substantially for either cat cohort, before and after adenoviral vector administration, which was consistent with the quantitation of plasma HDL-C by the precipitation method described (data not shown). These data confirm that the exogenous active hLPL appropriately influences the metabolism of feline TG-rich lipoproteins in vivo. Adenoviral-delivered hLPL not only reduced endogenous TG-rich lipoproteins in feline LPL deficiency, but also resulted in accelerated clearance of an exogenous TG emulsion introduced via intravenous infusion of Intralipid substrate when assessed at 9 days post gene delivery. Figure 7.10 demonstrates the slower clearance rate seen in cats with complete LPL deficiency prior to Ad-CMV-LPL administration, versus wild-type cats. Administration of control adenoviral vector did not significantly alter the clearance rate of exogenous TG. The rate of TG hydrolysis increased primarily during the initial phase of clearance in those cats receiving Ad-CMV-LPL with approximately a 60% reduction in plasma TG, at a rate closely paralleling that seen in wild type cats. Slower clearance to 80% of original TG levels occurred by 6 hours whereas approximately 50% of residual TG remained in those LPL -/- controls receiving Ad-CMV-AP. Toxicity Sera were analyzed at different times for presence of neutralizing antibodies to the recombinant adenovirus. Prior to adenoviral exposure none of the cats had neutralizing antibodies. After administration of either adenoviral vector (Ad-CMV-LPL or Ad-CMV-AP), neutralizing antibodies were detected in all 4 cats, 7 days later at a low titer (1:64 to 1:256). A further increase in antibody titer (up to 1:1024) was observed at 2 weeks and persisted at this level until the end of the study. 177 178 0=1 P o/0'91 euiseid | | | 179 To further investigate the short-term exogenous LPL expression in these LPL deficient cats, we analyzed inhibiting antibodies against hLPL. Cat sera collected at different times were incubated with a culture medium containing high levels of human LPL obtained from Ad-CMV-LPL infected 293 cells. At a 1:8 dilution, more than 70% inhibition of hLPL activity was observed 2 weeks after administration of Ad-CMV-LPL (Figure 7.11 A). At 4 weeks, human LPL activity was almost completely inhibited. The plasma from control Ad-CMV^AP cats had virtually no effect on hLPL activity. Therefore, strong inhibitory anti-LPL antibodies were detected as early as two weeks after intravenous infusion of adenovirus containing the human LPL gene. The feline Ab is found to completely inhibit LPL activity in serum samples collected from human, mouse, cat and bovine species (Figure 7.1 IB). Throughout the course of these studies, animals were monitored for potential toxicity due to adenoviral gene transfer. There was no loss of body weight or alteration in behavior. Biochemical analyses revealed a moderate elevation of the liver enzyme, alanine aminotransferase (ALT) at Day 7 after Ad-CMV-LPL infusion in cats 3 and 4 (Table 7.5), persisting in cat 4 to 14 days (normal ALT for cats is 65 IU/1). Blood levels of insulin and glucose remained within the normal range for all 4 cats before and after gene transfer. Ketone body formation in the plasma was at very low levels during the study and there was no defined trend or difference between cats receiving Ad-CMV-LPL or Ad-CMV-AP over the 28 days post gene transfer. Morphological analysis of hematoxylin and eosin (H/E) stained preparations of Ad-CMV-LPL and Ad-CMV-AP treated livers from LPL -/- cats, necropsied at Day 5 post-infection, revealed comparable degrees of mixed inflammatory cell (plasma T, B and macrophage cell) infiltration limited to the liver (data not shown). Multifocal areas of mixed inflammatory cell infiltration, interspersed with occasional sites of hepatocellular apoptosis and necrosis were also minimally apparent at Day 5. Examination of H/E stained spleen, lung, heart, kidney, skeletal and adipose tissues harvested from either Ad-CMV-LPL or Ad-CMV-AP infected LPL-deficient cats was unremarkable. ORO staining was performed to localize neutral lipid deposits amongst the tissues sampled in both Ad-CMV-LPL and Ad-CMV-AP LPL-deficient cohorts at Day 5 (Figure 7.12). Interestingly, a moderate and diffuse increase in ORO positivity was noted in the 180 CO ^ •4—' H—' ro ro o o T " CN _ _ _i _ i CD CO Q_ Q_ ° °!_] _] +3 +3 " D " D O O < < D 1 I B I CO CM > N CO Q CN CO Q • • w w . CO Q 2a ro o o o i n (%0(Hseo A B Q ) pauieuiaj - p v ~ld~l ueiunij io % (0 c CO +•> o c o O ) £ CO (0 >» ro O u 5-P H~* o u 3 co to •5 J 2 Ps CO 1 g£ •3 - f l cd - r CO l> ea O 0 is "3 J « x> i g DH CU i—i .s C i CU "—' ^H co CU .2 £ 1 1 ca -a i—) .S DH CO 1 03 5@ XJ CU b * |§ :5 <u I I O to cu P £ 3 C L co B fl 0) o > •& CU o 00 CU CN X! CU I <D co JO O CO i > fl cu co c3 ^ I—I CU 3 ! ca >» i 5 <+H C 0 cu -H OT s * .2 © .15 co 1 £ fl fl —1 o 3 I • f l « +-> CU cu "3 O OH B £ * o fl ° o oo cd T - H •'fl CN . CO P> O (U DH 3 J fl° DH cu o CO cu fl X) c ca fl o < -a cu — C O . f l DH i—l u 1 xi < kH CU <£ ca 181 Cat Day 1 2 3 4 0 55 35 38 45 2 49 30 42 40 7 30 35 114 104 14 50 32 46 72 21 41 25 36 37 28 51 49 13 41 Table 7.5 Serum ALT activity in LPL- deficient cats after administration of Ad-vectors. Cats infused with Ad-CMV-LPL and Ad-CMV-AP were analyzed for serum levels of alanine aminotransferase (ALT) activity at the indicated times before and after infusion of recombinant adenovirus. Cats 1 and 2 received Ad-CMV-AP whereas cat 3 and 4 received Ad-CMV-LPL. Activity expressed as international units per liter. 182 Figure 7.12 Oil red O staining for neutral lipid in liver (A and D), spleen (B and E), and lung (C and F) in Ad-CMV-AP (A-C) injected control versus Ad-CMV-LPL (D-F)-injected LPL-deficient cats 5 days after administration of Ad vectors. Histopathological evidence of increased Oil red O-positive lipid influx was seen in Ad-CMV-LPL-injected cats relative to controls. 183 liver of the Ad-CMV-LPL recipient cat (Figure 7.12D), finely divided within hepatocytes and Kupffer cells, yet was seen only at trace levels in the Ad-CMV-AP recipient (Figure 7.12A). Spleen and lung tissues from the Ad-CMV-LPL recipient (Figure 7.12E,F respectively) also demonstrated significantly greater ORO positivity relative to Ad-CMV-AP controls (Figure 7.12B,C). These findings clearly indicate that expression of hLPL in the tissues of LPL -/- cats results in increased influx of lipids (fatty acids and/or re-synthesized TG) in these tissues and that the liver is not the sole tissue targeted after peripheral venous administration of Ad-CMV-LPL in the cat. It was not assessed whether this ORO positivity was in part due to the initial lipid burden in the cats which may have later been resolved once the lipid levels dropped to those of normal cats. 7.4 Discussion The availability of a fully characterized feline model of LPL deficiency closely resembling the human disorder has enabled us to address whether liver directed expression of LPL was effective in ameliorating the biochemical and clinical phenotype of complete LPL deficiency. Here we have shown, for the first time, the phenotypic outcome of exogenous gene delivery in a naturally occurring large animal model of complete LPL deficiency. The affected phenotype in these cats is amenable to short-term correction by systemic, in vivo gene transfer of a simple, first-generation adenoviral vector after intravenous administration. Furthermore, the liver-directed adenoviral-mediated secretion of exogenous human LPL is also in direct contact with the plasma compartment, successful in resolving both endogenous and exogenously induced hypertriglyceridemia and is not associated with significant clinical, biochemical or histopathological evidence of somatic toxicity. Although adenovirus-mediated gene transfer by an intravenous route predominantly targets the liver, it is also expressed in disparate organs including spleen and lung. Thus if appropriate regulatory regions are included in future studies, coordinated secretion from multiple tissues sites may be feasible. LPL is naturally synthesized in extrahepatic differentiated tissues, primarily skeletal and heart muscle, adipose tissue, macrophages and lactating mammary gland as well as kidney, lung, ovaries and brain {106}. It remains unclear which LPL-producing cell type is the major contributor of LPL found at the endothelial surface and whether this reflects a 184 continuous flux of lipase molecules from various sites of synthesis in peripheral tissues {194,195}. Our data indicated exogenous LPL expression to be most significant in, yet not limited to, the feline liver. This finding is not unexpected in view of the natural hepatic affinity of adenoviral vectors, when administered intravenously {134}. Newborn mice with complete LPL deficiency caused by targeted disruption of the carboxy-terminal LPL gene die within the first 48 hours of life, presumably due to congestion and microinfarcts of the peripheral circulation by excess TG-rich lipoproteins {33,34}. Merkel et al {42} recently reported that liver LPL expression exclusively on a background of gene-targeted murine LPL deficiency rescued the neonatal lethal phenotype of the LPL-deficient mice. Surprisingly these mice survived with a persistent elevation of triglyceride levels. These same adult mice appeared to shunt circulating TGs to the liver leading to a futile cycle of LPL-mediated uptake and secretion of VLDL by the liver. Despite our success in normal and LPL heterozygous knock-out mice, these findings of unresolved hypertriglyceridemia and aberrant VLDL turnover in complete LPL deficiency by hepatic expression of LPL brought into question the feasibility of successfully treating LPL deficiency via an intravenous delivered adenoviral gene transfer approach. However, our data are in contrast to the results reported by Merkel et al {42} and could be accounted for by: (1) cat versus mouse species variation and their respective clinical presentations leading to viability versus lethality for complete LPL deficiency; (2) potential temporal and spatial differences in hepatic expression of LPL - whereas the mouse model was induced to express liver LPL at the embryonic stage of development via microinjection into a fertilized egg, the expression of LPL in the adult cat was mediated by systemic injection of an exogenous adenoviral vector; (3) the different distribution of LPL expression; the mice exclusively express LPL in the liver while LPL expression was observed in multiple tissues in the Ad-CMV-LPL injected cats. Taken together, systemic delivery of a newer generation adenoviral vector expressing LPL may in fact be feasible to replace the normal plasma function of LPL in lipoprotein metabolism in the LPL deficient state. Studies of transgenic mice which over express LPL {26,27,52} specifically in heart {44} or skeletal muscle tissues {25} have demonstrated a significantly improved lipid profile including reduction in plasma triglycerides and mild elevation of HDL-C. In contrast, cholesterol determinations in the Ad-CMV-LPL treated cats did not show any significant 185 change in HDL-C despite substantial plasma clearance of VLDL/CM lipoproteins. While the expressed hLPL resulted in appropriate hydrolysis of the feline TG-rich lipoproteins, HDL metabolism may be dependent upon multiple factors influenced by species-specific variation in lipoprotein metabolism, the potential disruption of normal liver lipoprotein metabolism due to the ectopic expression and release of LPL from this site or, a possible alteration in the usual systemic equilibrium of LPL expression and function between contributing tissues. We have previously shown the correction of hypertriglyceridemia and impaired fat tolerance in heterozygous (gene-targeted) LPL-deficient mice after adenoviral-mediated expression of human lipoprotein lipase (Chapter 4). Whereas the response to fat load was greatly impaired in untreated heterozygous mice relative to wt littermates, a single injection of a comparable dose of Ad-CMV-LPL completely restored normal postprandial hydrolysis of serum TGs, paralleling the clearance observed in wt mice. Ad-CMV-LPL mediated correction of fat tolerance in feline LPL deficiency showed that the rate of TG hydrolysis increased primarily during the initial phase of clearance. The presence of 20% of the plasma TG level relative to wt controls remaining after the initial clearance phase may reflect the metabolic limitation of hepatic gene therapy in complete LPL deficiency in efficiently resolving the rapid massive influx of lipid to the liver, particularly in competition with increased VLDL secretion and decreased VLDL turnover. It is well known that first generation adenoviral vectors elicit a significant host immune response against the viral vector, targeting the elimination of viral infected cells and preventing successful repeat vector delivery. In our novel feline model, we also consistently detected strong inhibitory anti-LPL antibodies as early as two weeks after intravenous infusion of adenovirus containing the human LPL gene. This finding, coupled with the presence of adenoviral-specific neutralizing antibody as early as 7 days post gene transfer suggests that the extinguishing of longer term expression of Ad-delivered human LPL in these LPL deficient cats resulted from a combined host immune response against both the vector and transgene. Therefore, overcoming the adenoviral-induced immunogenic host response alone is not the single exclusive hurdle to overcome before this technology can render prolonged transgene expression and be therapeutically applied. Moreover, the significant anti-LPL host response and cross-reactivity to other species including the cat, suggests that LPL protein is either completely absent or that the immune tolerance to self-186 protein is disrupted after adenoviral-mediated LPL delivery in this affected cat kindred. It therefore seems unlikely that this response could be ameliorated by virtue of using a feline versus human LPL cDNA transgene. Further cross-reactivity studies using a panel of different antibodies are now in progress in attempt to reassess whether LPL deficiency in these cats is due to complete absence of LPL protein or a catalytically defective protein, as previously described {35}. These findings also suggests that the duration of benefit from somatic cell LPL gene therapy may be extended further if applied in the presence of a catalytic defect resulting in circulating levels of LPL protein, in contrast to a null allele. In summary, adenoviral-mediated gene therapy to mediate the delivery and expression of a corrective LPL gene confirmed its anticipated effect in resolving the hypertriglyceridemia and dyslipoproteinemia of feline LPL deficiency. This work provides a further key advance supporting the development of LPL gene therapy as a viable therapeutic option in the treatment of complete LPL deficiency. The challenges of increasing duration of expression and reducing vector-mediated immunogenicity may be attainable using newer-generation, more extensively modified viral vector systems for LPL gene delivery. 187 Chapter 8: Conclusions and Recommendations for Further Work 188 8.1 Summary of results Lipoprotein lipase plays a critical role in the regulation of total body lipoprotein and energy metabolism. Although it has been studied for many years, with the first case of complete deficiency being documented in 1939 {196}, there are many regulatory features and roles of this enzyme that remain to be fully characterized and integrated. One of the major goals of our lab is to develop and test a comprehensive series of vectors and delivery systems for LPL gene transfer and expression in somatic cells in vitro and in vivo. Through the application of various vector systems, it is hoped that some of the complex cellular and biological features of LPL can be understood more fully and provide evidence towards the efficacy of LPL as a candidate gene for a gene based therapeutic strategy in patients with complete and perhaps partial LPL deficiency. This thesis has focused on one particular, well characterized, system of gene transfer: namely the adenovirus. This vector system has many advantages over other systems including an inherent efficiency and ability to infect non-cycling cells resulting in very high levels of gene expression. Although expression is transient, likely due to the immune reaction as well as the episomal nature of the adenovirus, studies herein clearly reveal the validity of utilizing adenovirus as a tool for both study of the gene and pre-clinical evaluation of the feasibility of LPL gene therapy. 8.1.1 In vitro analysis Initial studies in vitro provide an important and fundamental basis for predicting successful in vivo gene transfer and expression of LPL via any gene delivery approach. The studies presented revealed the efficacy of a first generation E1-/E3- recombinant adenoviral vector, bearing either an RSV- or a CMV-driven human LPL cDNA expression cassette to achieve efficient adenoviral-mediated LPL gene delivery, ectopic expression and biological function in various cell lines. In particular HepG2 cells, a relevant and accepted in vitro model for studies of human hepatocellular lipoprotein metabolism, were employed and revealed successful nonconstitutive, ectopic hepatic-derived LPL expression. Together, these studies provide critical preclinical evidence of the potential efficacy of adenovirus as a tool for LPL gene transfer. 189 8.1.2 Hepatic target in vivo The adult liver is an unnatural site for LPL gene expression. Thus initial hypothetical discussions focused on the tissue targeting issue for any gene transfer method applied to LPL deficiency. Due to the hepatotropic nature of adenoviruses, success was debated. Initially, in vitro studies in a human hepatocellular model provided a basis for predicting successful in vivo liver-directed adenoviral-mediated LPL gene transfer. However, definitive proof came from ectopic liver expression in normal and heterozygous LPL deficient mice as well as human LPL transgenic mice after a simple intravenous injection of an adenovirus vector containing the human LPL cDNA. In summary, these findings suggest that liver-targeted, adenovirus-mediated LPL gene transfer offers an effective means for transient correction of altered lipoprotein metabolism and impaired fat tolerance due to LPL deficiency. Additionally, LPL gene transfer may have some benefits in a normal system altering the lipoprotein profile to a more anti-atherogenic status. 8.1.3 Adipocyte and myocyte targets in vivo Despite the success of hepatic-directed adenoviral-mediated LPL gene transfer, it was of great interest to investigate gene transfer to other natural sites of LPL expression. Two major sites of endogenous LPL expression are the muscle and adipose tissues. Due to the natural LPL expression, the size and accessibility of these sites, as well as several studies in the gene transfer of other genes within the literature for guidance in study design, they were chosen for assessing gene transfer. Unfortunately, the level of expression attained within the tissue compartment was insufficient to translate into an alteration of plasma LPL activity, mass or TG levels. Whether an effect might have been observed upon a high fat diet, as was seen in one published study after plasmid-mediated LPL gene transfer to muscle {111}, is a source of future studies. Considering the results of plasmid based gene transfer, optimism remains for targeting the muscle compartment and perhaps the adipose as well, with a vector system able to transfect this tissue at a higher rate than that typically observed by adenoviral vectors. 190 8.1.4 LPL mutant analysis Some of the most remarkable results obtained using the adenovirus gene transfer system for LPL were achieved by incorporating the cDNA of a common polymorphism in the LPL gene which truncates the last 2 amino acids off the 448 aa protein (Ser447Ter). Although the lipid altering effect of this polymorphism has been well described in several human populations, the mechanism has been a matter of speculation. Two very interesting facts emerged immediately within the in vivo pilot study. One was the lack of a dose response and the other was the massive amount of LPL immunoreactive protein secreted into the plasma of these mice. The significance of the lack of dose response is not fully understood; however, when LPL plasma activity and protein levels were evaluated in a Dutch population of 447 carriers, a significant elevation in both parameters was also observed. This suggests that part of the mechanism of this polymorphism may be due to this increased protein level in addition to a slightly higher plasma activity. Strikingly, when the viral dose was reduced, a significant increase in HDL was observed only in the cohort of mice receiving Ad-CMV-447, mimicking the human condition. Although the plasma LPL activity was significantly increased, the magnitude of the increase was less that observed at the higher doses and the LPL protein level was in a range similar to that observed in the human population. The elevation of HDL was not observed with the wild type version of the human LPL gene and appears to be unique to this variant. This data suggests that this particular polymorphism may be of significance to many future studies as well as gene therapy strategies such that less gene transfer would be required to see a significant anti-atherogenic lipoprotein profile change. Preferential utilization of the Ser447Ter variant may translate into the lowering of the prevalence of cardiovascular disease in patients with dyslipidemia stemming from several sources including but not limited to LPL deficiency. 8.1.5 Feline correction of complete LPL deficiency Assessment of adenoviral-mediated LPL gene transfer in vitro followed by normal and heterozygous LPL knockout mice, prior to studies in our limited homozygous deficient cat colony, served to identify an efficient route of delivery as well as to help define the 191 immune response. Interestingly, although a vigorous immune response to the adenoviral vector was observed in both mice and cats, an immune response to the exogenously delivered human LPL was only observed in cats. This is surprising considering that the protein identity from cat to human is 94.5% while the mouse is approximately 85% {35}. Utilizing the feline cDNA in future studies is currently under investigation and is hoped to minimize the immune response allowing for prolonged gene expression. However, at this time there is no satisfactory and simple method for minimizing the observed immune response to either the vector or the transgene. With the advent of new vector systems it is hoped that the immunogenicity of the vector itself may be minimized. In the case of a completely deficient patient with no protein or gene expression at all, the immune reaction to the delivered gene product itself is a large concern with no current solution. 8.2 Further investigations Although work to date has provided a firm foundation indicating the feasibility of LPL gene transfer, many questions remain unanswered and several investigations will have to be performed before moving to gene therapy in humans. Our adenoviral vectors, in their current form, yield short-term expression and toxicity indicating that their true clinical potential is limited to preclinical studies such as these. However, the value of utilizing these adenovectors as tools for further understanding the role of LPL in lipid metabolism should not be underestimated. This system will likely continue to be a choice in vitro and in vivo gene expression system and despite the limitations, results provided herein clearly indicate the potential clinical benefits to LPL gene transfer. One important question that will continue to be a major focus in future studies is choice of expression site for LPL. The work presented within this thesis suggests that the best site for LPL gene transfer based expression may be the liver. However, due to the potential caveats and unknowns associated with long term hepatic-derived LPL, especially in a completely deficient patient, further work must be done to look at higher expression levels from the muscle or coordinated expression from several tissues, likely including the liver. Additionally, gene regulation will become an integral part of future studies. With the exception of one study that employed the mouse muscle creatine kinase promoter, ubiquitously expressing viral based promoters have dominated the gene transfer vectors for 192 LPL. Although suitable for proof-of-principle studies such as those presented herein, investigators must seriously examine the benefits of regulated and tissue specific expression promoters, as well as other expression elements (introns, enhancers, etc.), before moving to clinical studies. For example, if the muscle and adipose tissue compartments could be targeted with regulatory elements that are appropriately responsive to the nutritional status of the individual, a more natural LPL gene expression profile would be achieved. In this situation, the likelihood of toxicity (i.e. myopathy, obesity or cachexia) would be reduced. This thesis provides strong support towards employing the common LPL Ser447Ter polymorphism into any future gene transfer vehicles and studies. Our current Ad-vectors will continue to provide us with an efficient method to further investigate the mechanisms behind the beneficial effects of this alteration over wild type LPL both in vitro and in vivo. The pleiotropic effects of truncating the last two c-terminal amino acids of LPL on both LPL protein regulation as well as lipoprotein metabolism will provide the basis for many future investigations ranging from intracellular protein trafficking to studies in atherosclerosis. Current studies are focussed on the ability of the 447-LPL protein to participate in lipoprotein metabolism after intramuscular gene transfer as well as the alterations in plasma LPL protein turnover. Another major area for future investigation is vectorology. LPL has a few important advantages for this type of gene transfer study. It has a small cDNA allowing insertion into virtually all currently described vector systems along with various promoter elements, and it is secreted into the plasma resulting in ability to continuously evaluate gene expression in vivo over significant periods of time. Studies are currently underway in several laboratories evaluating the efficacy of adeno-associated virus (AAV) for LPL gene transfer. To date this vector appears promising providing long term expression and less immunogenicity {114-116}. Future studies will likely also involve new and improved vector systems as they are developed. Toxicity is one key facet that will have to be evaluated with each new gene transfer system, as well as target site(s). In summary, this thesis has investigated several critical cellular and biological questions surrounding the issue of gene transfer for LPL. The data herein provide a firm foundation for advancing to future studies involving other vectors such as the adeno-associated virus with a better, longer term clinical prospective than adenovirus. As the gene 193 transfer field itself moves forward, the reality of treating the phenotype associated with both complete and partial LPL deficiency through the utilization of a gene-based therapeutic strategy will hopefully become a reality. 194 References 1. Braun J. E. and Severson D. L. 1992. Regulation of the synthesis, processing and translocation of lipoprotein lipase. Biochem.J. 287[Pt. 2], 337-347. 2. Brunzell J. D. 1995. Familial lipoprotein lipase deficiency and other causes of the chylomicronemia syndrome. The Metabolic Basis of Inherited Disease. [59], 1913-1932. New York, McGraw-Hill Inc. 3. Patsch J. R., Gotto A. M., Jr., Olivecrona T., and Eisenberg S. 1978. Formation of high density lipoprotein2-like particles during lipolysis of very low density lipoproteins in vitro. Proceedings of the National Academy of Sciences, USA 75 [9], 4519-4523. 4. Schaefer E. J., Jenkins L. J., and Brewer H. B., Jr. 1978. Human chylomicron apolipoprotein metabolism. Biochem.Biophys.Res.Commun 80[2], 405-412. 5. Wion K. L., Kirchgessner T. G , Lusis A. J., Schotz M. C , and Lawn R. M. 1987. Human lipoprotein lipase complementary DNA sequence. Science 235, 1638-1641. 6. Deeb S. and Peng R. 1989. Structure of the human lipoprotein lipase gene. Biochemistry 28, 4131-4135. 7. Sparkes R. S., Zollman S., Klisak I., Kirchgessner T. G., Komaromy M. C , Mohandas T., Schotz M. C , and Lusis A. J. 1987. Human genes involved in lipolysis of plasma lipoproteins: mapping of loci for lipoprotein lipase to 8p22 and hepatic lipase to 15q21. Genomics 1, 138-144. 8. Murthy V., Julien P., and Gagne C. 1996. Molecular pathobiology of the human lipoprotein lipase gene. Pharmacol.Ther. 70[2], 101-135. 9. Gagne C , Brun L. D., Julien P., Moorjani S., and Lupien P. J. 1989. Primary lipoprotein lipase activity deficiency: clinical investigation of a French Canadian population. CMAJ 140[4], 405-411. 10. Kern P. A. 1991. Lipoprotein lipase and hepatic lipase. Current Opinion in Lipidology 2, 162-169. 11. Reymer P. W. A., Gagne E., Groenemeyer B. E., Zhang H , Forsythe I., Jansen H , Seidall J. C , Kromhout D., Lie K. E., Kastelein J., and Hayden M. R. 1995. A lipoprotein lipase mutation (Asn291Ser) is associated with reduced HDL cholesterol levels in premature atherosclerosis. Nature Genetics 10, 28-34. 12. Ma Y., Liu M.-S., Ginzinger D. G , Frolich J., Brunzell J. D., and Hayden M. R. 1993. Gene-environment interaction in the conversion of a mild-to-severe phenotype in a patient homozygous for a Serl72~>Cys mutation in the lipoprotein lipase gene. Journal of Clinical Investigation 91 [5], 1953-1958. 195 13. Miesenbock G., Holzl B., Foger B., Brandstatter E., Paulweber B., Sandhofer F., and Patsch J. R. 1993. Heterozygous lipoprotein lipase deficiency due to a missense mutation as the cause of impaired triglyceride tolerance with multiple lipoprotein abnormalities. Journal of Clinical Investigation 91, 448-455. 14. Pimstone S. N., Clee S. M., Gagne S. E., Miao L., Zhang H., Stein E. A., and Hayden M. R. 1996. A frequently occurring mutation in the lipoprotein lipase gene (Asn291Ser) results in altered postprandial chylomicron triglyceride and retinyl palmitate response in normolipidemic carriers. Journal of Lipid Research 37, 1675-1684. 15. Hokanson J. E. 1997. Lipoprotein lipase gene variants and risk of coronary disease: a quantitative analysis of population-based studies. International Journal of Clinical and Laboratory Research 27, 24-34. 16. Zhang Q., Cavanna J., Winkelman B. R., Shine B., Gross W., Marz W., and Galton D. J. 1995. Common genetic variants of lipoprotein lipase that relate to lipid transport in patients with premature coronary artery disease. Clinical Genetics 48, 293-298. 17. Mattu R. K., Needham E. W. A., Morgan R., Rees A., Hackshaw A. K., Stocks J., Elwood P. C , and Galton D. J. 1994. DNA variants at the LPL gene locus associated with angiographically defined severity of atherosclerosis and serum lipoprotein levels in a Welsh population. Arteriosclerosis and Thrombosis 14, 1090-1097. 18. Kuivenhoven J. A., Groenemeyer B. E., Boer J. M. A., Reymer P. W. A., Berghuis R., Bruin T., Jansen H., Seidell J. C , and Kastelein J. J. P. 1997. Ser447Stop mutation in lipoprotein lipase is associated with elevated HDL cholesterol levels in normolipidemic males. Arteriosclerosis, Thrombosis and Vascular Biology 17, 595-599. 19. Groenemeijer B. E., Hallman M. D., Reymer P. W. A., Gagne E., Kuivenhoven J. A., Bruin T., Jansen H., Lie K. I., Bruschke A. V. G., Boerwinkle E., Hayden M. R., and Kastelein J. J. P. 1997. Genetic variant showing a positive interaction with 13-blocking agents with a beneficial influence on lipoprotein lipase activity, HDL cholesterol, and triglyceride levels in coronary artery disease patients. The Ser447-Stop substitution in the lipoprotein lipase gene. Circulation 95, 2628-2635. 20. Fisher R. M., Humphries S. E., and Talmud P. J. 1997. Common variation in the lipoprotein lipase gene: Effects on plasma lipids and risk of atherosclerosis. Atherosclerosis 135, 145-159. 21. Gagne S. E., Larson M. G., Pimstone S. N., Shaefer E. J., Kastelein J. J. P., Wilson P. W. F., Ordovas J. M., and Hayden M. R. 1999. A common truncation variant of lipoprotein lipase (S447X) confers protection against coronary heart disease: the Framingham Offspring Study. Clinical Genetics 55,450-454. 196 22. Benlian P., de Gennes J. L., Foubert L., Zhang H., Gagne S. E., and Hayden M. 1996. Premature atherosclerosis in patients with familial chylomicronemia caused by mutations in the lipoprotein lipase gene. The New England Journal of Medicine 335, 848-854. 23. Bijvoet S., Gagne S. E., Moorjani S., Henderson H. E., Fruchart J.-C, Dallongeville J., Alaupovic P., Prins M., Kastelein J. J. P., and Hayden M. R. 1996. Alterations in plasma lipoproteins and apolipoproteins before the age of 40 in heterozygotes for lipoprotein lipase deficiency. Journal of Lipid Research 37, 640-650. 24. Gnudi L., Jensen D. R, Tozzo E., Eckel R. H , and Kahn B. B. 1996. Adipose-specific overexpression of GLUT-4 in transgenic mice alters lipoprotein lipase activity. Am.J.Physiol. 270[4 Pt 2], 785-792. 25. Levak-Frank S., Radner H , Walsh A., Stollberger R., Knipping G., Hoefler G., Sattler W., Weinstock P. H , Breslow J. L., and Zechner R. 1995. Muscle-specific overexpression of lipoprotein lipase causes a severe myopathy characterized by proliferation of mitochondria and peroxisomes in transgenic mice. Journal of Clinical Investigation 96[2], 976-986. 26. Liu M.-S., LeBoeuf R. C , Henderson H , Castellani L. W., Lusis A. J., Ma Y., Forsythe I. J., Zhang H , Kirk E., Brunzell J. D., and Hayden M. R. 1994. Alteration of lipid profiles in plasma of transgenic mice expressing human lipoprotein lipase. The Journal of Biological Chemistry 269[15], 11417-11424. 27. Shimada M., Shimano H , Gotoda T., Yamamato K., Kawamura M., Inaba T., Yazaki Y., and YamadaN. 1993. Overexpression of human lipoprotein lipase in transgenic mice. Resistance to diet-induced hypertriglyceridemia and hypercholesterolemia. The Journal of Biological Chemistry 268[24], 17924-17929. 28. Shimada M., Ishibashi S., Inaba T., Yagyu H , Harada K., Osuga J.-L, Ohashi K., Yazaki Y., and Nobuhiro Y. 1996. Suppression of diet-induced atherosclerosis in low density lipoprotein receptor knockout mice overexpressing lipoprotein lipase. Proceedings of the National Academy of Sciences, USA 93, 7242-7246. 29. Tsutsumi K., Inoue Y., Shima A., Iwasaki K., Kawamura M., and Murase T. 1997. The novel compound NO-1886 increases lipoprotein lipase activity with resulting elevation of high density lipoprotein cholesterol, and long-term administration inhibits atherogenesis in the coronary arteries of rats with experimental atherosclerosis. Journal of Clinical Investigation 92, 411-417. 30. Tsutsumi K., Inoue Y., Shima A., and Murase T. 1995. Correction of hypertriglyceridemia with low high-density lipoprotein cholesterol by the novel compound NO-1886, a lipoprotein lipase-promoting agent, in STZ-induced diabetic rats. Diabetes 44[4], 414-417. 31. Shepherd J. 1995. Fibrates and statins in the treatment of hyperlipidaemia: an appraisal of their efficacy and safety. European Heart Journal 16[1], 5-13. 197 32. Staels B., Schoonjans K., Fruchart J. C , and Auwerx J. 1997. The effects of fibrates and thiazolidinediones on plasma triglyceride metabolism are mediated by distinct peroxisome proliferator activated receptors (PPARs). Biochimie 79, 95-99. 33. Coleman T., Seip R. L., Gimble J. M., Lee D., Maeda N., and Semenkovich C. F. 1995. COOH-terminal disruption of lipoprotein lipase in mice is lethal in homozygotes, but heterozygotes have elevated triglycerides and impaired enzyme activity. The Journal of Biological Chemistry 270[21], 12518-12525. 34. Weinstock P. H., Bisgaier C. L., Aalto-Setala K., Radner H., Ramakrishnan R., Levak-Frank S., Essenburg A. D., Zechner R., and Breslow J. L. 1995. Severe hypertriglyceridemia, reduced high density lipoprotein, and neonatal death in lipoprotein lipase knockout mice. Mild hypertriglyceridemia with impaired very low density lipoprotein clearance in heterozygotes. Journal of Clinical Investigation 96[6], 2555-2568. 35. Ginzinger D. G., Lewis M. E. S., Ma Y., Jones B. R., Liu G., Jones S. D., and Hayden M. R. 1996. A mutation in the lipoprotein lipase gene is the molecular basis of chylomicronemia in a colony of domestic cats. Journal of Clinical Investigation 97, 1257-1266. 36. Ginzinger D. G., Clee S. M., Dallongeville J., Lewis M. E. S., Henderson H. E., Bauje E., Rogers Q. R., Jensen D. R., Eckel R. H., Dyer R., Innis S., Jones B., Fruchart J.-C, and Hayden M. R. 1999. Lipid and lipoprotein analysis of cats with lipoprotein lipase deficiency. European Journal of Clinical Investigation 29[1], 17-26. 37. Liu G., Ashbourne Excoffon K. J. D., Wilson J. E., McManus B. M., Rogers Q. R., Miao L., Kastelein J. J. P., Lewis M. E. S., and Hayden M. R. 2000. Phenotypic correction of feline lipoprotein lipase deficiency by adenoviral gene transfer. Human Gene Therapy 11, 21-32. 38. Christophersen B., Nordstoga K., Shen Y., Olivecrona T., and Olivecrona G. 1997. Lipoprotein lipase deficiency with pancreatitis in mink: biochemical characterization and pathology. J.Med.Genet. 38[5], 837-846. 39. Lindberg A., Nordstoga K., Christophersen B., Savonen R., van Tol A., and Olivecrona G. 1998. A mutation in the lipoprotein lipase gene associated with hyperlipoproteinemia type I in mink: studies on lipid and lipase levels in heterozygotes. Int.J.Mol.Med. 1[3], 529-538. 40. Savonen R., Nordstoga K., Christophersen B., Lindberg A., Shen Y., Hultin M., Olivecrona T., and Olivecrona G. 1999. Chylomicron metabolism in an animal model for hyperlipoproteinemia type I. Journal of Lipid Research 40[7], 1336-1346. 41. Scow R. O., Schultz C. J., Park J. W., and Blanchette-Mackie E. J. 1998. Combined lipase deficiency (cld/cld) in mice affects differently post-translational processing of 198 lipoprotein lipase, hepatic lipase and pancreatic lipase. Chemistry and Physics of Lipids 93, 149-155. 42. Merkel M., Weinstock P. H., Chajek-Shaul T., Radner H., Yin B., Breslow J. L., and Goldberg I. J. 1998. Lipoprotein lipase expression exclusively in the liver. A mouse model for metabolism in the neonatal period and during cachexia. Journal of Clinical Investigation 102[5], 893-901. 43. Levak-Frank S., Weinstock P. H., Hayek T., Verdery R., Hofmann W., Ramakrishnan R., Sattler W., Breslow J. L., and Zechner R. 1997. Induced mutant mice expressing lipoprotein lipase exclusively in muscle have subnormal triglycerides yet reduced high density lipoprotein cholesterol levels in plasma. The Journal of Biological Chemistry 272[27], 17182-17190. 44. Levak-Frank S., Hofmann W., Weinstock P. H , Radner H., Sattler W., Breslow J. L., and Zechner R. 1999. Induced mutant mouse lines that express lipoprotein lipase in cardiac muscle, but not in skeletal muscle and adipose tissue, have normal plasma triglyceride and high-density lipoprotein-cholesterol levels. Proceedings of the National Academy of Sciences, USA 96, 3165-3170. 45. Jones B. R, Wallace A., Harding D. R. K., Hancock W. S., and Campbell C. H. 1983. Occurrence of idiopathic, familial hyperchylomicronemia in a cat. Veterinary Record 112,543-547. 46. Peritz L. N., Brunzell J. D., Harvey-Clarke C , Pritchard H , Jones B., and Hayden M. R. 1990. Characterization of a lipoprotein lipase class III type defect in hypertriglyceridemic cats. Clinical and Investigative Medicine 13 [5], 259-263. 47. Thompson J. C , Johnstone A. C , Jones B. R., and Hancock W. S. 1989. The ultrastructural pathology of five lipoprotein lipase-deficient cats. Journal of Comparative Pathology 101,251-262. 48. Jones B. R, Johnstone A. C , Cahill J. I., and Hancock W. S. 1986. Peripheral neuropathy in cats with inherited primary hyperchylomicronaemia. Veterinary Record 119, 268-272. 49. Krapp A., Zhang H , Ginzinger D. G , Liu M.-S., Lindberg A., Olivecrona G , Hayden M. R., and Beisiegel U. 1995. Structural features in lipoprotein lipase necessary for the mediation of lipoprotein uptake into cells. Journal of Lipid Research 36[11], 2362-2373. 50. Demacker P., van Heijst P. J., Hak-Lemmers H. L. M., and Stalenhoef A. F. H. 1987. A study of the lipid transport system in the cat, Felix domesticus. Atherosclerosis 66, 113-123. 51. Stalenhoef A. F. H , Malloy M. J., Kane J. P., and Havel R. J. 1984. Metabolism of apolipoproteins B-48 and B-100 of triglyceride-rich lipoproteins in normal and lipoprotein lipase-deficient humans. Journal of Clinical Investigation 88,985-994. 199 52. Zsigmond E., Scheffler E., Forte T. M.^ Potenz R., Wu W., and Chan L. 1994. Transgenic mice expressing human lipoprotein lipase driven by the mouse metallothionein promoter. A phenotype associated with increased perinatal mortality and reduced plasma very low density lipoprotein of normal size. The Journal of Biological Chemistry 269[29], 18757-18766. 53. Shimada M., Ishibashi S., Yamamato K., Kawamura M., Watanabe Y., Gotoda T., Harada K., Inaba T., Ohsuga J., Yazaki Y., and Yamada N. 1995. Overexpression of human lipoprotein lipase increases hormone-sensitive lipase activity in the adipose tissue of mice. Biochemical and Biophysical Research Communications 211 [3], 761-766. 54. Shimada M., Ishibashi S., Gotoda T., Kawamura M., Yamamato K., Inaba T., Harada K., Ohsuga J., Perrey S., Yazaki Y., and Yamada N. 1995. Overexpression of human lipoprotein lipase protects diabetic transgenic mice from diabetic hypertriglyceridemia and hypercholesterolemia. Arteriosclerosis, Thrombosis and Vascular Biology 15[10], 1688-1694. 55. Sattler W., Frank-Levak S., Radner H., Kostner G. M., and Zechner R. 1996. Muscle-Specific overexpression of lipoprotein lipase in transgenic mice results in increased alpha-tocopherol levels in skeletal muscle. Biochem.J. 318, 15-19. 56. Hoefler G., Noehammer C , Levak-Frank S., El-Shabrawi Y., Schauer S., Zechner R., and Radner H. 1996. Muscle-specific overexpression of human lipoprotein lipase in mice causes increased intracellular free fatty acids and induction of peroxisomal enzymes. Biochimie 79, 163-168. 57. Jensen D. R., Schlaepfer C. L., Morin D. S., Pennington T. M., Ammon S. M., Gutierrez-Hartmann A., and Eckel R. H. 1997. Prevention of diet-induced obesity in transgenic mice overexpressing skeletal muscle lipoprotein lipase. AmJ.Physiol. 273[42],R683-R689. 58. Merkel M., Kako Y., Radner H., Cho I. S., Ramasamy R., Brunzell J. D., Goldberg I. J., and Breslow J. L. 1998. Catalytically inactive lipoprotein lipase expression in muscle of transgenic mice increases very low density lipoprotein uptake: direct evidence that lipoprotein lipase bridging occurs in vivo. Proceedings of the National Academy of Sciences, USA 95,13841-13846. 59. Weinstock P. H., Levak-Frank S., Hudgins L. C , Radner H., Friedman J. M., Zechner R., and Breslow J. L. 1997. Lipoprotein lipase controls fatty acid entry into adipose tissue, but fat mass is preserved by endogenous synthesis in mice deficient in adipose tissue lipoprotein lipase. Proceedings of the National Academy of Sciences, USA 94,10261-10266. 60. Yagyu H., Ishibashi S., Chen Z., Osuga J.-L, Okazaki M., Perrey S., Kitamine T., Shimada M., Ohashi K., Harada K., Shionoiri F., Yahagi N., Gotoda T., Yazaki Y., and Yamada N. 1999. Overexpressed lipoprotein lipase protects against 200 atherosclerosis in apolipoprotein E knockout mice. Journal of Lipid Research 40, 1677-1685. 61. Clee S. M., Bissada N., Miao F., Miao L., Marais A. D., Henderson H. E., Steures P., McManus J., McManus B. M., LeBoeuf R. C , Kastelein J. J. P., and Hayden M. R. 2000. Plasma and vessel wall lipoprotein lipase have different roles in atherosclerosis. Journal of Lipid Research 41 [4], 521 -531. 62. Semenkovich C. F., Coleman T., and Daugherty A. 1998. Effects of heterozygous lipoprotein lipase deficiency on diet-induced atherosclerosis in mice. Journal of Lipid Research 39, 1141 -1151. 63. Renier G., Skamene E., DeSanctis J. B., and Radzioch D. 1993. High macrophage lipoprotein lipase expression and secretion are associated in inbred murine strains with susceptibility to atherosclerosis. Arteriosclerosis and Thrombosis 13,190-196. 64. Vilaro S., Llobera M., Bengtsson-Olivecrona G , and Olivecrona T. 1988. Lipoprotein lipase uptake by the liver: localization, turnover, and metabolic role. Am.J.Physiol. 254[5 Pt 1], G711-G722. 65. Robbins P. D., Tahara H , and Ghivizzani S. C. 1998. Viral vectors for gene therapy. Trends in Biotechnology 16[1], 35-40. 66. Treco D. A. and Selden R. F. 1995. Non-viral gene therapy. Mol Med Today 1 [7], 314-321. 67. Graham F. L. and Prevec L. 1991. Manipulation of adenovirus vectors. Methods in molecular biology Vol. 7: Gene transfer and expression protocols. 109-127. NJ. 68. Liu G., Ashbourne Excoffon K. J. D., Benoit P., Ginzinger D. G , Miao L., Ehrenborg E., Duverger N., Denefle P., Hayden M. R., and Lewis M. E. S. 1997. Efficient adenovirus-mediated ectopic gene expression of human lipoprotein lipase in human hepatic (HepG2) cells. Human Gene Therapy 8, 205-214. 69. Enerbach S. and Gimble J. M. 1993. Lipoprotein lipase gene expression: physiological regulators at the transcriptional and post-transcriptional level. Biochimica et Biophysica Acta 1109, 107-125. 70. Kessler J. I. 1963. Effect of diabetes and insulin on the activity of myocardial and adipose tissue lipoprotein lipase of rats. Journal of Clinical Investigation 42[3], 362-367. 71. Kern P. A., Mandic A., and Eckel R. H. 1987. Regulation of lipoprotein lipase by glucose in primary cultures of isolated human adipocytes: relevance to hypertriglyceridemia of diabetes. Diabetes 36[11], 1238-1245. 201 72. Ong J. M., Kirchgessner T. G., Schotz M. C., and Kern P. A. 1988. Insulin increases the synthetic rate and messenger RNA level of lipoprotein lipase in isolated rat adipocytes. The Journal of Biological Chemistry 263[26], 12933-12938. 73. Bengtsson G. and Olivecrona T. 1980. Lipoprotein lipase: mechanism of product inhibition. European Journal of Biochemistry 106, 557-562. 74. Amri E.-Z., Teboul L., Vannier C , Grimaldi P.-A., and Ailhaud G. 1996. Fatty acids regulate the expression of lipoprotein lipase gene and activity in preadipose and adipose cells. Biochem. J. 314, 541-546. 75. Goodpaster B. H. and Kelley D. E. 1998. Role of muscle in triglyceride metabolism. Current Opinion in Lipidology 9, 231-236. 76. Karpe F., Olivecrona T., Walldius G., and Hamsten A. 1992. Lipoprotein lipase in plasma after an oral fat load: relation to free fatty acids. Journal of Lipid Research 33, 975-984. 77. Kern P. A., Saghizadeh M., Ong J. M., Bosch R. J., Deem R., and Simsolo R. B. 1995. The expression tumor necrosis factor in human adipose tissue: regulation by obesity, weight loss, and relationship to lipoprotein lipase. Journal of Clinical Investigation 95, 2111-2119. 78. Morin C. L. and Eckel R. H. 1995. Transcriptional regulation of the lipoprotein lipase gene. Atherosclerosis X. 231-235. Amsterdam, Elsevier Science B.V. 79. Larochelle N., Lochmuller H., Zhao J., Jani A., Hallauer P., Hastings K. E. M., Massie B., Prescott S., Petrof B. J., Karpati G., and Nalbantoglu J. 1997. Efficient muscle-specific transgene expression after adenovirus-mediated gene transfer in mice using a 1.35 kb muscle creatine kinase promoter/enhancer. Gene Therapy 4, 465-472. 80. Acsadi G., Agnes J., Massie B., Simoneau M., Holland P., Blaschuk K., and Karpati G. 1994. A differential efficiency of adenovirus-mediated in vivo gene transfer into skeletal muscle cells of different maturity. Human Molecular Genetics 3 [4], 579-584. 81. Davis H. L., Demeneix B. A., Quantin B., Coulombe J., and Whalen R. G. 1993. Plasmid DNA is superior to viral vectors for direct gene transfer into adult mouse skeletal muscle. Human Gene Therapy 4, 733-740. 82. Fisher K. J., Jooss K., Alston J., Yang Y., Haecker S. E., High K., Pathak R., Raper S. E., and Wilson J. M. 1997. Recombinant adeno-associated virus for muscle directed gene therapy. Nature Medicine 3[3], 306-312. 83. Rivera V. M., Ye X., Courage N. L., Sachar J., Cerasoli F., Wilson J. M., and Gilman M. 1999. Long-term regulated expression of growth hormone in mice after intramuscular gene transfer. Proceedings of the National Academy of Sciences, USA 96, 8657-8662. 202 84. Semenkovich C. F., Wims M., Noe L., Etienne J., and Chan L. 1989. Insulin regulation of lipoprotein lipase activity in 3T3-L1 adipocytes is mediated at posttranscriptional and posttranslational levels. The Journal of Biological Chemistry 264[15], 9030-9038. 85. Maheux P., Azhar S., Kern P. A., Chen Y.-D. I., and Reaven G. M. 1997. Relationship between insulin-mediated glucose disposal and regulation of plasma and adipose tissue lipoprotein lipase. Diabetologia 40, 850-858. 86. Amri E.-Z., Dani C , Doglio A., Grimaldi P.-A., and Ailhaud G. 1986. Coupling of growth arrest and expression of early markers during adipose conversion of preadipocyte cell lines. Biochem.Biophys.Res.Commun 137[2], 903-910. 87. Bensadoun A. 1991. Lipoprotein lipase. Annual Review of Nutrition 11, 217-237. 88. Levine J. A., Jensen M. D., Eberhardt N. L., and O'Brien T. 1998. Adipocyte macrophage colony-stimulating factor is a mediator of adipose tissue growth. Journal of Clinical Investigation 101, 1557-1564. 89. Kern P. 1997. Obesity: Common symptom of diverse gene-based metabolic dysregulation: Potential role of TNFa and lipoprotein lipase as candidate genes for obesity. Journal of Nutrition 127, 1917S-1922S. 90. Tanuma Y., Hiroki N., Esumi M., and Endo H. 1995. A silencer element for the lipoprotein lipase gene promoter and cognate double- and single-stranded DNA-binding proteins. Molecular and Cellular Biology 15[1], 517-523. 91. Boucher R. C. 1999. Status of gene therapy for cystic fibrosis lung disease. Journal of Clinical Investigation 103 [4], 441-445. 92. Imai E., Akagi Y., and Isaka Y. 1998. Towards gene therapy for renal diseases. Nephrologie 19[7], 397-402. 93. Goldberg I. J. 1993. Lipoprotein metabolism in normal and uremic patients. Am.J.Kidney Disease 21[1], 87-90. 94. Kozarsky K. F., McKinley D. F., Austin L. L., Raper S. E., Stratford-Perricaudet L. D., and Wilson J. M. 1994. In vivo correction of low density lipoprotein receptor deficiency in the Watanabe heritable hyperlipidemic rabbit with recombinant adenoviruses. The Journal of Biological Chemistry 269[18], 13695-13702. 95. Kozarsky K. F., Jooss K., Donahee M., Strauss J. F. 3., and Wilson J. M. 1996. Effective treatment of familial hypercholesterolaemia in the mouse model using adenovirus-mediated transfer of the VLDL receptor gene. Nature Genetics 13[1], 54-62. 96. Ishibashi S., Brown M. S., Goldstein J. L., Gerard R. D., Hammer R. E., and Herz J. 1993. Hypercholesterolemia in low density lipoprotein receptor knockout mice and 203 its reversal by adenovirus-mediated gene delivery. Journal of Clinical Investigation 92, 883-893. 97. Applebaum-Bowden D., Kobayashi.J., Kashyap V. S., Brown D. R., Berard A., Meyn S., Parrott C , Maeda N., Shamburek R. D., Brewer H. B., Jr., and Santamarina-Fojo S. 1996. Hepatic lipase gene therapy in hepatic lipase-deficient mice. Adenovirus-mediated replacement of a lipolytic enzyme to the vascular endothelium. Journal of Clinical Investigation 97[3], 799-805. 98. Kopfler W. P., Willard M., Betz T., Willard J. E., Gerard R. D., and Meidell R. S. 1994. Adenovirus-mediated transfer of a gene encoding human apolipoprotein A-I into normal mice increases circulating high-density lipoprotein cholesterol. Circulation 90, 1319-1327. 99. Stevenson S. C , Marshall-Neff J., Teng B., Lee C. B., Roy S., and McClelland.A. 1995. Phenotypic correction of hypercholesterolemia in apoE-deficient mice by adenovirus-mediated in vivo gene transfer. Arteriosclerosis and Thrombosis 15, 479-484. 100. Chan L. 1995. Use of somatic gene transfer to study lipoprotein metabolism in experimental animals in vivo. Current Opinion in Lipidology 6, 335-340. 101. Gerard R. D. and Chan L. 1996. Adenovirus-mediated gene transfer: strategies and applications in lipoprotein research. Current Opinion in Lipidology 7[2], 105-111. 102. Belalcazar M. and Chan L. 1999. Somatic gene therapy for dyslipidemias. J Lab Clin Med 134[3], 194-214. 103. Schiedner G., Morral N., Parks R. J., Wu Y., Koopmans S. C , Langston C , Graham F. L., Beaudet A. L., and Kochanek S. 1998. Genomic DNA transfer with a high-capacity adenovirus vector results in improved in vivo gene expression and decreased toxicity. Nature Genetics 18[2], 180-183. 104. Brenner M. 1999. Gene transfer by adenovectors. Blood 94[12], 3965-3967. 105. Huard J., Lochmuller H., Acsadi G., Jani A., Massie B., and Karpati G. 1995. The route of administration is a major determinant of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Therapy 2,107-115. 106. Camps L., Reina M., Llobera M., Vilaro S., and Olivecrona T. 1990. Lipoprotein lipase: cellular origin and functional distribution. AmJ.Physiol. 258, C673-C681. 107. Kobayashi J., Applebaum-Bowden D., Dugi K. A., Brown D. R., Kashyap V. S., Parrott C , Duarte C , Maeda N., and Santamarina-Fojo S. 1996. Analysis of protein structure-function in vivo: adenovirus-mediated transfer of lipase lid mutants in hepatic lipase-deficient mice. The Journal of Biological Chemistry 271 [42], 26296-26301. 204 \ 108. Zsigmond E., Kobayashi K., Tzung K.-W., Li L., Fuke Y., and Chan L. 1997. Adenovirus-mediated gene transfer of human lipoprotein lipase ameliorates the hyperlipidemias associated with apolipoprotein E and LDL receptor deficiencies in mice. Human Gene Therapy 8, 1921-1933. 109. Lewis M. E. S., Forsythe I., Marth J. D., Brunzell J. D., Hayden M. R., and Humphries R. K. 1995. Retroviral-mediated gene transfer and expression of human lipoprotein lipase in somatic cells. Human Gene Therapy 6, 853-863. 110. Pakkanen T. M., Laitinen M., Hippelainen M., Kallionpaa H , Lehtolainen P., Leppanen P., Luoma J. S., Tarvainen R., and Yla-Herttuala S. 1999. Enhanced plasma cholesterol lowering effect of retrovirus-mediated LDL receptor gene transfer to WHHL rabbit liver after improved surgical technique and stimulation of hepatocyte proliferation by combined partial liver resection and thymidine kinase-ganciclovir treatment. Gene Therapy 6[1], 34-41. 111. Schlaepfer I. R. and Eckel R. H. 1999. Plasma triglyceride reduction in mice after direct injections of muscle-specific lipoprotein lipase DNA. Diabetes 48, 223-227. 112. Isner J. M., Baumgartner I., Rauh G., Schainfeld R, Manor O., Razvi S., and Symes J. F. 1998. Treatment of thromboangiitis obliterans (Buerger's disease) by intramuscular gene transfer of vascular endothelial growth factor: preliminary clinical results. Journal of Vascular Surgery 28[6], 964-973. 113. Baumgartner I., Pieczek A., Manor O., Blair R, Kearney M., Walsh K., and Isner J. M. 1998. Constitutive expression of phVEGF165 after intramuscular gene transfer promotes collateral vessel development in patients with critical limb ischemia. Circulation 97[12], 1114-1123. 114. Monahan P. E. and Samulski R. J. 2000. AAV vectors: is clinical success on the horizon? Gene Therapy 7[1], 24-30. 115. Bartlett J. S., Wilcher R., and Samulski R. J. 2000. Infectious entry pathway of adeno-associated virus and adeno-associated virus vectors. J.Virology 74[6], 2777-2785. 116. Snyder R. O., Spratt S. K., Lagarde C , Bohl D., Kaspar B., Sloan B., Cohen L. K., and Danos O. 1997. Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Human Gene Therapy 8[16], 1891-1900. 117. Marshall B. A., Tordjman K., Host H. H , Ensor N. J., Kwon G., Marshall C. A., Coleman T., McDaniel M. L., and Semenkovich C. F. 1999. Relative hypoglycemia and hyperinsulinemia in mice with heterozygous lipoprotein lipase (LPL) deficiency. The Journal of Biological Chemistry 274[39], 27426-27432. 118. Kay M. A., Manno C. S., Ragni M. V., Larson P. J., Couto L. B., McClelland A., Glader B., Chew A. J., Tai S. J., Herzog R. W., Arruda V., Johnson F., Scallan C , 205 Skarsgard E., Flake A. W., and High K. A. 2000. Evidence for gene transfer and expression of factor IX in hemophilia B patients treated with an AAV vector. Nature Genetics 24[3], 257-261. 119. Ashbourne Excoffon K. J. D., Liu G., Miao L., Wilson J. E., McManus B. M., Semenkovich C. F., Coleman T., Benoit P., Duverger N., Branellec D., Denefle P., Hayden M. R., and Lewis M. E. S. 1997. Correction of hypertriglyceridemia and impaired fat tolerance in lipoprotein lipase-deficient mice by adenovirus-mediated expression of human lipoprotein lipase. Arteriosclerosis, Thrombosis and Vascular Biology 17,2532-2539. 120. Liu G., Ashbourne Excoffon K. J. D., Wilson J. E., McManus B. M., Miao L., Benoit P., Duverger N., Branellec D., Denefle P., Hayden M. R., and Lewis M. E. S. 1998. Enhanced lipolysis in normal mice expressing liver-derived human lipoprotein lipase after adenoviral gene transfer. Clinical and Investigative Medicine 21 [4-5], 172-185. 121. Nabel G. J. 1999. Development of optimized vectors for gene therapy. Proceedings of the National Academy of Sciences, USA 96, 324-326. 122. Zhang H., Henderson H., Gagn6 S. E., Clee S. M., Miao L., Liu G., and Hayden M. R. 1996. Common sequence variants of lipoprotein lipase: standardized studies of in vitro expression and catalytic function. Biochimica et Biophysica Acta 1302, 159-166. 123. Stratford-Perricaudet L. D., Makeh I., Perricaudet M., and Briand P. 1992. Widespread long-term gene transfer to mouse skeletal muscles and heart. Journal of Clinical Investigation 90, 626-630. 124. Mulligan R. C. and Berg P. 1980. Expression of a bacterial gene in mammalian cells. Science 209,1422-1427. 125. Cotten M., Baker A., Saltik M., Wagner E., and Buschle M. 1994. Lipopolysaccharide is a frequent contaminant of plasmid DNA preparations and can be toxic to primary human cells in the presence of adenovirus. Gene Therapy 1, 239-246. 126. Zhang W. W., Koch P. E., and Roth J. A. 1995. Detection of wild-type contamination in a recombinant adenoviral preparation by PCR. BioTechniques 18, 4 4 4 . 4 4 7 . 127. Nilsson-Ehle P. and Schotz M. C. 1976. A stable, radioactive substrate emulsion for assay of lipoprotein lipase. Journal of Lipid Research 17, 536-541. 128. Iverius P.-H. and Ostlund-Lindqvist A.-M. 1986. Preparation, characterization and measurement of lipoprotein lipase. Methods in Enzymology 129, 691-704. 206 129. Peterson J., Fujimoto W. Y., and Brunzell J. D. 1992. Human lipoprotein lipase: relationship of activity, heparin affinity, and conformation as studied with monoclonal antibodies. Journal of Lipid Research 33, 1165-1170. 130. Chomczynski P. and Sacchi N. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry 162, 156-159. 131. Levesque G., Lamarche B., Murthy M. R. V., Julien P., Despres J. P., and Deshaies Y. 1994. Determination of changes in specific gene expression by reverse transcription PCR using interspecies mRNAs as internal standards. BioTechniques 17[4], 738-741. 132. Auwerx J., Schoonjans K., and Staels B. 1995. Regulation of liver lipoprotein lipase gene expression. Atherosclerosis X. 241-245. Amsterdam, Elsevier Science. 133. Olivecrona T., Chernick S. S., Bengtsson-Olivecrona G., Garrison M., and Scow R. O. 1987. Synthesis and secretion of lipoprotein lipase in 3T3-L1 adipocytes. Demonstration of inactive forms of lipase in cells. The Journal of Biological Chemistry 262[22], 10748-10759. 134. Li Q., Kay M. A., Finegold M., Stratford-Perricaudet L. D., and Woo S. L. 1993. Assessment of recombinant Ad vectors for hepatic gene therapy. Human Gene Therapy 4, 403-409. 135. Barr D., Tubb J., Ferguson D., Scaria A., Lieber A., Wilson C , Perkins J., and Kay M. A. 1995. Strain related variations in adenovirally mediated transgene expression from mouse hepatocytes in vivo: comparisons between immunocompetent and immunondeficient inbred strains. Gene Therapy 2,151-155. 136. Kay M. A., Holterman A., Meuse L., Gown A., Ochs H. D., Linsley P. S., and Wilson C. B. 1995. Long-term hepatic adenovirus-mediated gene expression in mice following CTLA4Ig administration. Nature Genetics 11,191-197. 137. Cheng L., Ziegelhoffer P. R., and Yang N.-S. 1993. In vivo promoter activity and transgene expression in mammalian somatic tissues evaluated by using particle bombardment. Proceedings of the National Academy of Sciences, USA 90, 4455-4459. 138. Gorman C. M., Merlino G. T., Willingham M. C , Pastan I., and Howard B. H. 1982. The rous sarcoma virus long terminal repeat is a strong promoter when introduced into a variety of eukaryotic cells by DNA-mediated transfection. Proceedings of the National Academy of Sciences, USA 79, 6777-6781. 139. Levrero M., Barban V., Manteca S., Ballay A., Balsamo C , Avantaggiati M. L., Natoli G., Skellekens H., Tiollais P., and Perricaudet M. 1991. Defective and nondefective adenovirus vectors for expressing foreign genes in vitro and in vivo. Gene 101 [2], 195-202. 207 140. Stratford-Perricaudet L. D., Levrero M., Chasse J. F., Perricaudet M., and Briand P. 1990. Evaluation of the transfer and expression in mice of an enzyme-encoding gene using a human adenovirus vector. Human Gene Therapy 1[3], 241-256. 141. Jaffe H. A., Danel C , Longenecker G , Metzger M., Setoguchi Y., Rosenfeld M. A., Gant T. W., Thorgeirsson S. S., Stratford-Perricaudet L. D., and Perricaudet M. 1992. Adenovirus-mediated in vivo gene transfer and expression in normal rat liver. Nature Genetics 1 [5], 372-378. 142. Chevreuil O., Hultin M., Ostergaard P., and Olivecrona T. 1993. Depletion of lipoprotein lipase after heparin administration. Arteriosclerosis and Thrombosis 13, 1391-1396. 143. Karpe F., Humphreys S. M., Samra J. S., Summers L. K. M., and Frayn K. N. 1997. Clearance of lipoprotein remnant particles in adipose tissue and muscle in humans. Journal of Lipid Research 38, 2335-2343. 144. Karpe F., Olivecrona T., Olivecrona G , Samra J. S., Summers L. K. M., Humphreys S. M., and Frayn K. N. 1998. Lipoprotein lipase transport in plasma: role of muscle and adipose tissues in regulation of plasma lipoprotein lipase concentrations. Journal of Lipid Research 39,2387-2393. 145. Ranganathan S., Ciaraldi T. P., Henry R. R, Mudaliar S., and Kern P. A. 1998. Lack of effect of leptin on glucose transport, lipoprotein lipase, and insulin action in adipose and muscle cells. Endocrinology 139, 2509-2513. 146. Knowles B. B., Howe C. C , and Aden D. P. 1980. Human hepatocellular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen. Science 209[4455], 497-499. 147. Javitt N. B. 1990. Hep G2 cells as a resource for metabolic studies: lipoprotein, cholesterol, and bile acids. FASEB J 4[2], 161-168. 148. Spergel J. M. and Chen-Kiang S. 1991. Interleukin 6 enhances a cellular activity that functionally substitutes for El A protein in transactivation. Proceedings of the National Academy of Sciences, USA 88[15], 6472-6476. 149. Evans A. J., Sawyez C. G , Wolfe B. M., and Huff M. W. 1992. Lipolysis is a prerequisite for lipid accumulation in HepG2 cells induced by large hypertriglyceridemic very low density lipoproteins. Proceedings of the National Academy of Sciences, USA 267[15], 10743-10751. 150. Williams K. J., Fless G. M., Petrie K. A., Snyder M. L., Brocia R. W., and Swenson T. L. 1992. Mechanisms by which lipoprotein lipase alters cellular metabolism of lipoprotein(a), low density lipoprotein, and nascent lipoproteins. Roles for low density lipoprotein receptors and heparan sulfate proteoglycans. The Journal of Biological Chemistry 267[19], 13284-13292. 208 151. Gimenez-Llort L., Vilanova J., Skottova N., Bengtsson-Olivecrona G., Llobera M., and Robert M. Q. 1991. Lipoprotein lipase enables triacylglycerol hydrolysis by perfused newborn rat liver. AmJ.Physiol. 261 [4 Pt 1], G641-G647. 152. Lombardi P., Mulder M., van der Boom H., Frants R. R., and Havekes L. M. 1993. Inefficient degradation of triglyceride-rich lipoprotein by HepG2 cells is due to a retarded transport to the lysosomal compartment. The Journal of Biological Chemistry 268[35], 26113-26119. 153. Severson D. L. and Carroll R. 1995. Posttranscriptional regulation of lipoprotein lipase. Atherosclerosis X. 236-240. Amsterdam, Elsevier Science. 154. Saxena U., Klein M. G., and Goldberg I. J. 1991. Transport of lipoprotein lipase across endothelial cells. Proceedings of the National Academy of Sciences, USA 88[6], 2254-2258. 155. Goldberg I. J., Sivaram P., Choi S. Y., Parthasarathy N., and Wagner W. D. 1995. Lipoprotein lipase binding to endothelial cells: specific protein-protein and protein-glycosaminoglycan interactions. Atherosclerosis X. 246-249. Amsterdam, Elsevier Science. 156. Edwards C. P. and Aruffo A. 1993. Current applications of COS cell based transient expression systems. Curr Opin Biotechnol 4[5], 558-563. 157. Ishida B. Y., Blanche P. J., Nichols A. V., Yashar M., and Paigen B. 1991. Effects of atherogenic diet consumption on lipoproteins in mouse strains C57BL/6 and C3H. Journal of Lipid Research 32, 559-568. 158. Tripathy S. K., Black H. B., Goldwasser E., and Leiden J. M. 1996. Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replication-defective adenovirus vectors. Nature Medicine 2[5], 545-550. 159. Michou A., Santoro L., Christ M., Julliard V., Pavirani A., and Mehtali M. 1997. Adenovirus-mediated gene transfer: influence of transgene, mouse strain and type of immune response on persistence of transgene expression. Gene Therapy 4[5], 473-482. 160. Ducobu J. and Dupont P. 1981. Inherited raised alkaline phosphatase activity in the absence of disease. The Lancet 1, 1372-1373. 161. Coppack S. W., Jensen M. D., and Miles J. M. 1994. In vivo regulation of lipolysis in humans. Journal of Lipid Research 35, 177-193. 162. Auerbach B. J., Bisgaier C. L., WoTle J., and Saxena U. 1996. Oxidation of low density lipoproteins greatly enhances their association with lipoprotein lipase anchored to endothelial cell matrix. The Journal of Biological Chemistry 271 [3], 1329-1335. 209 163. Skottova N., Savonen R., Lookene A., Hultin M., and Olivecrona G. 1995. Lipoprotein lipase enhances removal of chylomicrons and chylomicron remnants by the perfused rat liver. Journal of Lipid Research 36, 1334-1344. 164. Beisiegel U., Weber W., and Bengtsson-Olivecrona G. 1991. Lipoprotein lipase enhances the binding of chylomicrons to low density lipoprotein receptor-related protein. Proceedings of the National Academy of Sciences, USA 88, 8342-8346. 165. Masuno H., Tsujita T., Nakanishi H., Yoshida A., Fukunishi R, and Okuda H. 1984. Lipoprotein lipase-like activity in the liver of mice with Sarcoma 180. Journal of Lipid Research 25, 419-427. 166. Tracey K. J. and Cerami A. 1992. Tumor necrosis factor in the malnutrition (cachexia) of infection and cancer. Am.J.Trop.Med.Hyg. 47, 2-7. 167. Feingold K. R, Marshall M., Gulli R, Moser A. FL, and Grunfeld G 1991. Effect of endotoxin and cytokines on lipoprotein lipase activity in mice. Arteriosclerosis and Thrombosis 14, 1866-1872. 168. Sabugal R., Robert M. Q., Julve J., Auwerx J., Llobera M., and Peinado-Onsurbe J. 1996. Hepatic regeneration induces changes in lipoprotein lipase activity in several tissues and it re-expression in the liver. Biochem.J. 318, 597-602. 169. O'Brien K. D., Gordon D., Deeb S., Ferguson M., and Chait A. 1992. Lipoprotein lipase is synthesized by macrophage-derived foam cells in human coronary atherosclerotic plaques. Journal of Clinical Investigation 89, 1544-1550. 170. Mattsson L., Johansson H , Ottosson M., Bondjers G , and Wiklund O. 1993. Expression of lipoprotein lipase mRNA and secretion in macrophages isolated from human atherosclerotic aorta. Journal of Clinical Investigation 92,1759-1765. 171. Guo Z. S., Wang L. H , Eisensmith R. C , and Woo S. L. 1996. Evaluation of promoter strength for hepatic gene expression in vivo following adenovirus-mediated gene transfer. Gene Therapy 3, 802-810. 172. Paigen B., Morrow A., Brandon C , Mitchell D., and Holmes P. A. 1985. Variation in susceptibility to atherosclerosis among inbred strains of mice. Atherosclerosis 57, 65-73. 173. Paigen B., Morrow A., Holmes P. A., Mitchell D., and Williams R. A. 1987. Quantitative assessment of atherosclerotic lesions in mice. Atherosclerosis 68,231-240. 174. Wickham T. J., Filardo E. J., Cheresh D. A., and Nemerow G. R. 1993. Integrins a o p3 and 0 ^ 5 promote adenovirus internalization but not virus attachment. Cell 73, 309-319. 210 175. Bergelson J. M. 1999. Receptors mediating adenovirus attachment and internalization. Biochem.Pharmacol. 57[9], 975-979. 176. van Deutekom J. C , Cao B., Pruchnic R., Wickham T. J., Kovesdi I., and Huard J. 1999. Extended tropism of an adenoviral vector does not circumvent the maturation-dependent transducibility of mouse skeletal muscle. J.Gene.Med. 1[6], 393-399. 177. Zechner R. 1997. The tissue-specific expression of lipoprotein lipase: implications for energy and lipoprotein metabolism. Current Opinion in Lipidology 8, 77-88. 178. Levy M. Y., Barron L. G., Meyer K. B., and Szoka F. C. Jr. 1996. Characterization of plasmid DNA transfer into mouse skeletal muscle: evaluation of uptake mechanism, expression and secretion of gene products into blood. Gene Therapy 3[3], 201-211. 179. Kobayashi J., Nishida T., Ameis D., Stahnke G., Schotz M. C , Hashimoto H., Fukamachi I., Shirai K., Saito Y., and Yoshida S. 1992. A Heterozygous mutation(the codon for Ser447> a stop codon) in lipoprotein lipase contributes to a defect in lipid interface recognition in a case with type I hyperlipidemia. Biochem.Biophys.Res.Commun 182[1], 70-77. 180. Henderson H. E., Kastelein J. J. P., Zwinderman A. H., Gagne E., Jukema J. W., Reymer P. W. A., Lie K. I., Bruschke A. V. G., Hayden M. R., and Jansen H. 1999. Lipoprotein lipase activity is.decreased in a large cohort of patients with coronary artery disease and is associated with changes in lipids and lipoproteins. Journal of Lipid Research 40, 735-743. 181. Wittrup H. H., Tybjaerg-Hansen A., and Nordestgaard B. G. 1999. Lipoprotein lipase mutations, plasma lipids and lipoproteins, and risk of ischemic heart disease. A meta-analysis. Circulation 99, 2901-2907. 182. Kozaki K., Gotoda T., Kawamura M., Shimano H., Yazaki Y., Ouchi Y., Orimo H., and Yamada N. 1993. Mutational analysis of human lipoprotein lipase by carboxy-terminal truncation. Journal of Lipid Research 34, 1765-1772. 183. Previato L., Guardamagna O., Dugi K. A., Ronan R., Talley G. D., Santamarina-Fojo S., and Brewer H. B., Jr. 1994. A novel missense mutation in the C-terminal domain of lipoprotein lipase (Glu410Val) leads to enzyme inactivation and familial chylomicronemia. Journal of Lipid Research 35,1552-1560. 184. Beisiegel U. 1996. New aspects on the role of plasma lipases in lipoprotein catabolism and atherosclerosis. Atherosclerosis 124,1-8. 185. Mulder M., Lombardi P., Jansen H., van Berkel T. J. C , Frants R. R., and Havekes L. M. 1993. Low density lipoprotein receptor internalizes low density and very low density lipoproteins that are bound to heparan sulfate proteoglycans via lipoprotein lipase. The Journal of Biological Chemistry 268, 9369-9375. 211 186. Mulder M., Lombardi P., Jansen H., van Berkel T. J. C , Frants R. R., and Havekes L. M. 1992. Heparan sulphate proteoglycans are involved in the lipoprotein lipase-mediated enhancement of the cellular binding of very low density and low density lipoproteins. Biochemical and Biophysical Research Communications 185[2], 582-587. 187. Rinninger F., Kaiser T., Mann W. A., Meyer N., Greten H., and Beisiegel U. 1998. Lipoprotein lipase mediates an increase in the selective uptake of high density lipoprotein-associated cholesterol esters by hepatic cells in culture. Journal of Lipid Research 39, 1335-1348. 188. Jukema J. W., Bruschke A. V. G , van Boven A. J., Reiber J. H. C , Bal E. T., Zwinderman A. H , Jansen H , Boerma G. J. M., van Rappard F. M., and Lie K. I. 1995. Effects of lipid lowering by pravastatin on progression and regression of coronary artery disease in symptomatic men with normal to moderately elevated serum cholesterol levels: the Regression Growth Evaluation Statin Study (REGRESS). Circulation 91, 2528-2540. 189. Kuusi T., Ehnholm C , Viikari J., Harkonen R., Vartiainen E., Puska P., and Taskinen M.-R. 1989. Postheparin plasma lipoprotein and hepatic lipase are determinants of hypo- and hyperalphalipoproteinemia. Journal of Lipid Research 30, 1117-1126. 190. Hata A., Robertson M., Emi M., and Lalouel J.-M. 1990. Direct detection and automated sequencing of individula alleles after electrophoretic strand separation: identification of a common nonsense mutation in exon 9 of the human lipoprotein lipase gene. Nucleic Acids Research 18[18], 5407-5411. 191. Thorn J. A., Needham E. W. A., Mattu R. K., Stocks J., and Galton D. J. 1998. The Ser447-Ter mutation of the lipoprotein lipase gene relates to variability of serum lipid and lipoprotein levels in monozygotic twins. Journal of Lipid Research 39,437-441. 192. Henderson H. E., Ma Y., Hassan M. F., Monsalve V., Marais A. D., Winkler F., Gubernator K., Peterson J., Brunzell J. D., and Hayden M. R. 1991. Amino acid substitution (He 194—>Thr) in exon 5 of the lipoprotein lipase gene causes lipoprotein lipase deficiency in three unrelated probands. Support for a multicentric origin. Journal of Clinical Investigation 87,2005-2011. 193. Benlian P., Etienne J., de Gennes J. L., Noe L., Brault D., Raisonnier A., Hamelin J., Foubert L., Chuat J.-C, Tse C , and Galibert F. 1995. Homozygous deletion of exon 9 causes lipoprotein lipase deficiency: possible intron-Alu recombination. Journal of Lipid Research 36, 356-366. 194. Olivecrona T. and Bengtsson-Olivecrona G. 1993. Lipoprotein lipase and hepatic lipase. Current Opinion in Lipidology 4, 187-196. 195. Eckel R. H. 1989. Lipoprotein lipase. A multifunctional enzyme relevant to common metabolic diseases. The New England Journal of Medicine 320, 1060-1068. 212 196. Holt L. E., Aylward F. X., and Timbers H. G. 1939. Idiopathic familial lipemia. Bull.Johns Hopkins Hosp. 64, 279-314. 213 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items